274 ■ Chapter 9: Survey Research importance of each student’s responding, and the The follow-up mailing stimulated a resurgence mechanics of returning the questionnaire. of returns, as expected, and the same logging pro- cedures continued. The returned postcards told us Students were assured that their responses which additional mailing labels to destroy. Unfor- to the survey were anonymous, and the postcard tunately, time and financial pressures made a third method was explained. A statement followed about mailing impossible, despite initial plans to do so, the auspices under which the study was being con- but the two mailings resulted in an overall return ducted, and a telephone number was provided for rate of 62 percent. those who might want more information about the study. (Five students called for information.) This illustration should give you a fairly good sense of what’s involved in the execution of mailed By printing the introductory letter on the ques- self-administered questionnaires. Let’s turn now to tionnaire, we avoided the necessity of enclosing a the second principal method of conducting surveys, separate letter in the outgoing envelope, thereby in-person interviews. simplifying the task of assembling mailing pieces. Interview Surveys The materials for the initial mailing were as- sembled as follows. (1) One mailing label for each The interview is an alternative method of collect- student was stuck on a postcard. (2) Another label ing survey data. Rather than asking respondents to was stuck on an outgoing manila envelope. (3) read questionnaires and enter their own answers, One postcard and one questionnaire were placed researchers send interviewers to ask the questions in each envelope—with a glance to ensure that the orally and record respondents’ answers. Interview- name on the postcard and on the envelope were ing is typically done in a face-to-face encounter, the same in each case. but telephone interviewing, discussed in the next section, follows most of the same guidelines. The distribution of the survey questionnaires had been set up for a bulk rate mailing. Once the Most interview surveys require more than questionnaires had been stuffed into envelopes, one interviewer, although you might undertake a they were grouped by zip code, tied in bundles, and small-scale interview survey yourself. Portions of delivered to the post office. this section will discuss methods for training and supervising a staff of interviewers assisting you Shortly after the initial mailing, questionnaires with a survey. and postcards began arriving at the research office. Questionnaires were opened, scanned, and as- This section deals specifically with survey inter- signed identification numbers as described earlier in viewing. Chapter 10 discusses the less-structured, this chapter. For every postcard received, a search in-depth interviews often conducted in qualitative was made for that student’s remaining labels, and field research. they were destroyed. The Role of the Survey After two or three weeks, the remaining mail- Interviewer ing labels were used to organize a follow-up mail- ing. This time a special, separate letter of appeal There are several advantages to having a question- was included in the mailing piece. The new letter naire administered by an interviewer rather than indicated that many students had returned their a respondent. To begin with, interview surveys questionnaires already, and it was very important typically attain higher response rates than mail for all others to do so as well. surveys do. A properly designed and executed interview survey ought to achieve a completion interview A data-collection encounter in which rate of at least 80 to 85 percent. (Federally funded one person (an interviewer) asks questions of an- other (a respondent). Interviews may be conducted face-to-face or by telephone.
Interview Surveys ■ 275 surveys often require one of these response rates.) The interviewer must also fit into this ideal Respondents seem more reluctant to turn down situation. The interviewer’s presence should affect an interviewer standing on their doorstep than to neither a respondent’s perception of a question nor throw away a mail questionnaire. the answer given. In other words, the interviewer should be a neutral medium through which ques- The presence of an interviewer also generally tions and answers are transmitted. decreases the number of “don’t knows” and “no answers.” If minimizing such responses is impor- As such, different interviewers should obtain tant to the study, the interviewer can be instructed exactly the same responses from a given respon- to probe for answers (“If you had to pick one of the dent. (Recall our earlier discussions of reliability.) answers, which do you think would come closest to This neutrality has a special importance in area your feelings?”). samples. To save time and money, a given inter- viewer is typically assigned to complete all the Further, if a respondent clearly misunderstands interviews in a particular geographic area—a city the intent of a question or indicates that he or she block or a group of nearby blocks. If the inter- does not understand, the interviewer can clarify viewer does anything to affect the responses matters, thereby obtaining relevant responses. (As obtained, the bias thus interjected might be inter- we’ll discuss shortly, such clarifications must be preted as a characteristic of that area. strictly controlled through formal specifications.) Let’s suppose that a survey is being done to de- Finally, the interviewer can observe respon- termine attitudes toward low-cost housing in order dents as well as ask questions. For example, the to help in the selection of a site for a new govern- interviewer can note the respondent’s race if this ment-sponsored development. An interviewer is considered too delicate a question to ask. Similar assigned to a given neighborhood might—through observations can be made regarding the quality word or gesture—communicate his or her own of the dwelling, the presence of various posses- distaste for low-cost housing developments. Re- sions, the respondent’s ability to speak English, the spondents might therefore tend to give responses respondent’s general reactions to the study, and in general agreement with the interviewer’s own so forth. In one survey of students, respondents position. The results of the survey would indicate were given a short, self-administered question- that the neighborhood in question strongly resists naire to complete—concerning sexual attitudes construction of the development in its area when and behavior—during the course of the interview. in fact their apparent resistance simply reflects the While respondents completed the questionnaire, interviewer’s attitudes. the interviewer made detailed notes regarding their dress and grooming. General Guidelines for Survey Interviewing This procedure raises an ethical issue. Some researchers have objected that such practices violate The manner in which interviews ought to be con- the spirit of the agreement by which the respondent ducted will vary somewhat by survey population has allowed the interview. Although ethical issues and survey content. Nevertheless, some general seldom are clear-cut in social research, it’s impor- guidelines apply to most interviewing situations. tant to be sensitive to them, as we saw in Chapter 3. Appearance and Demeanor Survey research is of necessity based on an unrealistic stimulus-response theory of cognition As a rule, interviewers should dress in a fashion and behavior. Researchers must assume that a similar to that of the people they’ll be interview- questionnaire item will mean the same thing to ing. A richly dressed interviewer will probably every respondent, and every given response must have difficulty getting good cooperation and re- mean the same when given by different respon- sponses from poorer respondents; a poorly dressed dents. Although this is an impossible goal, survey questions are drafted to approximate the ideal as closely as possible.
276 ■ Chapter 9: Survey Research interviewer will have similar difficulties with richer Ultimately, the interviewer must be able to read respondents. To the extent that the interviewer’s the questionnaire items to respondents without dress and grooming differ from those of the respon- error, without stumbling over words and phrases. dents, it should be in the direction of cleanliness A good model is the actor reading lines in a play or and neatness in modest apparel. If cleanliness is movie. The lines must be read as though they con- not next to godliness, it appears at least to be next stituted a natural conversation, but that conversa- to neutrality. Although middle-class neatness and tion must follow exactly the language set down in cleanliness may not be accepted by all sectors of the questionnaire. U.S. society, they remain the primary norm and are the most likely to be acceptable to the largest By the same token, the interviewer must be number of respondents. familiar with the specifications prepared in con- junction with the questionnaire. Inevitably some Dress and grooming are typically regarded as questions will not exactly fit a given respondent’s signs of a person’s attitudes and orientations. Torn situation, and the interviewer must determine how jeans, green hair, and razor blade earrings may the question should be interpreted in that situa- communicate—correctly or incorrectly—that the tion. The specifications provided to the interviewer interviewer is politically radical, sexually permis- should give adequate guidance in such cases, sive, favorable to drug use, and so forth. Any of but the interviewer must know the organization these impressions could bias responses or affect the and contents of the specifications well enough to willingness of people to be interviewed. refer to them efficiently. It would be better for the interviewer to leave a given question unanswered In demeanor, interviewers should be pleasant than to spend five minutes searching through the if nothing else. Because they’ll be prying into a specifications for clarification or trying to interpret respondent’s personal life and attitudes, they must the relevant instructions. communicate a genuine interest in getting to know the respondent, without appearing to spy. They Following Question Wording Exactly must be relaxed and friendly, without being too casual or clinging. Good interviewers also have the The first part of this chapter discussed the ability to determine very quickly the kind of person significance of question wording for the responses the respondent will feel most comfortable with, the obtained. A slight change in the wording of a given kind of person the respondent would most enjoy question may lead a respondent to answer “yes” talking to. Clearly, the interview will be more suc- rather than “no.” It follows that interviewers must cessful in this case. Further, because respondents be instructed to follow the wording of questions are asked to volunteer a portion of their time and exactly. Otherwise all the effort that the developers to divulge personal information, they deserve the have put into carefully phrasing the questionnaire most enjoyable experience the researcher and items to obtain the information they need and to interviewer can provide. ensure that respondents interpret items precisely as intended will be wasted. Familiarity with the Questionnaire While I hope the logic of this injunction is If an interviewer is unfamiliar with the question- clear, it is not necessarily a closed discussion. For naire, the study suffers and the respondent faces an example, Giampietro Gobo (2006) argues that we unfair burden. The interview is likely to take more might consider giving interviewers more latitude, time than necessary and be unpleasant. Moreover, suggesting that respondents sometimes make errors the interviewer cannot acquire familiarity by skim- that may be apparent to the interviewer on the ming through the questionnaire two or three times. spot. Allowing the interviewer to intervene, as he He or she must study it carefully, question by ques- notes, does increase the possibility that the inter- tion, and must practice reading it aloud. viewer will impact the data collected.
Interview Surveys ■ 277 Recording Responses Exactly respondent, however, may reply: “I think that’s true.” The interviewer should follow this reply with Whenever the questionnaire contains open-ended “Would you say you strongly agree or agree some- questions (ones soliciting the respondent’s own an- what?” If necessary, interviewers can explain that swers), the interviewer must record those answers they must check one or the other of the categories exactly as given. No attempt should be made to provided. If the respondent adamantly refuses to summarize, paraphrase, or correct bad grammar. choose, the interviewer should write in the exact response given by the respondent. This exactness is especially important because the interviewer will not know how the responses Probes are more frequently required in elicit- are to be coded. Indeed, the researchers them- ing responses to open-ended than closed-ended selves may not know the coding until they’ve questions. For example, in response to a question read a hundred or so responses. For example, the about traffic conditions, the respondent might questionnaire might ask respondents how they simply reply, “Pretty bad.” The interviewer could feel about the traffic situation in their community. obtain an elaboration on this response through One respondent might answer that there are too a variety of probes. Sometimes the best probe is many cars on the roads and that something should silence; if the interviewer sits quietly with pencil be done to limit their numbers. Another might say poised, the respondent will probably fill the pause that more roads are needed. If the interviewer re- with additional comments. (This technique is used corded these two responses with the same sum- effectively by newspaper reporters.) Appropriate mary—“congested traffic”—the researchers would verbal probes might be “How is that?” or “In what not be able to take advantage of the important ways?” Perhaps the most generally useful probe is differences in the original responses. “Anything else?” Sometimes, verbal responses are too inarticu- Often, interviewers need to probe for answers late or ambiguous to permit interpretation. How- that will be sufficiently informative for analyti- ever, the interviewer may be able to understand cal purposes. In every case, however, such probes the intent of the response through the respondent’s must be completely neutral; they must not in any gestures or tone. In such a situation, the inter- way affect the nature of the subsequent response. viewer should still record the exact verbal response Whenever you anticipate that a given question but also add marginal comments giving both the may require probing for appropriate responses, interpretation and the reasons for arriving at it. you should provide one or more useful probes next to the question in the questionnaire. This More generally, researchers can use any mar- practice has two important advantages. First, you’ll ginal comments explaining aspects of the response have more time to devise the best, most neutral not conveyed in the verbal recording, such as the probes. Second, all interviewers will use the same respondent’s apparent anger, embarrassment, un- probes whenever they’re needed. Thus, even if the certainty in answering, and so forth. In each case, probe isn’t perfectly neutral, all respondents will however, the exact verbal response should also be be presented with the same stimulus. This is the recorded. same logical guideline discussed for question word- ing. Although a question should not be loaded or Probing for Responses probe A technique employed in interviewing to Sometimes respondents in an interview will give solicit a more complete answer to a question. It is a an inappropriate or incomplete answer. In such nondirective phrase or question used to encourage cases, a probe, or request for an elaboration, can a respondent to elaborate on an answer. Examples be useful. For example, a closed-ended question include “Anything more?” and “How is that?” may present an attitudinal statement and ask the respondent to strongly agree, agree somewhat, disagree somewhat, or strongly disagree. The
278 ■ Chapter 9: Survey Research biased, it’s essential that every respondent be pre- present problems. Suppose a respondent says he or sented with the same question, even if it’s biased. she will be 25 next week. The interviewer might not be sure whether to take the respondent’s cur- Coordination and Control rent age or the nearest one. The specifications for that question should explain what should be done. Most interview surveys require the assistance of (Probably, you would specify that the age as of last several interviewers. In large-scale surveys, inter- birthday should be recorded in all cases.) viewers are hired and paid for their work. Student researchers might find themselves recruiting If you’ve prepared a set of specifications, friends to help them interview. Whenever more review them with the interviewers when you go than one interviewer is involved in a survey, their over the individual questions in the questionnaire. efforts must be carefully controlled. This control Make sure your interviewers fully understand the has two aspects: training interviewers and supervis- specifications and the reasons for them as well as ing them after they begin work. the questions themselves. The interviewers’ training session should begin This portion of the interviewer training is likely with a description of what the study is all about. to generate many troublesome questions from Even though the interviewers may be involved your interviewers. They’ll ask, “What should I do only in the data-collection phase of the project, it if . . . ?” In such cases, avoid giving a quick, offhand will be useful to them to understand what will be answer. If you have specifications, show how the done with the interviews they conduct and what solution to the problem could be determined from purpose will be served. Morale and motivation the specifications. If you do not have specifications, are usually lower when interviewers don’t know show how the preferred handling of the situa- what’s going on. tion fits within the general logic of the question and the purpose of the study. Giving unexplained The training on how to interview should answers to such questions will only confuse the begin with a discussion of general guidelines and interviewers and cause them to take their work less procedures, such as those discussed earlier in this seriously. If you don’t know the answer to such section. Then the whole group should go through a question when it’s asked, admit it and ask for the questionnaire together—question by ques- some time to decide on the best answer. Then tion. Don’t simply ask if anyone has any questions think out the situation carefully and be sure to give about the first page of the questionnaire. Read the all the interviewers your answer, explaining your first question aloud, explain the purpose of the reasons. question, and then entertain any questions or com- ments the interviewers may have. Once all their Once you’ve gone through the whole ques- questions and comments have been handled, go on tionnaire, conduct one or two demonstration to the next question in the questionnaire. interviews in front of everyone. Preferably, you should interview someone other than one of the It’s always a good idea to prepare specifications interviewers. Realize that your interview will be a to accompany an interview questionnaire. model for those you’re training, so make it good. It Specifications are explanatory and clarifying com- would be best, moreover, if the demonstration in- ments about handling difficult or confusing situ- terview were done as realistically as possible. Don’t ations that may occur with regard to particular pause during the demonstration to point out how questions in the questionnaire. When drafting you’ve handled a complicated situation: Handle it, the questionnaire, try to think of all the problem and then explain later. It’s irrelevant if the person cases that might arise—the bizarre circumstances you’re interviewing gives real answers or takes on that might make a question difficult to answer. some hypothetical identity for the purpose, as long The survey specifications should provide detailed as the answers are consistent. guidelines on how to handle such situations. For example, even as simple a matter as age might After the demonstration interviews, pair off your interviewers and have them practice on each
Telephone Surveys ■ 279 other. When they’ve completed the questionnaire, emphasis on practice applies equally to the one- have them reverse roles and do it again. Interview- person project and to the complex funded survey ing is the best training for interviewing. As your in- with a large interviewing staff. terviewers practice on each other, wander around, listening in on the practice so you’ll know how Telephone Surveys well they’re doing. Once the practice is completed, the whole group should discuss their experiences For years telephone surveys had a rather bad and ask any other questions they may have. reputation among professional researchers. By definition, telephone surveys are limited to people The final stage of the training for interview- who have telephones. Years ago, this method ers should involve some “real” interviews. Have produced a substantial social-class bias by exclud- them conduct some interviews under the actual ing poor people from the surveys. This was vividly conditions that will pertain to the final survey. You demonstrated by the Literary Digest fiasco of 1936. may want to assign them people to interview, or Recall that, even though voters were contacted perhaps they may be allowed to pick people them- by mail, the sample was partially selected from selves. Don’t have them practice on people you’ve telephone subscribers, who were hardly typical in selected in your sample, however. After each a nation just recovering from the Great Depression. interviewer has completed three to five interviews, By 2003, however, the U.S. Bureau of the Census have him or her check back with you. Look over (2006: 737, Table 1117) estimated that 95.5 percent the completed questionnaires for any evidence of of all housing units had telephones, so the earlier misunderstanding. Again, answer any questions form of class bias has substantially diminished. that the interviewers have. Once you’re convinced that a given interviewer knows what to do, assign A related sampling problem involves unlisted some actual interviews, using the sample you’ve numbers. A survey sample selected from the pages selected for the study. of a local telephone directory would omit all those people—typically richer—who requested that It’s essential to continue supervising the work their numbers not be published. This potential bias of interviewers over the course of the study. You has been erased through a technique that has ad- should check in with them after they conduct no vanced telephone sampling substantially: random- more than 20 or 30 interviews. You might assign digit dialing (RDD). 20 interviews, have the interviewer bring back those questionnaires when they’re completed, look Imagine that you were to select a set of seven- them over, and assign another 20 or so. Although digit telephone numbers at random. Even those this may seem overly cautious, you must continu- whose numbers were unlisted would have the ally protect yourself against misunderstandings that same chance of selection as those who were in may not be evident early in the study. Moreover, the directory would. However, if you were to start Kristen Olson and Andy Peytchev (2007) have dialing randomly selected numbers, a high propor- discovered that interviewers’ behavior continues tion of those would turn out to be “not in service,” to change over the course of a survey project. For government offices, commercial enterprises, and example, interviewers speed through the interview so forth. Fortunately, you can obtain ranges of more quickly and are more likely to judge respon- numbers that are (mostly) active residential num- dents as uninterested in it. bers. Selecting a set of those numbers at random If you’re the only interviewer in your study, random-digit dialing (RDD) A sampling tech- these comments may not seem relevant. How- nique in which random numbers are selected from ever, it would be wise, for example, to prepare within the ranges of numbers assigned to active specifications for potentially troublesome questions telephones. in your questionnaire. Otherwise, you run the risk of making ad hoc decisions, during the course of the study, that you’ll later regret or forget. Also, the
280 ■ Chapter 9: Survey Research will provide a representative sample of residential the greatest advantages are money and time, in that households. As a consequence, random-digit dial- order. In a face-to-face, household interview, you ing has become a standard procedure in telephone may drive several miles to a respondent’s home, surveys. find no one there, return to the research office, and drive back the next day—possibly finding no The growth in popularity of cell phones has one there again. It’s cheaper and quicker to let your become a new source of concern for survey re- fingers make the trips. searchers, however, since cell phone numbers are typically not included in phone surveys. Those who Interviewing by telephone, you can dress any use cell phones exclusively, moreover, tend to be way you please without affecting the answers re- younger. This, of course, can affect survey out- spondents give. And sometimes respondents will be comes. For example, younger voters in 2004 were more honest in giving socially disapproved answers more likely to vote for John Kerry than older voters if they don’t have to look you in the eye. Similarly, were. In 2008 they were more likely than the aver- it may be possible to probe into more-sensitive age voter to support Barack Obama. Further, in a areas, though this isn’t necessarily the case. People study of this matter, Scott Keeter and his colleagues are, to some extent, more suspicious when they (2008) found a distinct bias by age and the vari- can’t see the person asking them questions— ables closely related to it (such as marital status) perhaps a consequence of “surveys” aimed at distinguishing those who were reachable only by selling magazine subscriptions and time-share cell phone and those reachable by landline: condominiums. One of the most striking differences between Interviewers can communicate a lot about cell-only respondents and people reached on themselves over the phone, however, even though a landline telephone is their age. Nearly half they can’t be seen. For example, researchers of the cell-only respondents (46%) are under worry about the impact of an interviewer’s name age 30 compared to only 12% in the landline (particularly if ethnicity is relevant to the study) sample. Related to their younger age, only 26% and debate the ethics of having all interviewers of cell-only respondents are married, compared use bland “stage names” such as Smith or Jones. with 57% percent of those in the landline (Female interviewers sometimes ask permission to sample. Similarly, about half of cell-only do this, to avoid subsequent harassment from men respondents have never been married (51%), they interview.) compared with only 16% in the landline sample. Telephone surveys can allow greater con- trol over data collection if several interviewers (Keeter et al. 2008) are engaged in the project. If all the interviewers are calling from the research office, they can get At the 2008 meetings of the American Associa- clarification from the person in charge whenever tion for Public Opinion Research (AAPOR), several problems occur, as they inevitably do. Alone in research papers examined the implications of cell the boondocks, an interviewer may have to wing phone popularity. Overall, most of the research- it between weekly visits with the interviewing ers found that, for most purposes, ignoring those supervisor. with only cell phones does not seriously bias survey results, because these customers represent a Finally, another important factor involved in relatively small portion of all telephone customers. the growing use of telephone surveys has to do However, virtually all of the researchers concluded with personal safety. Don Dillman (1978: 4) de- by saying that this situation was likely to change in scribes the situation this way: the years ahead. The role of cell phones is clearly a development that social researchers will continue Interviewers must be able to operate comfort- to examine and deal with. ably in a climate in which strangers are viewed with distrust and must successfully counter Telephone surveys offer many advantages that respondents’ objections to being interviewed. underlie the popularity of this method. Probably Increasingly, interviewers must be willing to
Telephone Surveys ■ 281 work at night to contact residents in many sociodemographic characteristics of owners will households. In some cases, this necessitates change.” This fact made it likely that “different providing protection for interviewers working behavior patterns associated with the utilization of in areas of a city in which a definite threat to the answering machine” could emerge (1991: 216). the safety of individuals exists. More-recent research has shown that sev- Concerns for safety, thus, work in two ways to eral factors, including answering machines, have hamper face-to-face interviews. Potential re- reduced response rates in telephone surveys. Peter spondents may refuse to be interviewed, fearing Tuckel and Harry O’Neill (2002) and others have the stranger-interviewer. And the interviewers examined the impact of such factors as Caller ID, themselves may incur some risks. All this is made answering machines, telemarketing, and phone even worse by the possibility of the researchers be- lines being tied up by faxes and Internet access. ing sued for huge sums if anything goes wrong. All these constitute difficulties modern survey researchers must deal with. Telephone interviewing presents its own problems, however. As I’ve already mentioned, Computer-Assisted Telephone the method is hampered by the proliferation of Interviewing (CATI) bogus “surveys” that are actually sales campaigns disguised as research. If you have any questions In Chapter 14, we’ll see some of the ways comput- about any such call you receive, by the way, ask the ers have influenced the conduct of social re- interviewer directly whether you’ve been selected search—particularly data processing and analysis. for a survey only or if a sales “opportunity” is in- Computers are also changing the nature of tele- volved. It’s also a good idea, if you have any doubts, phone interviewing. One innovation is computer- to get the interviewer’s name, phone number, and assisted telephone interviewing (CATI). This company. Hang up if the caller refuses to provide method is increasingly used by academic, govern- any of these. ment, and commercial survey researchers. Though there are variations in practice, here’s what CATI For the researcher, the ease with which people can look like. can hang up is another shortcoming of telephone surveys. Once you’ve been let inside someone’s Imagine an interviewer wearing a telephone home for an interview, the respondent is unlikely headset, sitting in front of a computer terminal to order you out of the house in midinterview. It’s and its video screen. The central computer selects much easier to terminate a telephone interview a telephone number at random and dials it. (Recall abruptly, saying something like, “Whoops! Some- that random-digit dialing avoids the problem of un- one’s at the door. I gotta go.” or “Omigod! The pigs listed telephone numbers.) On the video screen is are eating my Volvo!” (That sort of thing is much an introduction (“Hello, my name is . . .”) and the harder to fake when the interviewer is sitting in first question to be asked (“Could you tell me how your living room.) many people live at this address?”). Another potential problem for telephone inter- When the respondent answers the phone, the viewing is the prevalence of answering machines or interviewer says hello, introduces the study, and voicemail. A study conducted by Walker Research asks the first question displayed on the screen. (1988) found that half of the owners of answering machines acknowledged using their machines to computer-assisted telephone interviewing “screen” calls at least some of the time. Research by (CATI) A data-collection technique in which a Peter Tuckel and Barry Feinberg (1991), however, telephone-survey questionnaire is stored in a com- showed that answering machines had not yet puter, permitting the interviewer to read the ques- had a significant effect on the ability of telephone tions from the monitor and enter the answers on the researchers to contact prospective respondents. computer keyboard. Nevertheless, the researchers concluded that as answering machines continue to proliferate, “the
282 ■ Chapter 9: Survey Research Voice Capture responses of the young men, young women, middle-aged men, and so forth. In one such study we found the younger and older men tending James E.Dannemiller to watch one TV news show, while the middle-aged men watched something else. Listening to the responses of the middle-aged men, one SMS Research,Honolulu after another, we heard a common comment: “Well, now that I’m older . . .” This kind of aside might have been lost in the notes hastily typed The development of various CATI techniques has been a boon to by interviewers, but such comments stood out dramatically in the oral survey and marketing research,though mostly it has supported data. The middle-aged men seemed to be telling us they felt “maturity” the collection,coding,and analysis of“data as usual.”The Voice Capture required them to watch a particular show, while more years under their technique developed by Survey Systems,however,offers quite unusual belts let them drift back to what they liked in the first place. possibilities,which we are only beginning to explore. These kinds of data are especially compelling to clients, particularly In the course of a CATI-based telephone interview,the interviewer in customer satisfaction studies. Rather than summarize what we feel can trigger the computer to begin digitally recording the conversa- a client’s customers like and don’t like, we can let the respondents tion with the respondent.Having determined that the respondent has speak directly to the client in their own words. It’s like a focus group recently changed his or her favorite TV news show,for example,the on demand. Going one step further, we have found that letting line interviewer can ask,“Why did you change?” and begin recording the employees (bank tellers, for example) listen to the responses has more verbatim response. (Early in the interview, the interviewer has asked impact than having their supervisors tell them what they are doing right permission to record parts of the interview.) or wrong. Later on, coders can play back the responses and code them— As exciting as these experiences are, I have the strong feeling that much as they would do with the interviewer’s typescript of the we have scarcely begun to tap into the possibilities for such unconven- responses. This offers an easier and more accurate way of accomplishing tional forms of data. a conventional task. But that’s a tame use of the new capability. It’s also possible to incorporate such oral data as parts of a cross- tabulation during analysis. We may create a table of gender by age by reasons for switching TV news shows. Thus, we can hear, in turn, the When the respondent answers the question, the It’s also possible to set up questionnaires in interviewer types that answer into the computer personal data assistants (PDAs) for use by an in- terminal—either the verbatim response to an terviewer or for direct data entry by respondents. open-ended question or the code category for the Some of these systems include the possibility of appropriate answer to a closed-ended question. voice capture, as described in the box. The answer is immediately stored in the computer. The second question appears on the video screen, is Response Rates asked, and the answer is entered into the com- in Interview Surveys puter. Thus, the interview continues. Earlier in this chapter we looked at the issue of re- In addition to the obvious advantages in terms sponse rates in mail surveys, and this is an equally of data collection, CATI automatically prepares important issue for interview surveys. In Chapter the data for analysis; in fact, the researcher can 7, when we discussed formulas for calculating sam- begin analyzing the data before the interviewing pling error to determine the accuracy of survey es- is complete, thereby gaining an advanced view of timates, the implicit assumption was that everyone how the analysis will turn out. Sill another innova- selected in a sample would participate—which is tion that computer technology makes possible is described in “Voice Capture.”
Online Surveys ■ 283 almost never the case. Lacking perfection, research- evidenced by a fax entitled “Should Hand Guns ers must maximize participation by those selected. Be Outlawed?” Two fax numbers were provided Although interview surveys tend to produce higher for expressing either a “Yes” or “No” opinion. The response rates than mail surveys do, interview suc- smaller print noted, “Calls to these numbers cost cess has recently declined. $2.95 per minute, a small price for greater democ- racy. Calls take approx. 1 or 2 minutes.” You can By analyzing response-rate trends in the imagine where the $2.95 went. University of Michigan’s Survey of Consumer Attitudes, Richard Curtin, Stanley Presser, and Online Surveys Eleanor Singer (2005) have sketched a pattern of general decline over recent years. Between 1979 An increasingly popular method of survey re- and 1996, the response rate in this telephone sur- search involves the use of the Internet and the vey dropped from 72 to 60 percent, representing an World Wide Web—two of the most far-reaching average annual decline of three-quarters of a per- developments of the late twentieth century. Some cent. Since 1996, the rate of decline has doubled. researchers feel that the Internet can be used to The increased non-responses reflected both refusals conduct meaningful survey research, and this and those they were unable to contact. technique has been getting especially popular in marketing research, for example. Some online sur- By contrast, the General Social Survey, using veys are conducted completely via e-mail; others personal interviews, experienced response rates are conducted via websites. Commonly, potential between 73.5 and 82.4 percent in the years from respondents will receive an e-mail asking them to 1975 to 1998. In the 2000 and 2002 surveys, go to a web link where the survey resides. however, the GSS completion rate was 70 percent. Their decline came primarily from refusals rather Mick Couper (2001: 464) provides an excellent than being unable to contact respondents, because overview of the issues concerning the present and household interviews produce higher rates of con- prospective state of online surveys. tact than telephone surveys do. The rapid development of surveys on the World In recent years, both household and telephone Wide Web (WWW) is leading some to argue surveys have experienced a decline in response that soon Internet (and, in particular, Web) rates. A special issue of the Public Opinion Quarterly surveys will replace traditional methods of sur- (2006) was devoted entirely to analyzing the many vey data collection. Others are urging caution dimensions of the decline in response rates in or even voicing skepticism about the future household surveys. As the analyses show, lower re- role Web surveys will play. Clearly, we stand at sponse rates do not necessarily produce inaccurate the threshold of a new era for survey research, estimates of the population being studied, but the but how this will play out is not yet clear. variations on this issue defy a simple summary. As we’ve seen, one immediate objection that Many researchers believe that the widespread many social researchers make to online surveys growth of telemarketing has been a big part of the concerns representativeness: Will the people who problems experienced by legitimate telephone sur- can be surveyed online be representative of mean- veys, and there are hopes that the state and national ingful populations, such as all U.S. adults, all voters, “do not call” lists may ease that problem. Further, and so on? This is the criticism raised with regard we saw that other factors such as answering mach- to surveys via fax or by telephone interviewers. ines and voicemail also contribute to these problems (Tuckel and O’Neill 2002). Response rate is likely to Camilo Wilson (1999), founder of Cogix (see remain an issue of high concern in survey research. the link on this book’s website), points out that some populations are ideally suited to online As a consumer of social research, you should surveys: specifically, those who visit a particular be wary of “surveys” whose apparent purpose is to raise money for the sponsor. This practice has already invaded the realm of “fax surveys,”
284 ■ Chapter 9: Survey Research website. For example, Wilson indicates that market Harris Interactive has demonstrated success in research for online companies should be conducted predicting election results. online, and his firm has developed software called ViewsFlash for precisely that purpose. Although Many of the cautions urged in relation to website surveys could easily collect data from all online surveys today are similar to those urged in who visit a particular site, Wilson suggests that relation to telephone surveys in the first edition survey-sampling techniques can provide sufficient of this book, in 1975. Mick Couper (2001: 466) consumer data without irritating thousands or mil- makes a similar observation: lions of potential customers. Several years ago, I predicted that the rapid But how about general population surveys? spread of electronic data collection methods How about political polling? These are probably such as the Internet would produce a bifur- the main issues raised regarding online surveys cation in the survey industry between high- today. Not everyone of interest can be reached via quality surveys based on probability samples Internet nor feels comfortable using it for partici- and using traditional data collection methods, pation in surveys. Moreover, people who are less on the one hand, and surveys focused more available to online surveys do not represent a ran- on low cost and rapid turnaround than on rep- dom segment of the overall population. The poor resentativeness and accuracy on the other. In and the elderly, for example, are likely to be under- hindsight, I was wrong, and I underestimated represented in online surveys. At the same time, as the impact of the Web on the survey industry. more and more people gain access to the Internet, It has become much more of a fragmentation this problem will decline. (An early criticism of than a bifurcation (in terms of Web surveys at telephone surveys was that not everyone had a least), with vendors trying to find or create a phone.) niche for their particular approach or product. No longer is it just “quick and dirty” in one In one solution to this problem, the National corner and “expensive but high quality” in Opinion Research Center, which conducts the peri- the other; rather, there is a wide array of ap- odic General Social Survey (GSS), used probability proaches representing varying levels of quality sampling methods to create a representative sample and cost. of potential respondents (T. Smith 2001). Each person in the sample was provided with WebTV In the meantime, researchers are amassing a access to the Internet, with an agreement that body of experience with this new technique, yield- they would participate in polls from time to time. ing lessons for increasing success. For example, While these online respondents were demographi- Survey Sampling, Inc., suggests the following dos cally representative, there were differences in their and don’ts for conducting online surveys: responses on survey issues that will require further study. For example, the online respondents were Do use consistent wording between the invita- more likely to choose extreme responses (such as tion and the survey. Don’t use terms such as “strongly agree”) than those surveyed in face-to- “unique ID number” in the invitation, then ask face interviews were. respondents to type their “password” when they get to the survey. Changing terminology Commercial research firms, such as Harris can be confusing. Interactive and Knowledge Networks (see the links on this book’s website) report they have Do use plain, simple language. developed large-scale panels of online respondents from whom they are able to select samples that Don’t force the respondent to scroll down the are representative of whatever populations are of screen for the URL for the study location. interest for study. Because their specific methods are proprietary, assessing their methodological Do offer to share selected results from the strengths and weaknesses is difficult. However, study with everyone who completes the survey. Respondents will often welcome information as
Comparison of the Different Survey Methods ■ 285 a reward for taking the study, especially when The Gallup Organization they are young adults and teens. SMS Research Do plan the time of day and day of week to mail, depending on the subject of the study The Survey/Marketing Research e-Store and type of respondent. Send the invitation late afternoon, evening, or weekend, when Zogby International respondents are most likely to be reading mail at home, especially if the study requests Online surveys appear to have response rates respondents to check an item in the kitchen approximately comparable to mail surveys, ac- or other area in the home. If a parent-child cording to a large-scale study of Michigan State questionnaire is planned, send the invita- University students (Kaplowitz, Hadlock, and tion late afternoon when children are home, Levine 2004), especially when the online survey not early in the day, when respondents can’t is accompanied by a postcard reminder encourag- complete the study because children are at ing respondents to participate. While producing a school. comparable response rate, the online survey costs substantially less than a conventional mail survey. Do be aware of technical limitations. For The cost of paper, printing, and postage alone can example, WebTV users currently cannot access constitute a large expense. surveys using Java. If respondents’ systems need to be Java-enabled or require access to In another study of ways to improve re- streaming video, alert panelists at the beginning sponse rates in online surveys, Stephen Porter of the study, not midway through. and Michael Whitcomb (2003) found that some of the techniques effective in mail surveys, such Do test incentives, rewards, and prize draw- as personalizing the appeal or varying the appar- ings to determine the optimal offer for best ent status of the researcher, had little or no impact response. Longer surveys usually require larger in the new medium. At the same time, specifying incentives. that the respondents had been specially selected for the survey and setting a deadline for participa- Do limit studies to 15 minutes or less.* tion did increase response rates. The years ahead will see many experiments aimed at improving the Over the years, members of industrialized na- effectiveness of online surveys. See “How to Do It: tions have become familiar with the format and Conducting an Online Survey” for more. process of self-administered questionnaires, but the web presents a new challenge for many. Leah During the 2004 election campaign, the role Christian, Don Dillman, and Jolene Smyth (2007) and nature of polls—in-person, telephone, and provide a wealth of guidance on the formatting online—drew considerable public attention. Mark of web surveys. Their aim is, as their article title Blumenthal (2005) reviewed the public discussion suggests, “helping respondents get it right the first and considered the implications for the future of time.” polling. Clearly, this discussion will not end any time soon. The web is already seeing extensive use as a marketplace for surveys and other research tech- Comparison of the Different niques. As only a few illustrative examples, see the Survey Methods following links on this book’s website: http://www .cengage.com/sociology/babbie: Now that we’ve seen several ways to collect survey data, let’s take a moment to compare them directly. *Source: http://www.worldopinion.com/the_frame/ frame4.html. Reprinted with permission. Self-administered questionnaires are gener- ally cheaper and quicker than face-to-face inter- view surveys. These considerations are likely to be
286 ■ Chapter 9: Survey Research How to Do It: Conducting an Online Survey If you’re interested in conducting an online survey,you can experiment The program is quite user-friendly with regard to designing with a limited version of an online program called Survey Monkey,at questionnaire items.To reach your intended respondents,you enter their no charge.To get started,go to the Survey Monkey link on this book’s e-mail addresses,and they then receive an e-mail invitation to visit the website (http://www.cengage.com/sociology/babbie) and click survey web page and participate.The free beginner package will also “Create Survey.” provide you with a basic analysis of the survey results. important for an unfunded student wishing to un- With interviews, you can conduct a survey dertake a survey for a term paper or thesis. More- based on a sample of addresses or phone numbers over, if you use the self-administered mail format, rather than on names. An interviewer can arrive it costs no more to conduct a national survey than at an assigned address or call the assigned num- a local one of the same sample size. In contrast, a ber, introduce the survey, and even—following national interview survey (either face-to-face or instructions—choose the appropriate person at by telephone) would cost far more than a local one. that address to respond to the survey. In contrast, Also, mail surveys typically require a small staff: self-administered questionnaires addressed to “oc- You could conduct a reasonable mail survey by cupant” receive a notoriously low response. yourself, although you shouldn’t underestimate the work involved. Further, respondents are sometimes Finally, as we’ve seen, interviewers questioning reluctant to report controversial or deviant attitudes respondents face-to-face can make important ob- or behaviors in interviews but are willing to respond servations aside from responses to questions asked to an anonymous self-administered questionnaire. in the interview. In a household interview, they may note the characteristics of the neighborhood, Interview surveys also offer many advan- the dwelling unit, and so forth. They can also note tages. For example, they generally produce fewer characteristics of the respondents or the quality of incomplete questionnaires. Although respondents their interaction with the respondents—whether may skip questions in a self-administered question- the respondent had difficulty communicating, was naire, interviewers are trained not to do so. In CATI hostile, seemed to be lying, and so on. surveys, the computer offers a further check on this. Interview surveys, moreover, have typically The chief advantages of telephone surveys over achieved higher completion rates than self-admin- those conducted face-to-face center primarily on istered questionnaires have. time and money. Telephone interviews are much cheaper and can be mounted and executed quickly. Although self-administered questionnaires may Also, interviewers are safer when interviewing be more effective for sensitive issues, interview sur- people living in high-crime areas. Moreover, the veys are definitely more effective for complicated impact of the interviewers on responses is some- ones. Prime examples include the enumeration what lessened when the respondents can’t see of household members and the determination of them. As only one indicator of the popularity of whether a given address corresponds to more than telephone interviewing, when Johnny Blair and one housing unit. Although the concept of housing his colleagues (1995) compiled a bibliography on unit has been refined and standardized by the Cen- sample designs for telephone interviews, they listed sus Bureau and interviewers can be trained to deal over 200 items. with the concept, it’s extremely difficult to com- municate this idea in a self-administered question- Online surveys have many of the strengths naire. This advantage of interview surveys pertains and weaknesses of mail surveys. Once the avail- generally to all complicated contingency questions. able software has been further developed, they will likely be substantially cheaper. An important weak-
Strengths and Weaknesses of Survey Research ■ 287 ness, however, lies in the difficulty of assuring that generally. Earlier chapters have discussed the respondents to an online survey will be representa- ambiguous nature of most concepts: They have no tive of some more general population. ultimately real meanings. One person’s religios- ity is quite different from another’s. Although you Clearly, each survey method has its place in must be able to define concepts in those ways most social research. Ultimately, you must balance the relevant to your research goals, you may not find advantages and disadvantages of the different it easy to apply the same definitions uniformly to methods in relation to your research needs and all subjects. The survey researcher is bound to this your resources. requirement by having to ask exactly the same questions of all subjects and having to impute the Strengths and Weaknesses same intent to all respondents giving a particular of Survey Research response. Regardless of the specific method used, surveys— Survey research also has several weaknesses. like other modes of observation in social research— First, the requirement of standardization often have special strengths and weaknesses. You should seems to result in the fitting of round pegs into keep these in mind when determining whether a square holes. Standardized questionnaire items survey is appropriate for your research goals. often represent the least common denominator in assessing people’s attitudes, orientations, circum- Surveys are particularly useful in describing stances, and experiences. By designing questions the characteristics of a large population. A carefully that will be at least minimally appropriate to all re- selected probability sample in combination with spondents, you may miss what is most appropriate a standardized questionnaire offers the possibil- to many respondents. In this sense, surveys often ity of making refined descriptive assertions about appear superficial in their coverage of complex a student body, a city, a nation, or any other large topics. Although this problem can be partly offset population. Surveys determine unemployment by sophisticated analyses, it is inherent in survey rates, voting intentions, and so forth with uncanny research. accuracy. Although the examination of official documents—such as marriage, birth, or death Similarly, survey research can seldom deal with records—can provide equal accuracy for a few the context of social life. Although questionnaires topics, no other method of observation can provide can provide information in this area, the survey this general capability. researcher rarely develops the feel for the total life situation in which respondents are thinking and Surveys—especially self-administered ones— acting that, say, the participant observer can (see make large samples feasible. Surveys of 2,000 re- Chapter 10). spondents are not unusual. A large number of cases is very important for both descriptive and explana- In many ways, surveys are inflexible. Stud- tory analyses, especially wherever several variables ies involving direct observation can be modified are to be analyzed simultaneously. as field conditions warrant, but surveys typically require that an initial study design remain un- In one sense, surveys are flexible. Many ques- changed throughout. As a field researcher, for tions can be asked on a given topic, giving you example, you can become aware of an important considerable flexibility in your analyses. Whereas new variable operating in the phenomenon you’re an experimental design may require you to com- studying and begin making careful observations of mit yourself in advance to a particular operational it. The survey researcher would probably be un- definition of a concept, surveys let you develop aware of the new variable’s importance and could operational definitions from actual observations. do nothing about it in any event. Finally, standardized questionnaires have an Finally, surveys are subject to the artificiality important strength in regard to measurement mentioned earlier in connection with experi- ments. Finding out that a person gives conservative
288 ■ Chapter 9: Survey Research answers in a questionnaire does not necessarily As with all methods of observation, a full mean the person is conservative; finding out that a awareness of the inherent or probable weaknesses person gives prejudiced answers in a questionnaire of survey research can partially resolve them in does not necessarily mean the person is prejudiced. some cases. Ultimately, though, researchers are on This shortcoming is especially salient in the realm the safest ground when they can employ several of action. Surveys cannot measure social action; research methods in studying a given topic. they can only collect self-reports of recalled past ac- tion or of prospective or hypothetical action. Secondary Analysis The problem of artificiality has two aspects. As a mode of observation, survey research involves First, the topic of study may not be amenable to the following steps: (1) questionnaire construc- measurement through questionnaires. Second, the tion, (2) sample selection, and (3) data collection, act of studying that topic—an attitude, for exam- through either interviewing or self-administered ple—may affect it. A survey respondent may have questionnaires. As you’ve gathered, surveys are given no thought to whether the governor should usually major undertakings. It’s not unusual for a be impeached until asked for his or her opinion by large-scale survey to take several months or even an interviewer. He or she may, at that point, form more than a year to progress from conceptualiza- an opinion on the matter. tion to data in hand. (Smaller-scale surveys can, of course, be done more quickly.) Through a method Survey research is generally weak on validity called secondary analysis, however, researchers and strong on reliability. In comparison with field can pursue their particular social research inter- research, for example, the artificiality of the survey ests—analyzing survey data from, say, a national format puts a strain on validity. As an illustration, sample of 2,000 respondents—while avoiding the people’s opinions on issues seldom take the form enormous expenditure of time and money such a of strongly agreeing, agreeing, disagreeing, or survey entails. strongly disagreeing with a specific statement. Their survey responses in such cases must be regarded Secondary analysis is a form of research in as approximate indicators of what the researchers which the data collected and processed by one had in mind when they framed the questions. This researcher are reanalyzed— often for a different comment, however, needs to be held in the context purpose—by another. Beginning in the 1960s, of earlier discussions of the ambiguity of validity survey researchers became aware of the potential itself. To say something is a valid or an invalid mea- value that lay in archiving survey data for analy- sure assumes the existence of a “real” definition of sis by scholars who had nothing to do with the what’s being measured, and many scholars now survey design and data collection. Even when one reject that assumption. researcher had conducted a survey and analyzed the data, those same data could be further analyzed Reliability is a clearer matter. By presenting by others who had slightly different interests. Thus, all subjects with a standardized stimulus, survey if you were interested in the relationship between research goes a long way toward eliminating un- political views and attitudes toward gender equal- reliability in observations made by the researcher. ity, you could examine that research question Moreover, careful wording of the questions can also through the analysis of any data set that happened significantly reduce the subject’s own unreliability. to contain questions relating to those two variables. secondary analysis A form of research in which The initial data archives were very much like the data collected and processed by one researcher book libraries, with a couple of differences. First, are reanalyzed— often for a different purpose—by instead of books, the data archives contained another. This is especially appropriate in the case of data sets: first as punched cards, then as magnetic survey data. Data archives are repositories or librar- tapes. Today they’re typically contained on com- ies for the storage and distribution of data for sec- ondary analysis.
Secondary Analysis ■ 289 puter disks, CD-ROMs, or online servers. Second, FIGURE 9-7 whereas you’re expected to return books to a con- Subject Index for the General Social Survey ventional library, you can keep the data obtained from a data archive. might get away with an offhand, unsubstantiated assertion, imagine how much more powerful your The best-known current example of a resource paper would be with the following addition. for secondary analysis is the General Social Survey (GSS). The National Opinion Research Center 1. Access the General Social Survey data set at the (NORC) at the University of Chicago conducts this National Opinion Research Center, at the link major national survey, currently every other year, on this book’s website. to collect data on a large number of social science variables. These surveys are conducted precisely for 2. Follow the instructions by clicking successively, the purpose of making data available to scholars at in the left column, little or no cost and are supported by a combina- a. The plus sign beside “GSS” tion of private and government funding. Recall that b. The plus sign beside “General Social Sur- the GSS was created by James A. Davis in 1972; veys, 1972–2006” it is currently directed by Davis, Tom W. Smith, c. The plus sign beside “Variable Description” and Peter V. Marsden. Their considerable ongo- d. The plus sign beside “Subject Index” ing efforts make an unusual contribution to social science research and to education in social science. Your screen should now contain the excerpt shown You can learn more about the GSS at the link on in Figure 9-7. this book’s website: http://www.cengage.com/ sociology/babbie. In the header to the right of this excerpt on your screen, you’ll see three tabs: DESCRIPTION, Numerous other resources are available for TABULATION, AND ANALYSIS. You’ll see that identifying and acquiring survey data for second- DESCRIPTION has been selected. To begin creating ary analysis. The Roper Center for Public Opinion Research at the University of Connecticut is one excellent resource. The center also publishes the journal Public Perspective, on public opinion poll- ing. Polling the Nations is an online repository for thousands of polls conducted in the United States and 70 other nations. A paid subscription allows users to obtain specific data results from studies they specify, rather than obtaining whole studies. Although the cost of the subscription may be too steep for the average student, you might check to see if your school’s library has subscribed. (See the links for the Roper Center and Polling the Nations on this book’s website.) Because secondary analysis has typically involved obtaining a data set and undertaking an extensive analysis, I would like you to consider another approach as well. Often you can do limited analyses by investing just a little time. Let’s say you’re writing a term paper about the impact of religion in contemporary American life. You want to comment on the role of the Roman Catholic church in the debate over abortion. Although you
290 ■ Chapter 9: Survey Research FIGURE 9-8 Instructions for Creating a Table the analysis for your term paper, click the TABULA- A TION tab. A portion of your screen will now look like Figure 9-8. Abortion Let’s select “Religious Preference” as the col- In case of . . . umn variable in our table. You’ll have to explore a bit, beginning with the plus sign beside the letter any reason R in the Subject Index. Once you’ve opened that letter, open the entry for “Religion.” Then open This will show the GSS questionnaire items “R’s Religious Preference.” Finally, open “Respon- that asked respondents whether a woman should dent (current).” I know these are a lot of steps, have the right to a legal abortion for any rea- but it demonstrates the wealth of data available for son. Opening the arrow next to ABORTION IF analysis. Once you get more familiar with the data WOMAN WANTS FOR ANY REASON should give set and the program, you’ll move around deftly. you the display shown in Figure 9-10. At this point, you’ll find several options under Click on the phrase “add to row” and you’ll be “Respondent (current).” We’re interested in the presented with a two-variable table, a portion of first of these: RS RELIGIOUS PREFERENCE. Notice which is shown in Figure 9-11. (I’ve limited the that instead of a plus sign, this notation is preceded display to those religious preferences with more by a right-pointing arrow. Click on the arrow, and than a few responses.) In reviewing these analytic you should see the display shown in Figure 9-9. results, you should realize that they represent the cumulation of surveys done between 1972 and Click on the phrase “Add to column.” Your 2006. If you want to limit your analysis to a single table is halfway complete. To add attitudes toward year, instructions on the website will tell you how abortion, repeat the process just outlined but select to do that. the following: The results of this analysis may surprise you. Whereas Catholics are less supportive of abortion
Secondary Analysis ■ 291 FIGURE 9-10 Selecting ABORTION IF WOMAN WANTS FOR ANY REASON as the Row Variable FIGURE 9-9 Selecting RS RELIGIOUS PREFERENCE as the Column Variable (35.4 percent) than Jews (77.8 percent) and those FIGURE 9-11 with no religion (66.3 percent) are, they do not The Relationship between Religion and Attitudes toward differ substantially from Protestants (36.0 percent). Abortion Imagine a term paper that says, “Whereas that the original researcher asked a question that the Roman Catholic Church has taken a strong, “comes close” to measuring what you’re interested official position on abortion, many Catholics do in, but you’ll wish the question had been asked not necessarily agree, as shown in Table . . .“ More- just a little differently— or that another, related over, this might be just the beginning of an analysis question had also been asked. For example, you that looks a bit more deeply into the matter, as may want to study how religious various people described in Chapter 14, on quantitative analysis. are and the survey data available to you only asked about attendance at worship services. Your quan- The key advantage of secondary analysis is that dary, then, is whether the question that was asked it’s cheaper and faster than doing original surveys, provides a valid measure of the variable you want and, depending on who did the original survey, to analyze. Nevertheless, secondary analysis can you may benefit from the work of topflight profes- be immensely useful. Moreover, it illustrates once sionals. The ease of secondary analysis has also again the range of possibilities available in finding enhanced the possibility of meta-analysis, in which the answers to questions about social life. Although a researcher brings together a body of past research no single method unlocks all puzzles, there is no on a particular topic. To gain confidence in your limit to the ways you can find out about things. understanding of the relationship between religion And when you zero in on an issue from several and abortion, for example, you could go beyond independent directions, you gain that much more the GSS to analyze similar data collected in dozens expertise. or even hundreds of other studies. I’ve discussed secondary analysis in this chapter There are disadvantages inherent in secondary on survey research because it’s the type of analy- analysis, however. The key problem involves the sis most associated with the technique. However, recurrent question of validity. When one researcher there is no reason that the reanalysis of social collects data for one particular purpose, you have no assurance that those data will be appropriate for your research interests. Typically, you’ll find
292 ■ Chapter 9: Survey Research research data needs to be limited to those collected • Questionnaires provide a method of collecting in surveys. Nigel Fielding (2004), for example, has examined the possibilities for the archiving and data by (1) asking people questions or (2) asking reanalysis of qualitative data as well. them to agree or disagree with statements repre- senting different points of view. Questions may Ethics and Survey Research be open-ended (respondents supply their own answers) or closed-ended (they select from a list Survey research almost always involves a request of provided answers). that people provide us with information about the mselves that is not readily available. Sometimes, we Guidelines for Asking Questions ask for information (about attitudes and behaviors, for example) that would be embarrassing to the • Items in a questionnaire should follow several respondents if that information became publicly known. In some cases, such revelations could result guidelines: (1) The form of the items should be in the loss of a job or a marriage. Hence, maintain- appropriate to the project; (2) the items must be ing the norm of confidentiality, mentioned earlier clear and precise; (3) the items should ask only in the book, is particularly important in survey about one thing (that is, double-barreled ques- research. tions should be avoided); (4) respondents must be competent to answer the item; (5) respondents Another ethical concern relates to the possibil- must be willing to answer the item; (6) questions ity of psychological injury to respondents. Even if should be relevant to the respondent; (7) items the information they provide is kept confidential, should ordinarily be short; (8) negative terms simply forcing them to think about some matters should be avoided so as not to confuse respon- can be upsetting. Imagine asking people for their dents; (9) the items should be worded to avoid attitudes toward suicide when one of them has re- biasing responses. cently experienced the suicide of a family member or close friend. Or asking people to report on their Questionnaire Construction attitudes about different racial groups, which may cause them to reflect on whether they might be • The format of a questionnaire can influence the racists or at least appear as such to the interviewers. The possibilities for harming survey respondents quality of data collected. are endless. While this fact should not prevent you from doing surveys, it should increase your consid- • A clear format for contingency questions is neces- ered efforts to avoid the problem wherever possible. sary to ensure that the respondents answer all the MAIN POINTS questions intended for them. Introduction • The matrix question is an efficient format for pre- • Survey research, a popular social research senting several items sharing the same response categories. method, is the administration of questionnaires to a sample of respondents selected from some • The order of items in a questionnaire can population. influence the responses given. Topics Appropriate for Survey Research • Clear instructions are important for getting ap- • Survey research is especially appropriate for mak- propriate responses in a questionnaire. ing descriptive studies of large populations; survey data may be used for explanatory purposes as well. • Questionnaires should be pretested before being administered to the study sample. • Questionnaires are usually administered in one of three main ways: through self-administered ques- tionnaires, face-to-face interviews, or telephone surveys. Researchers are exploring online surveys as well. Self-Administered Questionnaires • It’s generally advisable to plan follow-up mailings in the case of self-administered questionnaires, sending new questionnaires to those respondents who fail to respond to the initial appeal. Properly monitoring questionnaire returns will provide a good guide to when a follow-up mailing is appropriate.
Key Terms ■ 293 Interview Surveys cheaper to conduct, ensuring that the respon- dents represent a more general population can be • Interviewers must be neutral in appearance and difficult. actions; their presence in the data-collection pro- Strengths and Weaknesses cess must have no effect on the responses given to of Survey Research questionnaire items. • Survey research in general offers advantages in • Interviewers must be carefully trained to be famil- terms of economy, the amount of data that can be iar with the questionnaire, to follow the question collected, and the chance to sample a large popu- wording and question order exactly, and to record lation. The standardization of the data collected responses exactly as they are given. represents another special strength of survey research. • Interviewers can use probes to elicit an elabora- • Survey research has several weaknesses: It is some- tion on an incomplete or ambiguous response. Probes should be neutral. Ideally, all interviewers what artificial, potentially superficial, and relati- should use the same probes. vely inflexible. Using surveys to gain a full sense of social processes in their natural settings is difficult. Telephone Surveys In general, survey research is comparatively weak on validity and strong on reliability. • Telephone surveys can be cheaper and more Secondary Analysis efficient than face-to-face interviews, and they can permit greater control over data collection. • Secondary analysis provides social research- • Random-digit dialing (RDD) is a useful technique ers with an important option for “collecting” data cheaply and easily but at a potential cost in for eliminating potential bias in selecting numbers. validity. • The development of computer-assisted telephone Ethics and Survey Research interviewing (CATI) is especially promising. • Surveys often ask for private information, Online Surveys and researchers must keep such information confidential. • New technologies offer additional opportunities • Because asking questions can cause psychological for social researchers, surveys over the Internet. This method, however, must be used with caution discomfort or harm to respondents, the researcher because respondents may not be representative of should minimize this risk. the intended population. KEY TERMS Comparison of the Different Survey Methods The following terms are defined in context in the chapter and at the bottom of the page where the term • The advantages of a self-administered question- is introduced, as well as in the comprehensive glossary at the back of the book. naire over an interview survey are economy, speed, lack of interviewer bias, and the possibility bias probe of anonymity and privacy to encourage candid closed-ended questions questionnaire responses on sensitive issues. random-digit computer-assisted telephone dialing (RDD) • The advantages of an interview survey over a interviewing (CATI) contingency question respondent self-administered questionnaire are fewer incom- interview response rate plete questionnaires and fewer misunderstood open-ended questions secondary analysis questions, generally higher completion rates, and greater flexibility in terms of sampling and special observations. • The principal advantages of telephone surveys over face-to-face interviews are the savings in cost and time. Telephone interviewers have more safety than in-person interviewers do, and they may have a smaller effect on the interview itself. • Online surveys have many of the strengths and weaknesses of mail surveys. Although they’re
294 ■ Chapter 9: Survey Research PROPOSING SOCIAL RESEARCH: Consider at least five of the questions in it and SURVEY RESEARCH critique each one. 4. Look at your appearance right now. Identify If you’re planning a survey, you’ll have already aspects of your appearance that might create a described the sampling you’ll employ, and your problem if you were interviewing a general cross discussion of measurement will have presented at section of the public. least portions of your questionnaire. At this point you 5. Locate a survey being conducted on the web. need to describe the type of survey you’ll conduct: Briefly describe the survey and discuss its self-administered, telephone, face-to-face, or Internet. strengths and weaknesses. Whichever you plan, there will be numerous logistical details to spell out in the proposal. How will you deal SPSS EXERCISES with non-respondents, for example? Will you have follow-up mailing in a self-administered question- See the booklet that accompanies your text for ex- naire, follow-up calls in a telephone survey, and so ercises using SPSS (Statistical Package for the Social forth? Will you have a target completion rate? Sciences). There are exercises offered for each chapter, and you’ll also find a detailed primer on using SPSS. In the case of interview surveys, you should say something about the way you’ll select and train the Online Study Resources interviewers. You should also say something about the time frame within which the survey will be If your book came with an access code card, visit conducted. www.cengage.com/login to register. To purchase access, please visit www.ichapters.com. REVIEW QUESTIONS AND EXERCISES 1. Before you do your final review of the chapter, 1. For each of the following open-ended questions, take the CengageNOW pretest to help identify the construct a closed-ended question that could be areas on which you should concentrate. You’ll used in a questionnaire. find information on this online tool, as well as instructions on how to access all of its great re- a. What was your family’s total income last year? sources, in the front of the book. 2. As you review, take advantage of the Cengage- b. How do you feel about the space shuttle NOW personalized study plan, based on your quiz program? results. Use this study plan with its interactive ex- ercises and other resources to master the material. c. How important is religion in your life? 3. When you’re finished with your review, take the posttest to confirm that you’re ready to move on d. What was your main reason for attending to the next chapter. college? WEBSITE FOR THE PRACTICE e. What do you feel is the biggest problem facing OF SOCIAL RESEARCH 12TH EDITION your community? Go to your book’s website at www.cengage.com/ 2. Construct a set of contingency questions for use sociology/babbie for tools to aid you in studying for in a self-administered questionnaire that would your exams. You’ll find Tutorial Quizzes with feed- solicit the following information: back, Internet Exercises, Flash Cards, Glossaries, and Essay Quizzes, as well as InfoTrac College Edition search terms, a. Is the respondent employed? suggestions for additional reading, Web Links, and primers for using data-analysis software such as SPSS. b. If unemployed, is the respondent looking for work? c. If the unemployed respondent is not looking for work, is he or she retired, a student, or a homemaker? d. If the respondent is looking for work, how long has he or she been looking? 3. Find a questionnaire printed in a magazine or newspaper (for a reader survey, for example).
CHAPTER 10 Qualitative Field Research CHAPTER OVERVIEW Introduction Institutional Ethnography Participatory Action Qualitative field research enables Topics Appropriate for Field Research researchers to observe social life in Research its natural habitat:to go where the Conducting Qualitative Field action is and watch.This type of Special Considerations in Research research can produce a richer Qualitative Field Research understanding of many social Preparing for the Field phenomena than can be achieved The Various Roles of the Qualitative Interviewing through other observational Observer Focus Groups methods,provided that the Relations to Subjects Recording Observations researcher observes in a deliberate, well-planned,and active way. Some Qualitative Field Strengths and Weaknesses of Research Paradigms Qualitative Field Research Naturalism Validity Ethnomethodology Reliability Grounded Theory Case Studies and the Ethics and Qualitative Field Extended Case Method Research CengageNOW for Sociology Use this online tool to help you make the grade on your next exam. After reading this chapter, go to “Online Study Resources” at the end of the chapter for instructions on how to benefit from CengageNOW.
296 ■ Chapter 10: Qualitative Field Research Introduction numbers. Thus, for example, a field researcher may note the “paternalistic demeanor” of leaders at a Several chapters ago, I suggested that you’ve been political rally or the “defensive evasions” of a public doing social research all your life. This idea should official at a public hearing without trying to express become even clearer as we turn to what probably either the paternalism or the defensiveness as a seems like the most obvious method of making numerical quantity or degree. Although field re- observations: qualitative field research. In a sense, search can be used to collect quantitative data—for we do field research whenever we observe or par- example, noting the number of interactions of vari- ticipate in social behavior and try to understand it, ous specified types within a field setting—typically, whether in a college classroom, in a doctor’s wait- field research is qualitative. ing room, or on an airplane. Whenever we report our observations to others, we’re reporting our field Field observation also differs from some other research efforts. models of observation in that it’s not just a data- collecting activity. Frequently, perhaps typically, Such research is at once very old and very it’s a theory-generating activity as well. As a field new in social science, stretching at least from the researcher, you’ll seldom approach your task nineteenth-century studies of preliterate societies, with precisely defined hypotheses to be tested. through firsthand examinations of urban commu- More typically, you’ll attempt to make sense out nity life in the “Chicago School” of the 1930s and of an ongoing process that cannot be predicted in 1940s, to contemporary observations of chat-room advance—making initial observations, developing interactions on the web. Many of the techniques tentative general conclusions that suggest particu- discussed in this chapter have been used by social lar types of further observations, making those researchers for centuries. Within the social sci- observations and thereby revising your conclusions, ences, anthropologists are especially associated and so forth. In short, the alternation of induction with this method and have contributed to its and deduction discussed in Part 1 of this book is development as a scientific technique. Moreover, perhaps nowhere more evident and essential than something similar to this method is employed by in good field research. For expository purposes, many people who might not, strictly speaking, be however, this chapter focuses primarily on some of regarded as social science researchers. Newspaper the theoretical foundations of field research and on reporters are one example; welfare department techniques of data collection. Chapter 13 discusses case workers are another. how to analyze qualitative data. Although these are “natural” activities, they are Topics Appropriate also skills to be learned and honed. This chapter for Field Research discusses these skills in some detail, examining some of the major paradigms of field research and One of the key strengths of field research is how describing some of the specific techniques that comprehensive a perspective it can give research- make scientific field research more useful than the ers. By going directly to the social phenomenon casual observation we all engage in. under study and observing it as completely as pos- sible, researchers can develop a deeper and fuller I use the term qualitative field research to dis- understanding of it. As such, this mode of observa- tinguish this type of observational method from tion is especially, though not exclusively, appropri- methods designed to produce data appropriate for ate to research topics and social studies that appear quantitative (statistical) analysis. Thus, surveys to defy simple quantification. Field researchers may provide data from which to calculate the percent- age unemployed in a population, mean incomes, and so forth. Field research more typically yields qualitative data: observations not easily reduced to
Topics Appropriate for Field Research ■ 297 recognize several nuances of attitude and behavior 8. Settlements and habitats: Small-scale “societies” that might escape researchers using other methods. such as villages, ghettos, and neighborhoods, as opposed to large societies such as nations, Field research is especially appropriate for which are difficult to study the study of those attitudes and behaviors best understood within their natural setting, as opposed 9. Social worlds: Ambiguous social entities with to the somewhat artificial settings of experiments vague boundaries and populations, such as and surveys. For example, field research provides “the sports world” and “Wall Street” a superior method for studying the dynamics of religious conversion at a revival meeting, just as a 10. Subcultures and lifestyles: How large numbers of statistical analysis of membership rolls would be a people adjust to life in groups such as a “rul- better way of discovering whether men or women ing class” or an “urban underclass” were more likely to convert. In all these social settings, field research can reveal Finally, field research is well suited to the things that would not otherwise be apparent. Here’s study of social processes over time. Thus, the field a concrete example. researcher might be in a position to examine the rumblings and final explosion of a riot as events One issue I’m particularly interested in (Bab- actually occur rather than afterward in a recon- bie 1985) is the nature of responsibility for public struction of the events. matters: Who’s responsible for making the things that we share work? Who’s responsible for keeping Other good places to apply field research meth- public spaces—parks, malls, buildings, and so on— ods include campus demonstrations, courtroom clean? Who’s responsible for seeing that broken proceedings, labor negotiations, public hearings, or street signs get fixed? Or, if a strong wind knocks similar events taking place within a relatively lim- over garbage cans and rolls them around the street, ited area and time. Several such observations must who’s responsible for getting them out of the street? be combined in a more comprehensive examina- tion over time and space. On the surface, the answer to these questions is pretty clear. We have formal and informal agree- In Analyzing Social Settings (2006:123–132), ments in our society that assign responsibility for John Lofland and his colleagues discuss several ele- these activities. Government custodians are respon- ments of social life appropriate to field research: sible for keeping public places clean. Transporta- tion department employees are responsible for the 1. Practices: Various kinds of behavior, such as talk- street signs, and perhaps the police are responsible ing or reading a book for the garbage cans rolling around on a windy day. And when these responsibilities are not fulfilled, 2. Episodes: A variety of events such as divorce, we tend to look for someone to blame. crime, and illness What fascinates me is the extent to which the 3. Encounters: Two or more people meeting and assignment of responsibility for public things to interacting specific individuals not only relieves others of the responsibility but actually prohibits them from 4. Roles and social types: The analysis of the posi- taking responsibility. It’s my notion that it has be- tions people occupy and the behavior associ- come unacceptable for someone like you or me to ated with those positions: occupations, family take personal responsibility for public matters that roles, ethnic groups haven’t been assigned to us. 5. Social and personal relationships: Behavior ap- Let me illustrate what I mean. If you were propriate to pairs or sets of roles: mother–son walking through a public park and you threw relationships, friendships, and the like down a bunch of trash, you’d discover that your action was unacceptable to those around you. 6. Groups and cliques: Small groups, such as friend- People would glare at you, grumble to each other; ship cliques, athletic teams, and work groups perhaps someone would say something to you 7. Organizations: Formal organizations, such as hospitals or schools
298 ■ Chapter 10: Qualitative Field Research about it. Whatever the form, you’d be subjected to by a neighbor. Another student decided to clean definite, negative sanctions for littering. Now here’s out a clogged storm drain on his street and found the irony. If you were walking through that same himself being yelled at by a neighbor who insisted park, came across a bunch of trash that someone that the mess should be left for the street clean- else had dropped, and cleaned it up, it’s likely that ers. Everyone who picked up litter was sneered at, your action would also be unacceptable to those laughed at, and generally put down. One young around you. You’d probably face negative sanctions man was picking up litter scattered around a for cleaning it up. trash can when a passerby sneered, “Clumsy!” It became clear to us that there are only three accept- When I first began discussing this pattern with able explanations for picking up litter in a public students, most felt the notion was absurd. Although place: we would be negatively sanctioned for littering, cleaning up a public place would obviously bring 1. You did it and got caught—somebody forced positive sanctions: People would be pleased with you to clean up your mess. us for doing it. Certainly, all my students said they would be pleased if someone cleaned up a public 2. You did it and felt guilty. place. It seemed likely that everyone else would be pleased, too, if we asked them how they would 3. You’re stealing litter. react to someone’s cleaning up litter in a public place or otherwise taking personal responsibility for In the normal course of things, it’s simply not ac- fixing some social problem. ceptable for people to take responsibility for public things. To settle the issue, I suggested that my students start fixing the public problems they came across Clearly, we could not have discovered the in the course of their everyday activities. As they nature and strength of agreements about taking did so, I asked them to note the answers to two personal responsibility for public things except questions: through field research. Social norms suggest that taking responsibility is a good thing, sometimes 1. How did they feel while they were fixing a referred to as good citizenship. Asking people what public problem they had not been assigned they thought about taking responsibility would responsibility for? have produced a solid consensus that it was good. Only going out into life, doing it, and watching 2. How did others around them react? what happened gave us an accurate picture. My students picked up litter, fixed street signs, As an interesting footnote to this story, my put knocked-over traffic cones back in place, students and I found that whenever people could cleaned and decorated communal lounges in their get past their initial reactions and discover that dorms, trimmed trees that blocked visibility at the students were simply taking responsibility for intersections, repaired public playground equip- fixing things for the sake of having them work, the ment, cleaned public restrooms, and took care of a passersby tended to assist. Although there are some hundred other public problems that weren’t “their very strong agreements making it “unsafe” to responsibility.” take responsibility for public things, the willing- ness of one person to rise above those agreements Most reported feeling very uncomfortable seemed to make it safe for others to do so, and doing whatever they did. They felt foolish, goody- they did. goody, conspicuous, and all the other feelings that keep us from performing these activities normally. Field research is not to be confused with jour- In almost every case, their personal feelings of nalism. Social scientists and journalists may use discomfort were increased by the reactions of those similar techniques, but they have quite a different around them. One student was removing a dam- relationship to data. For instance, individual inter- aged and long-unused newspaper box from the viewing is a common technique in journalism and bus stop, where it had been a problem for months, sociology; nevertheless, sociologists are not simply when the police arrived, having been summoned concerned with reporting about a subject’s attitude, belief, or experience. A sociologist’s goal is to treat
Special Considerations in Qualitative Field Research ■ 299 an interview as data that need to be analyzed to fixed public things). In this chapter, I’ve used the understand social life more generally. term field research rather than the frequently used term participant observation, because field research- Two important aspects of qualitative research ers need not always participate in what they’re need to be stressed. First, a wide range of studies studying, though they usually do study it directly at fall under the umbrella “qualitative field research.” the scene of the action. As Catherine Marshall and As we’ll see in this chapter, various epistemologies Gretchen Rossman point out: within different paradigms have quite different approaches to basic questions such as “What are The researcher may plan a role that entails data?” “How should we collect data?” and “How varying degrees of “participantness”—that is, should we analyze data?” Second, we should the degree of actual participation in daily life. remember that the questions we want to answer At one extreme is the full participant, who in our research determine the types of methods we goes about ordinary life in a role or set of roles need to use. A question such as “How do women constructed in the setting. At the other extreme construct their everyday lives in order to perform is the complete observer, who engages not at their roles as mothers, partners, and breadwin- all in social interaction and may even shun in- ners?” could be addressed by in-depth interviews volvement in the world being studied. And, of and direct observations. The assessment of advertis- course, all possible complementary mixes along ing campaigns might profit from focus group dis- the continuum are available to the researcher. cussions. In most cases, we’ll find that researchers have alternate methods to choose from. (1995: 60) In summary, then, field research offers the ad- The complete participant, in this sense, may be vantage of probing social life in its natural habitat. a genuine participant in what he or she is studying Although some things can be studied adequately (for example, a participant in a campus demonstra- through questionnaires or in the laboratory, others tion) or may pretend to be a genuine participant. cannot. And direct observation in the field lets re- In any event, whenever you act as the complete searchers observe subtle communications and other participant, you must let people see you only as events that might not be anticipated or measured a participant, not as a researcher. For instance, if otherwise. you’re using this technique to study a group made up of uneducated and inarticulate people, it would Special Considerations not be appropriate for you to talk and act like a in Qualitative Field Research university professor or student. There are specific things to take into account in ev- This type of research introduces an ethical ery research method, and qualitative field research issue, one on which social researchers themselves is no exception. When you use field research meth- are divided. Is it ethical to deceive the people you’re ods, you’re confronted with decisions about the studying in the hope that they will confide in you role you’ll play as an observer and your relations as they will not confide in an identified researcher? with the people you’re observing. Let’s examine Do the potential benefits to be gained from the some of the issues involved in these decisions. research offset such considerations? Although many professional associations have addressed this The Various Roles issue, the norms to be followed remain somewhat of the Observer ambiguous when applied to specific situations. In field research, observers can play any of several Related to this ethical consideration is a roles, including participating in what they want to scientific one. No researcher deceives his or her observe (this was the situation of the students who subjects solely for the purpose of deception. Rather, it’s done in the belief that the data will be more valid and reliable, that the subjects will be more natural and honest if they do not know the researcher is doing a research project. If the people
300 ■ Chapter 10: Qualitative Field Research being studied know they’re being studied, they your teammates know what you’re doing. There might modify their behavior in a variety of ways. are dangers in this role also, however. The people This is known as the problem of reactivity. being studied may shift much of their attention to the research project rather than focusing on the First, they might expel the researcher. Second, natural social process, making the process being they might modify their speech and behavior to observed no longer typical. Or, conversely, you appear more “respectable” than would otherwise yourself may come to identify too much with the be the case. Third, the social process itself might be interests and viewpoints of the participants. You radically changed. Students making plans to burn may begin to “go native” and lose much of your down the university administration building, for scientific detachment. example, might give up the plan altogether once they learn that one of their group is a social scien- At the other extreme, the complete observer tist conducting a research project. studies a social process without becoming a part of it in any way. Quite possibly, because of the On the other side of the coin, if you’re a researcher’s unobtrusiveness, the subjects of study complete participant, you may affect what you’re might not realize they’re being studied. Sitting at a studying. Suppose, for example, that you’re asked bus stop to observe jaywalking at a nearby inter- for your ideas about what the group should do next. section is one example. Although the complete No matter what you say, you will affect the process observer is less likely to affect what’s being studied in some fashion. If the group follows your sugges- and less likely to “go native” than the complete tion, your influence on the process is obvious. If the participant, she or he is also less likely to develop group decides not to follow your suggestion, the a full appreciation of what’s being studied. Observa- process whereby the suggestion is rejected may af- tions may be more sketchy and transitory. fect what happens next. Finally, if you indicate that you just don’t know what should be done next, you Fred Davis (1973) characterizes the extreme may be adding to a general feeling of uncertainty roles that observers might play as “the Martian” and indecisiveness in the group. and “the Convert.” The latter involves delving more and more deeply into the phenomenon under Ultimately, anything the participant-observer study, running the risk of “going native.” We’ll does or does not do will have some effect on what’s examine this risk further in the next section. being observed; it’s simply inevitable. More seri- ously, there is no complete protection against this To appreciate the “Martian” approach, imagine effect, though sensitivity to the issue may provide that you were sent to observe some newfound life a partial protection. (This influence, called the on Mars. Probably you would feel yourself ines- Hawthorne effect, was discussed more fully in capably separate from the Martians. Some social Chapter 8.) scientists adopt this degree of separation when observing cultures or social classes different from Because of these several considerations, ethi- their own. cal and scientific, the field researcher frequently chooses a different role from that of complete par- Marshall and Rossman (1995: 60–61) also ticipant. You could participate fully with the group note that the researcher can vary the amount of under study but make it clear that you were also time spent in the setting being observed: You can undertaking research. As a member of the volley- be a full-time presence on the scene or just show ball team, for example, you might use your position up now and then. Moreover, you can focus your to launch a study in the sociology of sports, letting attention on a limited aspect of the social setting or seek to observe all of it—framing an appropriate reactivity The problem that the subjects of social role to match your aims. research may react to the fact of being studied, thus altering their behavior from what it would have When Jeffrey Kidder set out to study the cul- been normally. ture of bike messengers in New York City, he found it appropriate to identify his research role to some of those he observed but not others (2005: 349):
Special Considerations in Qualitative Field Research ■ 301 While I did have an academic motivation in whether or not you actually take on the beliefs, working as a messenger, it should be made attitudes, and other points of view shared by the clear that my participation within the messen- “real” members. If the cult members believe that ger world was neither forced nor faked. To the Jesus will come next Thursday night to destroy the contrary, my lifelong interest in bicycles and world and save the members of the cult, do you alternative transportation melded seamlessly believe it or do you simply pretend to believe it? with the messenger lifestyle. Traditionally, social scientists have tended to During the course of my fieldwork, most of emphasize the importance of “objectivity” in such the messengers with whom I came in contact matters. In this example, that injunction would were unaware of my research; this was a mat- be to avoid getting swept up in the beliefs of the ter of necessity. In New York City, a messenger group. Without denying the advantages associated crosses paths with hundreds of messengers with such objectivity, social scientists today also a day. The numerous individuals that helped recognize the benefits gained by immersing them- form my understandings of messenger style selves in the points of view they’re studying, what could not all be approached to sign consent John Lofland and his colleagues (2006: 70) refer to forms. Messengers with whom I had reoccur- as “selective competence” or “insider knowledge, ring contact were informed of my sociological skill, or understanding.” Ultimately, you won’t be interest. able to fully understand the thoughts and actions of the cult members unless you can adopt their points Different situations ultimately require different of view as true—at least temporarily. To fully ap- roles for the researcher. Unfortunately, there are no preciate the phenomenon you’ve set out to study, clear guidelines for making this choice—you must you need to believe that Jesus is coming Thursday rely on your understanding of the situation and night. In some settings, this can also help you gain your own good judgment. In making your decision, rapport with your subjects (see the discussion on however, you must be guided by both methodolog- rapport later in this chapter). ical and ethical considerations. Because these often conflict, your decision will frequently be difficult, Adopting an alien point of view is an uncom- and you may find sometimes that your role limits fortable prospect for most people. It can be hard your study. enough merely to learn about views that seem strange to you; you may sometimes find it hard Relations to Subjects just to tolerate certain views. But to take them on as your own can be ten times worse. Robert Bellah Having introduced the different roles field research- (1970, 1974) has offered the term symbolic realism ers might play in connection with their observa- to indicate the need for social researchers to treat tions, we now focus more specifically on how the beliefs they study as worthy of respect rather researchers may relate to the subjects of their study than as objects of ridicule. The difficulty of adopt- and to the subjects’ points of view. ing other’s views led William Shaffir and Robert Stebbins (1991: 1) to conclude that “fieldwork We’ve already noted the possibility of pretend- must certainly rank with the more disagreeable ing to occupy social statuses we don’t really occupy. activities that humanity has fashioned for itself.” Consider now how you would think and feel in such a situation. There is, of course, a danger in adopting the points of view of the people you’re studying. When Suppose you’ve decided to study a religious cult you abandon your objectivity in favor of adopting that has enrolled many people in your neighbor- such views, you lose the possibility of seeing and hood. You might study the group by joining it or understanding the phenomenon within frames of pretending to join it. Take a moment to ask yourself reference unavailable to your subjects. On the one what the difference is between “really” joining hand, accepting the belief that the world will end and “pretending” to join. The main difference is Thursday night allows you to appreciate aspects
302 ■ Chapter 10: Qualitative Field Research of that belief available only to believers; stepping characteristics can affect what you see and how outside that view, however, makes it possible for you interpret it. The issue is broader than that, you to consider some reasons why people might however, and applies to the subjects as well as to adopt such a view. You may discover that some the researcher. Imagine yourself interviewing a did so as a consequence of personal trauma (such homeless person (1) on the street, (2) in a home- as unemployment or divorce), whereas others less shelter, or (3) in a social welfare office. The were brought into the fold through their participa- research setting could affect the person’s responses. tion in particular social networks (for example, In other words, you might get different results their whole bowling team joined the cult). Notice depending on where you conducted the interview. that the cult members might disagree with those Moreover, you might act differently as a researcher “objective” explanations, and you might not come in those different settings. If you reflect on this up with them to the extent that you had operated issue, you’ll be able to identify other aspects of legitimately within the group’s views. the research encounter that complicate the task of “simply observing what’s so.” Anthropologists sometimes use the term emic perspective in reference to taking on the point of The problem we’ve just been discussing could view of those being studied. In contrast, the etic be seen as psychological, occurring mostly inside perspective maintains a distance from the native the researchers’ or subjects’ heads. There is a corre- point of view in the interest of achieving more sponding problem at a social level, however. When objectivity. you become deeply involved in the lives of the people you’re studying, you’re likely to be moved The apparent dilemma here is that both of by their personal problems and crises. Imagine, for these postures offer important advantages but also example, that one of the cult members becomes ill seem mutually exclusive. In fact, it’s possible to and needs a ride to the hospital. Should you pro- assume both postures. Sometimes you can simply vide transportation? Sure. Suppose someone wants shift viewpoints at will. When appropriate, you can to borrow money to buy a stereo. Should you loan fully assume the beliefs of the cult; later, you can it? Probably not. Suppose they need the money step outside those beliefs (more accurately, you can for food? step inside the viewpoints associated with social science). As you become more adept at this kind There are no black-and-white rules for re- of research, you may come to hold contradictory solving situations such as these, but you should viewpoints simultaneously, rather than switching realize that you’ll need to deal with them regard- back and forth. less of whether or not you reveal that you’re a researcher. Such problems do not tend to arise in During my study of trance channeling—in other types of research—surveys and experiments, which people allow spirits to occupy their bodies for example—but they are part and parcel of field and speak through them—I found I could partici- research. pate fully in channeling sessions without becoming alienated from conventional social science. Rather Caroline Knowles (2006) raises a somewhat than “believing” in the reality of channeling, I different issue with regard to the researcher’s re- found it possible to suspend beliefs in that realm: lationship to subjects in the field. In her interview neither believing it to be genuine (like most of the study of British expatriates living in Hong Kong, other participants) nor disbelieving it (like most she noticed that some were particularly difficult for scientists). Put differently, I was open to either her to deal with. When she found herself writing possibility. Notice how this differs from our normal research notes explaining why the project would need to “know” whether such things are legitimate not profit from her interviewing them further, she or not. forced herself to look more deeply into the inter- actional dynamics in question—with an emphasis Social researchers often refer to the concerns on her side of the relationships. She examined why just discussed as a matter of reflexivity, in the sense certain informants made her uncomfortable and of things acting on themselves. Thus, your own
Some Qualitative Field Research Paradigms ■ 303 then pressed through the discomfort to continue The more social researchers have gone into interviewing. She found that factors such as the the field to study their fellow humans face-to-face, attitudes they expressed, their rude interaction however, the more they have become conscious of styles, and the nature of the relationship she was these implicit assumptions about researcher superi- establishing with them contributed to her reaction. ority, and the more they have considered alterna- In the end, she gained a much deeper understand- tives. As we turn now to the various paradigms of ing of her subjects than would have been possible if field research, we’ll see some of the ways in which she had limited herself to those who were coopera- that ongoing concern has worked itself out. See tive and nice. “Keeping Humanity in Focus” for an example of field research on status. This discussion of the field researcher’s relation- ships to subjects flies in the face of the usual view Some Qualitative Field of “scientific objectivity.” Before concluding this Research Paradigms section, let’s take the issue one step further. Although I’ve described field research as simply go- In the conventional view of science, differ- ing where the action is and observing it, there are ences of power and status separate the researcher actually many different approaches to this research from the subjects of research. When we discussed method. This section examines several field re- experimental designs in Chapter 8, for example, it search paradigms: naturalism, ethnomethodology, was obvious who was in charge: the experimenter, grounded theory, case studies and the extended who organized things and told the subjects what case method, institutional ethnography, and partic- to do. Often the experimenter was the only person ipatory action research. Although this survey won’t who even knew what the research was really exhaust the variations on the method, it should about. Something similar might be said about give you a broad appreciation of the possibilities. survey research. The person running the survey designs the questions, decides who will be selected It’s important to recognize that there are no for questioning, and analyzes the data collected. specific methods attached to these paradigms. You could do ethnomethodology or institutional eth- Sociologists often look at these sorts of re- nography by analyzing court hearings or conduct- lationships as power or status relationships. In ing group interviews, for example. The important experimental and survey designs, the researcher distinctions of this section are epistemological, hav- clearly has more power and a higher status than ing to do with what data mean, regardless of how the people being studied do. The researchers have they were collected. a special knowledge that the subjects don’t enjoy. They’re not so crude as to say they’re superior to Naturalism their subjects, but there is a sense in which that’s implicitly assumed. (Notice that there is a similar, Naturalism is an old tradition in qualitative implicit assumption about the writers and readers research. The earliest field researchers operated on of textbooks.) the positivist assumption that social reality was “out In field research, such assumptions can be prob- naturalism An approach to field research based on lematic. When the early European anthropologists the assumption that an objective social reality exists set out to study what were originally called “primi- and can be observed and reported accurately. tive” societies, there was no doubt that the anthro- pologists knew best. Whereas the natives “believed” in witchcraft, for example, the anthropologists “knew” it wasn’t really true. Whereas the natives said some of their rituals would appease the gods, the anthropologists explained that the “real” func- tions of these rituals were the creation of social iden- tity, the establishment of group solidarity, and so on.
304 ■ Chapter 10: Qualitative Field Research Class Acts: Service and Inequality in Luxury Hotels by Rachel Sherman. © 2007 Rachel Sherman. Published by the University of California Press. Photograph © Karen Beard/Getty Images Keeping Humanity In Focus What could seem like a clearer status relationship than between a guest in a luxury hotel and the room service and other staff who serve that guest’s needs? In fact,Rachel Sherman has found a far more complex process than you might imagine.She is particularly interested in how service workers balance their relationships with management and their relationships with guests.Unlike manufacturing workers,the hotel service staff must deal with both supervisors and consumers,even when the demands of the two conflict.In part,she discovered that service workers in hotels often receive more discretion regarding how to serve guests’needs than we might expect.This has a positive impact on the worker’s sense of self as well as providing a good experience for guests. Sherman’s observations and conclusions came from months spent as a service worker in two luxury hotels.She made her research identity known to management and was able to move around through many of the different service jobs:making reservations,delivering room-service meals,parking cars,carrying bags,housekeeping,and many other tasks that the guests in luxury hotels expect.Her immersion in the research allowed her access to data she would not have found out otherwise. there,” ready to be naturally observed and reported believed that in order to learn fully about social by the researcher as it “really is” (Gubrium and life on the streets, he needed to become more of Holstein 1997). This tradition started in the 1930s an insider. He made contact with “Doc,” his key and 1940s at the University of Chicago’s sociology informant, who appeared to be one of the street- department, whose faculty and students fanned gang leaders. Doc let Whyte enter his world, and out across the city to observe and understand local Whyte got to participate in the activities of the neighborhoods and communities. The researchers people of Cornerville. His study offered something of that era and their research approach are now that surveys could not: a richly detailed picture of often referred to as the Chicago School. life among the Italian immigrants of Cornerville. One of the earliest and best-known studies that An important feature of Whyte’s study is that illustrates this research tradition is William Foote he reported the reality of the people of Cornerville Whyte’s ethnography of Cornerville, an Italian on their terms. The naturalist approach is based on American neighborhood, in his book Street Corner telling “their” stories the way they “really are,” not Society (1943). An ethnography is a study that the way the ethnographer understands “them.” focuses on detailed and accurate description rather The narratives collected by Whyte are taken at than explanation. Like other naturalists, Whyte face value as the social “truth” of the Cornerville residents. ethnography A report on social life that focuses on detailed and accurate description rather than Forty-five years later, David Snow and Leon explanation. Anderson (1987) conducted exploratory field re- search into the lives of homeless people in Austin, Texas. Their main task was to understand how the
Some Qualitative Field Research Paradigms ■ 305 homeless construct and negotiate their identity future of U.S. society, but most were more inter- while knowing that the society they live in attaches ested in creating alternative lives and cultures for a stigma to homelessness. Snow and Anderson themselves than in blowing anyone up. That’s believed that, to achieve this goal, the collection not to suggest none of the survivalists poses a of data had to arise naturally. Like Whyte in Street threat, but Mitchell’s examination moves beyond Corner Society, they found some key informants the McVeighs, Koreshes, and Weavers to draw a whom they followed in their everyday journeys, broader picture of the whole phenomenon. such as at their day-labor pickup sites or under bridges. Snow and Anderson chose to memorize While ethnographers seek to discover and the conversations they participated in or the “talks” understand the patterns of living among those that homeless people had with each other. At the they are studying, Mitchell Duneier (1999) has end of the day, the two researchers debriefed and warned against what he calls the “ethnographic wrote detailed field notes about all the “talks” they fallacy.” This refers to an overgeneralization and encountered. They also taped in-depth interviews oversimplification of the patterns observed. Despite with their key informants. the existence of patterns within groups, there is also diversity, and you need to be wary of broad Snow and Anderson reported “hanging out” assertions suggesting that “the poor,” “the French,” with homeless people over the course of 12 months or “cheerleaders” act or think in certain ways as for a total of 405 hours in 24 different settings. Out though all members of the group do so. of these rich data, they identified three related pat- terns in homeless people’s conversations. First, the Whereas this chapter aims at introducing you homeless showed an attempt to “distance” them- to some of the different approaches available to you selves from other homeless people, from the low- in qualitative field research, please realize that this status job they currently had, or from the Salvation discussion of ethnography merely sketches some of Army they depended on. Second, they “embraced” the many avenues social researchers have estab- their street-life identity—their group membership lished. If you’re interested in this general approach, or a certain belief about why they are homeless. you might want to explore the idea of virtual eth- Third, they told “fictive stories” that always con- nography, which uses ethnographic techniques for trasted with their everyday life. For example, they inquiry into cyberspace. Or, in a different direction, would often say that they were making much more autoethnography intentionally assumes a personal money than they really were, or even that they stance, breaking with the general proscription were “going to be rich.” against the researcher getting involved at that level. Lest autoethnography seem a simple and/or Richard Mitchell (2002) offers another, timely trivial undertaking, you might look at Sarah Wall’s illustration of the power of ethnographic reporting. 2008 article, “Easier Said than Done: Writing an Recent U.S. history has raised the specter of vio- Autoethnography.” lence from secretive survivalist groups, dramatized by the 1992 siege at Ruby Ridge, Idaho, which left You can learn more about these variants on the wife and son of the white supremacist Randy ethnography by searching the web or your campus Weaver dead; the 1993 shootout with David Ko- library. A later section of this chapter will examine resh and his Branch Davidians in Waco, Texas; and institutional ethnography, which links individuals and Timothy McVeigh’s 1995 bombing, which left 168 organizations. dead under the rubble of the nine-story Murrah Federal Building in Oklahoma City. In Chapter 9, we saw how the Internet is af- fecting survey research. Eric Anderson used the Mitchell describes a variety of survivalist Internet to launch a qualitative, in-depth inter- individuals and groups, seeking to understand viewing study of male cheerleaders: their reasoning, their plans, and the threat they may pose for the rest of us. He found the surviv- Twelve collegiate male cheerleaders were con- alists disillusioned with and uncertain about the tacted for interviews by using the member pro- file search on America Online which provides a
306 ■ Chapter 10: Qualitative Field Research search engine for accessing the stated interests day world. As we saw in Chapter 2, the sociologist of AOL’s 33 million subscribers. After com- Harold Garfinkel suggested that researchers break municating with these cheerleaders through the rules so that people’s taken-for-granted expecta- instant messaging, I asked them for in-depth, tions would become apparent. This is the technique taped telephone interviews. that Garfinkel called ethnomethodology. (2005: 340) Garfinkel became known for engaging his students to perform a series of what he called Anderson then used snowball sampling to increase “breaching experiments” designed to break away the number of cheerleaders to study. This is just from the ordinary (Heritage 1984). For instance, another example of the wide variety of venues for Garfinkel (1967) asked his students to do a “con- ethnographic study. versation clarification experiment.” Students were told to engage in an ordinary conversation with an Ethnomethodology acquaintance or a friend and to ask for clarification about any of this person’s statements. Through this Ethnomethodology, which I introduced as a re- technique, they uncovered elements of conversa- search paradigm in Chapter 2, is a unique approach tion that are normally taken for granted. Here are to qualitative field research. It has its roots in the two examples of what Garfinkel’s students reported philosophical tradition of phenomenology, which (1967: 42): can explain why ethnomethodologists are skepti- cal about the way people report their experience of Case 1 reality (Gubrium and Holstein 1997). Alfred Schutz The subject was telling the experimenter, (1967, 1970), who introduced phenomenology, argued that reality was socially constructed rather a member of the subject’s car pool, about than being “out there” for us to observe. People having had a flat tire while going to work the describe their world not “as it is” but “as they make previous day. sense of it.” Thus, phenomenologists would argue that Whyte’s street-corner men were describing I had a flat tire. their gang life as it made sense to them. Their (E) What do you mean, you had a flat tire? reports, however, would not tell us how and why She appeared momentarily stunned. Then it made sense to them. For this reason, researchers she answered in a hostile way: “What do you cannot rely on their subjects’ stories to depict social mean, ‘What do you mean?’ A flat tire is a realities accurately. flat tire. That is what I meant. Nothing special. What a crazy question.” Whereas traditional ethnographers believe in immersing themselves in a particular culture Case 6 and reporting their informants’ stories as if they The victim waved his hand cheerily. represented reality, phenomenologists see a need (S) How are you? to “make sense” out of the informants’ perceptions (E) How I am in regard of what? My of the world. Following in this tradition, some field researchers have felt the need to devise techniques health, my finances, my school work, my peace that reveal how people make sense of their every- of mind, my . . . ? ethnomethodology An approach to the study of (S) (Red in the face and suddenly out of social life that focuses on the discovery of implicit, control.) Look I was just trying to be polite. usually unspoken assumptions and agreements; this Frankly, I don’t give a damn how you are. method often involves the intentional breaking of agreements as a way of revealing their existence. By setting aside or “bracketing” their expec- tations from these everyday conversations, the experimenters made visible the subtleties of mun- dane interactions. For example, although “How are you?” has many possible meanings, none of us has any trouble knowing what it means in casual
Some Qualitative Field Research Paradigms ■ 307 interactions, as the unsuspecting subject revealed are more intent on identifying the methods in his final comment. through which understanding occurs. In the case of the convict code, Wieder came to see that convicts Ethnomethodologists, then, are not simply used the code to make sense of their own interac- interested in subjects’ perceptions of the world. In tions with other convicts and with the staff. The these cases, we could imagine that the subjects may ethnography of the halfway house thus shifted have thought that the experimenters were rude, to an ethnography of the code. For instance, the stupid, or arrogant. The conversation itself, not the convicts would say, “You know I won’t snitch,” informants, is the object of ethnomethodological referring to the code as a way to justify their refusal studies. In general, ethnomethodology focuses on to answer Wieder’s question (p. 168). According to the “underlying patterns” of interactions that regu- Wieder, the code “operated as a device for stopping late our everyday lives. or changing the topic of conversation” (p. 175). Even the staff would refer to the code to justify Ethnomethodologists believe that research- their reluctance to help the convicts. Although ers who use a naturalistic analysis “[lose] the the code was something that constrained behav- ability to analyze the commonsense world and its ior, it also functioned as a tool for the control of culture if [they use] analytical tools and insights interactions. that are themselves part of the world or culture being studied” (Gubrium and Holstein 1997: 43). Grounded Theory D. L. Wieder provides an excellent example of how different a naturalistic approach is from an Grounded theory originated from the collabora- ethnomethodological approach (Gubrium and tion of Barney Glaser and Anselm Strauss, sociolo- Holstein 1997). In his study Language and Social gists who brought together two main traditions of Reality: The Case of Telling the Convict Code (1988), research: positivism and interactionism. Essentially, Wieder started to approach convicts in a halfway grounded theory is the attempt to derive theories house in a traditional ethnographic style: He was from an analysis of the patterns, themes, and com- going to become an insider by befriending the in- mon categories discovered in observational data. mates and by conducting participant observations. The first major presentation of this method can be He took careful notes and recorded interactions found in Glaser and Strauss’s book, The Discovery of among inmates and between inmates and staff. His Grounded Theory (1967). Grounded theory can be first concern was to describe the life of the convicts described as an approach that attempts to combine of the halfway house the way it “really was” for a naturalist approach with a positivist concern for a them. Wieder’s observations allowed him to report “systematic set of procedures” in doing qualitative on a “convict code” that he thought was the source research. of the deviant behavior expressed by the inmates toward the staff. This code, which consisted of Anselm Strauss and Juliet Corbin (1998: a series of rules such as “Don’t kiss ass,” “Don’t 43– 46) have suggested that grounded theory al- snitch,” and “Don’t trust the staff,” was followed by lows the researcher to be scientific and creative at the inmates who interfered with the staff members’ the same time, as long as the researcher follows attempts to help them make the transition between these guidelines: prison and the community. grounded theory An inductive approach to the It became obvious to Wieder that the code was study of social life that attempts to generate a theory more than an explanation for the convicts’ devi- from the constant comparing of unfolding observa- ant behavior; it was a “method of moral persua- tions. This is very different from hypothesis testing, sion and justification” (Wieder 1988: 175). At this in which theory is used to generate hypotheses to be point he changed his naturalistic approach to an tested through observations. ethnomethodological one. Whereas naturalistic field researchers aim to understand social life as the participants understand it, ethnomethodologists
308 ■ Chapter 10: Qualitative Field Research Grounded theory emphasizes research proce- dures. In particular, systematic coding is important for achieving validity and reliability in the data analysis. Because of this somewhat positivistic view of data, grounded theorists are quite open to the use of qualitative studies in conjunction with quantitative ones. Here are two examples of the implementation of this approach. © Lynda Koolish Studying Academic Change Anselm L. Strauss, a pioneer qualitative researcher, was a Clifton Conrad’s (1978) study of academic change principal founder of the Grounded Theory Method. in universities is an early example of the grounded theory approach. Conrad hoped to uncover the • Think comparatively: The authors suggest that it major sources of changes in academic curricula and at the same time understand the process of change. is essential to compare numerous incidents as Using the grounded theory idea of theoretical a way of avoiding the biases that can arise from sampling—whereby groups or institutions are se- interpretations of initial observations. lected on the basis of their theoretical relevance— Conrad chose four universities for the purpose of • Obtain multiple viewpoints: In part this refers to his study. In two, the main vehicle of change was the formal curriculum committee; in the other two, the different points of view of participants in the vehicle of change was an ad hoc group. the events under study, but Strauss and Corbin suggest that different observational techniques Conrad explained, step by step, the advantage may also provide a variety of viewpoints. of using the grounded theory approach in building his theory of academic change. He described the • Periodically step back: As data accumulate, you’ll process of systematically coding data in order to create categories that must “emerge” from the data begin to frame interpretations about what is and then assessing the fitness of these categories going on, and it’s important to keep check- with each other. Going continuously from data to ing your data against those interpretations. As theory and theory to data allowed him to reassess Strauss and Corbin (1998: 45) say, “The data the validity of his initial conclusions about aca- themselves do not lie.” demic change. • Maintain an attitude of skepticism: As you begin to For instance, it first seemed that academic change was mainly caused by an administrator interpret the data, you should regard all those who was pushing for it. By reexamining the data interpretations as provisional, using new ob- and looking for more-plausible explanations, Con- servations to test those interpretations, not just rad found the pressure of interest groups a more confirm them. convincing source of change. The emergence of these interest groups actually allowed the adminis- • Follow the research procedures: Grounded theory trator to become an agent of change. allows for flexibility in data collection as Assessing how data from each of the two types theories evolve, but Strauss and Corbin (1998: of universities fit with the other helped refine the 46) stress that three techniques are essential: theory building. Conrad concluded that changes in “making comparisons, asking questions, and university curricula are based on the following pro- sampling.” cess: Conflict and interest groups emerge because of internal and external social structural forces;
Some Qualitative Field Research Paradigms ■ 309 they push for administrative intervention and they also began to develop a more or less standard- recommendation to make changes in the current ized set of questions to ask shoppers. Initially, all academic program; these changes are then made by the questions were open-ended, but they eventu- the most powerful decision-making body. ally developed closed-ended items as well. Shopping Romania This study illustrates the key, inductive prin- ciples of grounded theory: Data are collected in the Much has been written about large-scale changes absence of hypotheses. The initial data are used to caused by the shift from socialism to capitalism in determine the key variables as perceived by those the former USSR and its Eastern European allies. being studied, and hypotheses about relationships Patrick Jobes and his colleagues (1997) wanted to among the variables are similarly derived from the learn about the transition on a smaller scale among data collected. Continuing data collection yields average Romanians. They focused on the task of refined understanding and, in turn, sharpens the shopping. focus of data collection itself. Noting that shopping is normally thought of as Case Studies a routine, relatively rational activity, the research- and the Extended Case Method ers suggested that it could become a social problem in a radically changing economy. They used the Social researchers often speak of case studies. A Grounded Theory Method to examine Romanian case study focuses attention on a single instance shopping as a social problem, looking for the ways of some social phenomenon, such as a village, a in which ordinary people solved the problem. family, or a juvenile gang. As Charles Ragin and Howard Becker (1992) point out, there is little Their first task was to learn something about consensus on what may constitute a “case,” and how Romanians perceived and understood the the term is used broadly. The case being studied, task of shopping. The researchers—participants for example, might be a period of time rather than in a social problems class—began by interviewing a particular group of people. The limitation of at- 40 shoppers and asking whether they had experi- tention to a particular instance of something is the enced problems in connection with shopping and essential characteristic of the case study. what actions they had taken to cope with those problems. The chief purpose of case studies may be de- scriptive, as when an anthropologist describes the Once the initial interviews were completed, culture of a preliterate tribe. Or the in-depth study the researchers reviewed their data, looking for of a particular case can yield explanatory insights, categories of responses—the shoppers’ most com- as when the community researchers Robert and mon problems and solutions. One of the most com- Helen Lynd (1929, 1937) and W. Lloyd Warner mon problems was a lack of money. This led to the (1949) sought to understand the structure and pro- researchers’ first working hypothesis: The “socio- cess of social stratification in small-town USA. economic position of shoppers would be associated with how they perceived problems and sought Case study researchers may seek only an idio- solutions” (1997: 133). This and other hypotheses graphic understanding of the particular case under helped the researchers focus their attention on examination, or—as we’ve seen with grounded more-specific variables in subsequent interviewing. theory—case studies can form the basis for the development of more-general, nomothetic theories. As they continued, they also sought to inter- view other types of shoppers. When they inter- case study The in-depth examination of a single viewed students, for example, they discovered that instance of some social phenomenon, such as a vil- different types of shoppers were concerned with lage, a family, or a juvenile gang. different kinds of goods, which in turn affected the problems faced and the solutions tried. As the researchers developed additional hy- potheses in response to the continued interviewing,
310 ■ Chapter 10: Qualitative Field Research Michael Burawoy and his colleagues (1991) experiences as a student, teachers had total rights have suggested a somewhat different relation- over the minds, bodies, and souls of their pupils. ship between case studies and theory. For them, She observed something quite different at a school the extended case method has the purpose of in a lower-middle-class neighborhood in Berkeley, discovering flaws in, and then modifying, existing California—Emerald Junior High School, where social theories. This approach differs importantly she volunteered as a tutor. She had access to the from some of the others already discussed. classroom of Mr. Henry (an eighth-grade English teacher) as well as other teachers’ classrooms, the Whereas the grounded theorists seek to enter lunchroom, and English Department meetings. the field with no preconceptions about what they’ll She wrote field notes based on the negotiations find, Burawoy suggests just the opposite: to try “to between students and teachers. She explained the lay out as coherently as possible what we expect nature of the student–teacher negotiations she wit- to find in our site before entry” (Burawoy et al. nessed by focusing on the separation of functions 1991: 9). Burawoy sees the extended case method among the school, the teacher, and the family. as a way to rebuild or improve theory instead of approving or rejecting it. Thus, he looks for all the In Hurst’s observation, the school fulfilled the ways in which observations conflict with exist- function of controlling its students’ “bodies”—for ing theories and what he calls “theoretical gaps example, by regulating their general movements and silences” (1991: 10). This orientation to field and activities within the school. The students’ research implies that knowing the literature before- “minds” were to be shaped by the teacher, whereas hand is actually a must for Burawoy and his col- students’ families were held responsible for their leagues, whereas grounded theorists would worry “souls”; that is, families were expected to socialize that knowing what others have concluded might students regarding personal values, attitudes, sense bias their observations and theories. of property, and sense of decorum. When students don’t come to school with these values in hand, the To illustrate the extended case method, I’ll pre- teacher, according to Hurst, “must first negotiate sent two examples of studies by Burawoy’s students. with the students some compromise on how the students will conduct themselves and on what will Teacher–Student Negotiations be considered classroom decorum” (1991: 185). Leslie Hurst (1991) set out to study the patterns Hurst explained that the constant bargaining of interaction between teachers and students of a between teachers and students is an expression of junior high school. She went into the field armed the separation between “the body,” which is the with existing contradictory theories about the school’s concern, and “the soul” as family domain. “official” functions of the school. Some theories The teachers, who had limited sanctioning power suggested that the purpose of schools is to promote to control their students’ minds in the classroom, social mobility, whereas others suggested that were using forms of negotiations with students so schools mainly reproduce the status quo in the that they could “control . . . the student’s body and form of a stratified division of labor. The official sense of property” (1991: 185), or as Hurst defines roles assigned to teachers and students could be it, “babysit” the student’s body and soul. interpreted in terms of either view. Hurst says she differs from the traditional socio- Hurst was struck, however, by the contrast logical perspectives as follows: between these theories and the types of interac- tions she observed in the classroom. In her own I do not approach schools with a futuristic eye. I do not see the school in terms of train- extended case method A technique developed by ing, socializing, or slotting people into future Michael Burawoy in which case study observations hierarchies. To approach schools in this manner are used to discover flaws in and to improve existing is to miss the negotiated, chaotic aspects of social theories. the classroom and educational experience.
Some Qualitative Field Research Paradigms ■ 311 A futurist perspective tends to impose an order showed how the AIDS research project devel- and purpose on the school experience, missing oped the bureaucratic inertia typical of established its day-to-day reality. organizations: Its goal became that of sustaining itself. (1991: 186) Both of these studies illustrate how the ex- In summary, what emerges from Hurst’s study tended case method can operate. The researcher is an attempt to improve the traditional sociological enters the field with full knowledge of existing understanding of education by adding the idea that theories but aims to uncover contradictions that classroom, school, and family have separate func- require the modification of those theories. tions, which in turn can explain the emergence of “negotiated order” in the classroom. One criticism of the case study method is the limited generalizability of what may be observed The Fight against AIDS in a single instance of some phenomenon. This risk is reduced, however, when more than one Katherine Fox (1991) set out to study an agency case is studied in depth: the comparative case study whose goal was to fight the AIDS epidemic by method. You can find examples of this in the bringing condoms and bleach for cleaning needles discussion of comparative and historical methods in to intravenous drug users. It’s a good example Chapter 11 of this book. of finding the limitations of well-used models of theoretical explanation in the realm of understand- Institutional Ethnography ing deviance—specifically, the “treatment model” that predicted that drug users would come to the Institutional ethnography is an approach clinic and ask for treatment. Fox’s interactions with originally developed by Dorothy Smith (1978) to outreach workers—most of whom were part of the better understand women’s everyday experiences community of drug addicts or former prostitutes— by discovering the power relations that shape those contradicted that model. experiences. Today this methodology has been ex- tended to the ideologies that shape the experiences To begin, it was necessary to understand the of any oppressed subjects. drug users’ subculture and use that knowledge to devise more-realistic policies and programs. The Smith and other sociologists believe that if target users had to be convinced, for example, that researchers ask women or other members of subor- the program workers could be trusted, that they dinated groups about “how things work,” they can were really interested only in providing bleach and discover the institutional practices that shape their condoms. The target users needed to be sure they realities (M. L. Campbell 1998; D. Smith 1978). were not going to be arrested. The goal of such inquiry is to uncover forms of oppression that more-traditional types of research Fox’s field research didn’t stop with an ex- often overlook. amination of the drug users. She also studied the agency workers, discovering that the outreach Dorothy Smith’s methodology is similar to program meant different things to the research ethnomethodology in the sense that the subjects directors and the outreach workers. Some of the themselves are not the focus of the inquiry. The volunteers who were actually providing the bleach institutional ethnographer starts with the personal and condoms were frustrated about the minor changes they felt they could make. Many thought institutional ethnography A research technique the program was just a bandage on the AIDS and in which the personal experiences of individuals drug-abuse problems. Some resented having to are used to reveal power relationships and other take field notes. Directors, on the other hand, characteristics of the institutions within which they needed reports and field notes so that they could operate. validate their research in the eyes of the federal and state agencies that financed the project. Fox’s study
312 ■ Chapter 10: Qualitative Field Research Courtesy Dorothy E. Smith mothering, schooling, and children’s development. Griffith started by interviewing mothers from three Dorothy Smith, a pioneering social researcher and founder of cities of southern Ontario about their everyday institutional ethnography. work of creating a relationship between their families and the school. This was the starting point experiences of individuals but proceeds to uncover for other interviews with parents, teachers, school the institutional power relations that structure administrators, social workers, school psychologists, and govern those experiences. In this process, the and central office administrators. researcher can reveal aspects of society that would have been missed by an inquiry that began with In her findings, Griffith explained how the dis- the official purposes of institutions. course about mothering had shifted its focus over time from a mother– child interaction to “child- This approach links the “microlevel” of every- centered” recommendations. She saw a distinct day personal experiences with the “macrolevel” of similarity in the discourse used by schools, the institutions. As M. L. Campbell puts it, media (magazines and television programs), the state, and child-development professionals. Institutional ethnography, like other forms of ethnography, relies on interviewing, observa- Teachers and child-development profession- tions, and documents as data. Institutional als saw the role of mothers in terms of a necessary ethnography departs from other ethnographic collaboration between mothers and schools for the approaches by treating those data not as the child to succeed not only in school but also in life. topic or object of interest, but as “entry” into Because of unequal resources, all mothers do not the social relations of the setting. The idea is to participate in this discourse of “good” child devel- tap into people’s expertise. opment the same way. Griffith found that working- class mothers were perceived as weaker than (1998: 57) middle-class mothers in the “stimulation” effort of schooling. Griffith argues that this child-develop- Here are two examples of this approach. ment discourse, embedded in the school institution, perpetuates the reproduction of class by making Mothering, Schooling, middle-class ideals for family–school relations the and Child Development norm for everyone. Our first example of institutional ethnography is a Compulsory Heterosexuality study by Alison Griffith (1995), who collected data with Dorothy Smith on the relationship among The second illustration of institutional ethnography is taken from Didi Khayatt’s (1995) study of the institutionalization of compulsory heterosexual- ity in schools and its effects on lesbian students. In 1990, Khayatt began her research by interviewing 12 Toronto lesbians, 15 to 24 years of age. Begin- ning with the young women’s viewpoint, she expanded her inquiry to other students, teachers, guidance counselors, and administrators. Khayatt found that the school’s administrative practices generated a compulsory heterosexuality, which produced a sense of marginality and vulner- ability among lesbian students. For example, the school didn’t punish harassment and name- calling directed at gay students. The issue of
Some Qualitative Field Research Paradigms ■ 313 homosexuality was excluded from the curricu- selves as researchers, they automatically regain lum lest it appear to students as an alternative to power over knowledge. heterosexuality. Participatory action research poses a special In both of the studies I’ve described, the inquiry challenge to researchers. On the one hand, par- began with the women’s standpoint—mothers and ticipants in the social situation ideally become em- lesbian students. However, instead of emphasiz- powered to frame research relevant to their needs, ing the subjects’ viewpoints, both analyses focused as they define those needs. At the same time, the on the power relations that shaped these women’s researcher brings special skills and insights that experiences and reality. nonresearchers lack. So who should be in charge? Andrew Sense (2006: 1) suggests that this decision Participatory Action Research may have to be made in the moment: “Do I take the ‘passenger’ position on the bus or do I take Our final field research paradigm takes us further the ‘driver’ seat and be a little more provocative to along in our earlier discussion of the status and energise the session[?] My view at this moment is power relationships linking researchers to the sub- to judge it on the day.” jects of their research. Within the participatory action research (PAR) paradigm, the researcher’s Examples of this approach include research function is to serve as a resource to those being on community power structures, corporate re- studied—typically, disadvantaged groups—as an search, and “right-to-know” movements (Whyte, opportunity for them to act effectively in their own Greenwood, and Lazes 1991). Here are two interest. The disadvantaged subjects define their examples of corporate research that used a PAR problems, define the remedies desired, and take the approach. lead in designing the research that will help them realize their aims. The Xerox Corporation This approach began in Third World research A participatory action research project took place at development, but it spread quickly to Europe and the Xerox corporation at the instigation of leaders North America (Gaventa 1991). It comes from a of both management and the union. Management’s vivid critique of classical social science research. Ac- goal was to lower costs so that the company could cording to the PAR paradigm, traditional research thrive in an increasingly competitive market. The is perceived as an “elitist model” (Whyte, Green- union suggested a somewhat broader scope: im- wood, and Lazes 1991) that reduces the “subjects” proving the quality of working life while lowering of research to “objects” of research. According to manufacturing costs and increasing productivity. many advocates of the PAR perspective, the distinc- tion between the researcher and the researched Company managers began by focusing at- should disappear. They argue that the subjects who tention on shop-level problems; they were less will be affected by research should also be respon- concerned with labor contracts or problematic sible for its design. managerial policies. At the time, management had a plan to start an “outsourcing” program that Implicit in this approach is the belief that re- would lay off 180 workers, and the union had search functions not only as a means of knowledge production but also as a “tool for the education and participatory action research (PAR) An ap- development of consciousness as well as mobiliza- proach to social research in which the people being tion for action” (Gaventa 1991: 121–22). Advocates studied are given control over the purpose and pro- of participatory action research equate access to cedures of the research; intended as a counter to the information with power and argue that this power implicit view that researchers are superior to those has been kept in the hands of the dominant class, they study. sex, ethnicity, or nation. Once people see them-
314 ■ Chapter 10: Qualitative Field Research begun mobilizing to oppose the plan. Peter Lazes, affect their own lives. Bernita Quoss, Margaret a consultant hired by Xerox, spent the first month Cooney, and Terri Longhurst (2000) report a re- convincing management and the union to create a search project involving welfare policy in Wyo- “cost study team” (CST) that included workers in ming. University students, many of them welfare the wire harness department. recipients, undertook research and lobbying efforts aimed at getting Wyoming to accept postsecondary Eight full-time workers were assigned to the education as “work” under the state’s new welfare CST for six months. Their task was to study the regulations. possibilities of making changes that would save the company $3.2 million and keep the 180 jobs. This project began against the backdrop of the The team had access to all financial information 1996 Personal Responsibility and Work Opportu- and was authorized to call on anyone within the nity Act (PRWORA), which company. This strategy allowed workers to make suggestions outside the realm usually available eliminated education waivers that had been to them. According to Whyte and his colleagues, available under the previous welfare law, the “reshaping the box enabled the CST to call upon 1988 Family Support Act (FSA). These waivers management to explain and justify all staff ser- had permitted eligible participants in the cash vices” (1991: 27). Because of the changes suggested assistance AFDC program to attend college as by the CST and implemented by management, the an alternative to work training requirements. company saved the targeted $3.2 million. Empirical studies of welfare participants who received these waivers have provided evidence Management was so pleased by this result that that education, in general, is the most effective it expanded the wire harness CST project to three way to stay out of poverty and achieve self- other departments that were threatened by compe- sufficiency. tition. Once again, management was happy about the money saved by the teams of workers. (Quoss, Cooney, and Longhurst 2000: 47) The Xerox case study is an interesting example The students began by establishing an organiza- of participatory action research because it shows tion called Empower and by making presentations how the production of knowledge does not always on campus to enlist broad student and faculty have to be an elitist enterprise. The “experts” do support. They compiled existing research relevant not necessarily have to be the professionals. Ac- to the issue and established relationships with cording to Whyte and his colleagues, “At Xerox, members of the state legislature. By the time the participatory action research created and guided 1997 legislative session opened, the students were a powerful process of organizational learning—a actively engaged in the process of modifying state process whereby leaders of labor and management welfare laws to offset the shift in federal policy. learned from each other and from the consultant/ facilitator, while he learned from them” (1991: 30). The students prepared and distributed fact sheets and other research reports that would be PAR and Welfare Policy relevant to the legislators’ deliberations. They at- tended committee meetings and lobbied legislators Participatory action research often involves poor on a one-to-one basis. When erroneous or mislead- people, as they are typically less able than other ing data were introduced into the discussions, the groups to influence the policies and actions that student-researchers were on hand to point out the errors and offer corrections. emancipatory research Research conducted for the purpose of benefiting disadvantaged Ultimately, they succeeded. Welfare recipients groups. in Wyoming were allowed to pursue postsecondary education as an effective route out of poverty. Some researchers speak of emancipatory research, which Ardha Danieli and Carol Wood- hams (2005: 284) define as “first and foremost a
Some Qualitative Field Research Paradigms ■ 315 process of producing knowledge which will be of Recall the discussion of informed consent benefit to oppressed people; a political outcome.” in Chapter 3, a method of protecting research Both qualitative and quantitative methods can be subjects. In this case, as Lofland notes elsewhere, used to pursue this goal, but it goes well beyond explicit consent was not necessarily needed here, simply learning what’s so, even as seen from the because the behavior being studied was public. Still, subjects’ point of view. The authors focus on the his instincts as a social researcher were to ensure study of disability, and they note similarities in the that he treat subjects appropriately. development of emancipatory research and early feminist research. One of Lofland’s purposes was to study this failed attempt to secure “historic preservation” Blocking a Demolition status for a building, thus providing useful informa- tion to activists in the future. This indicates that In another example of researchers being directly there can be many different forms of participatory involved in what they study, John Lofland (2003) action research. detailed the demolition of a historic building in Davis, California, and community attempts to block At the same time, this is a valuable study for a the demolition. One thing that makes the book study of research methods, because Lofland, as the especially unusual is its reliance on photographs author of research methods textbooks, is particu- and facsimile news articles and government docu- larly sensitive to the methodological aspects of the ments as raw data for the analysis (and for the study. reader): what Lofland refers to as “documentary sociology.” The deptth and intensity of my involvement is a two-edged sword. On the one edge, my As Lofland explains, he was involved in the involvement provided me with a view closer issue first as an active participant, joining with than that of some other people. I was one type other community members in the attempt to block of “insider.” This means I could gather data of demolition of the Hotel Aggie (also known as the certain sorts that were not available to the less “Terminal Building” and “Terminal Hotel”). Built involved. in 1924 in a town of around a thousand inhabit- ants, the hotel fell victim to population growth On the other edge, my partisanship clearly and urban development. Lofland says his role as poses the threat of bias. I have always been researcher began on September 18, 2000, as the aware of this, and I have tried my best to cor- demolition of the building began. rect for it. But, in the end, I cannot be the final judge. Each reader will have to form her or Before that, I was only and simply an involved his own assessment. I can hope, however, that citizen. Along with many other people, I was the “digital documentary” evidence I mention attempting to preserve the Terminal Building in above helps the study tell itself, so to speak. It some manner. This also explains why there are makes the reader less dependent on me than is so few photographs in this book taken by me the case with some other methods of represent- before that date, but many after that date. I had ing what happened. then begun seriously to document what was going on with a camera and field notes. (Lofland 2003: 20) Therefore, questions of “informed consent” As you can see, the seemingly simple process (now so often raised regarding research) were of observing social action as it occurs has subtle not pertinent before September 18. After that though important variations. As we saw in Chapter day, it was my practice to indicate to everyone I 2, all our thoughts occur within and are shaped by encountered that I was “writing a book” about paradigms, whether we’re conscious of it or not. the building. Qualitative field researchers have been unusually deliberate in framing a variety of paradigms to (Lofland 2003: 20) enrich the observation of social life.
316 ■ Chapter 10: Qualitative Field Research The impact of researcher paradigms on the and learning what others have said about it. (Li- conduct of research is nowhere more explicitly rec- brary research is discussed at length in Appendix A.) ognized than in the case of kaupapa Maori research, a form of participatory action research developed In the next phase of your research, you might within the indigenous Maori community of New wish to discuss the student political group with oth- Zealand. As Shayne Walker, Anaru Eketone, and ers who have already studied it or with anyone else Anita Gibbs (2006) report, an adherence to Maori likely to be familiar with it. In particular, you might culture shapes not only the purposes of such find it useful to discuss the group with one or more research but also its processes and practices. In a informants (discussed in Chapter 7). Perhaps you study of foster care, for example, the purpose of the have a friend who is a member, or you can meet study was established by those most directly con- someone who is. This aspect of your preparation cerned. The method of collecting data conformed is likely to be more effective if your relationship to Maori practices, including public gatherings. The with the informant extends beyond your research implications derived from the analysis of data were role. In dealing with members of the group as tailored to Maori ways of doing things. informants, you should take care that your initial discussions do not compromise or limit later aspects Conducting Qualitative of your research. Keep in mind that the impression Field Research you make on the informant, the role you establish for yourself, may carry over into your later effort. So far in this chapter we’ve examined the kinds of For example, creating the initial impression that topics appropriate to qualitative field research, spe- you may be an undercover FBI agent is unlikely to cial considerations in doing this kind of research, facilitate later observations of the group. and a sampling of paradigms that direct different types of research efforts. Along the way we’ve You should also be wary about the information seen some examples that illustrate field research in you get from informants. Although they may have action. To round out the picture, we turn now to more direct, personal knowledge of the subject specific ideas and techniques for conducting field under study than you do, what they “know” is research, beginning with how researchers prepare probably a mixture of fact and point of view. Mem- for work in the field. bers of the political group in our example (as well as members of opposing political groups) would be Preparing for the Field unlikely to provide completely unbiased informa- tion. Before making your first contact with the Suppose for the moment that you’ve decided to student group, then, you should already be quite undertake field research on a campus political familiar with it, and you should understand its organization. Let’s assume further that you’re not general philosophical context. a member of that group, that you do not know a great deal about it, and that you’ll identify yourself There are many ways to establish your initial to the participants as a researcher. This section will contact with the people you plan to study. How use this example and others to discuss some of the you do it will depend, in part, on the role you in- ways you might prepare yourself before undertak- tend to play. Especially if you decide to take on the ing direct observations. role of complete participant, you must find a way to develop an identity with the people to be stud- As is true of all research methods, you would be ied. If you wish to study dishwashers in a restau- well advised to begin with a search of the relevant rant, the most direct method would be to get a job literature, filling in your knowledge of the subject as a dishwasher. In the case of the student political group, you might simply join the group. Many of the social processes appropriate to field research are open enough to make your con- tact with the people to be studied rather simple and straightforward. If you wish to observe a
Conducting Qualitative Field Research ■ 317 How to Do It: Establishing Rapport In qualitative field research,it’s almost always vital that you be able to talk primarily (1) to elicit more information from the other person or (2) establish rapport with those you’re observing,especially if your obser- to answer questions they may have about you and your research.While vations include in-depth interviews and interactions.Rapport might be you don’t have to agree with any points of view expressed by your sub- defined as an open and trusting relationship.But how do you do that? jects,you should never argue with them nor try to change their minds. Keep reminding yourself that your genuine purpose is to understand Let’s assume that you’ll be identifying yourself as a researcher.You’ll their world and how it makes sense to them—whether it works for need to explain your research purpose in a nonthreatening way.Say that you or not.A little humility may help with this.You’ll be able to hear and you are there to learn about them and understand them,not to judge understand people better if you don’t start out feeling superior to them. them or cause them any problems.This will work best if you actually have a genuine interest in understanding the people you’re observing Be relaxed and appropriate to the setting.Some people are more and can communicate that interest to them.This gives them a sense of formal or informal than others,and you’ll do well to take on their general self-worth,which will increase their willingness to open up to you.Pre- style or at least find a way to relax with whatever style is most comfort- tending to be interested is not the same as really being interested.In fact, able for them.If you can get them to relax and enjoy the interaction, if you aren’t interested in learning what things look like from the point of you’ll have achieved the rapport you need.And you’ll probably enjoy the view of those you’re observing,you might consider another activity and interaction yourself. not waste their time and your own. It follows that you’ll function best as an attentive listener rather than as a talker.You should not remain mute,of course,but you should mass demonstration, just be there. If you wish to you observe and interpret events subsequently— observe patterns in jaywalking, hang around busy especially if you’re unaware of the influence. streets. Second, if the administrators approve of your Whenever you wish to make more-formal research project and encourage students and contact with the people, identifying yourself as a faculty to cooperate with you, the latter groups will researcher, you must establish a rapport with them. probably look on you as somehow aligned with the You might contact a participant with whom you administration, which can affect what they say to feel comfortable and gain that person’s assistance. you. For example, faculty members might be reluc- In studying a formal group, you might approach tant to tell you about plans to organize through the the groups’ leaders, or you might find that one of teamsters’ union. your informants can introduce you. (See “How to Do It: Establishing Rapport” for more on this.) In making direct, formal contact with the people you want to study, you’ll be required to Although you’ll probably have many options in give them some explanation of the purpose of your making your initial contact with the group, realize study. Here again, you face an ethical dilemma. that your choice can influence your subsequent Telling them the complete purpose of your research observations. Suppose, for example, that you’re might eliminate their cooperation altogether or studying a university and begin with high-level significantly affect their behavior. On the other administrators. This choice is likely to have a couple hand, giving only what you believe would be an of important consequences. First, your initial im- pressions of the university will be shaped to some rapport An open and trusting relationship; espe- extent by the administrators’ views, which will cially important in qualitative research between re- differ quite a bit from those of students or faculty. searchers and the people they’re observing. This initial impression may influence the way
318 ■ Chapter 10: Qualitative Field Research acceptable explanation may involve outright decep- structures than we would have imagined. Although tion. Your decisions in this and other matters will it was important for new immigrants to have a probably be largely determined by the purpose of support structure of family members already in the your study, the nature of what you’re studying, the United States, Menjívar found that her interview- observations you wish to use, and similar factors, ees were often reluctant to call on relatives for but you must also take ethical considerations into help, for several reasons. On the one hand, they account. might jeopardize those family members who were here illegally and living in poverty. At the same Previous field research offers no fixed rule— time, asking for help would put them in debt to methodological or ethical—to follow in this regard. those helping them out. Menjívar also discovered Your appearance as a researcher, regardless of your that Salvadoran gender norms put women im- stated purpose, may result in a warm welcome migrants in an especially difficult situation, because from people who are flattered that a scientist finds they were largely prohibited from seeking the help them important enough to study. Or, it may result of men they weren’t related to, lest they seem to in your being totally ostracized or worse. It prob- obligate themselves sexually. These are the kinds ably wouldn’t be a good idea, for example, to burst of discoveries that can emerge from open-ended, into a meeting of an organized crime syndicate in-depth interviewing. and announce that you’re writing a term paper on organized crime. See “Pencils and Photos in the We’ve already discussed interviewing in Chap- Hands of Research Subjects” for a look at some ter 9, and much of what was said there applies to inventive data-collection techniques. qualitative field interviewing. The interviewing you’ll do in connection with field observation, Qualitative Interviewing however, is different enough to demand a separate treatment. In surveys, questionnaires are rigidly In part, field research is a matter of going where structured; however, less-structured interviews are the action is and simply watching and listening. more appropriate to field research. Herbert and As the baseball legend Yogi Berra said, “You can Riene Rubin (1995: 43) describe the distinction as see a lot just by observing”—provided that you’re follows: “Qualitative interviewing design is flexible, paying attention. At the same time, as I’ve already iterative, and continuous, rather than prepared in indicated, field research can involve more-active advance and locked in stone.” They elaborate in inquiry. Sometimes it’s appropriate to ask people this way: questions and record their answers. Your on-the- spot observations of a full-blown riot will lack Design in qualitative interviewing is iterative. something if you don’t know why people are riot- That means that each time you repeat the basic ing. Ask somebody. process of gathering information, analyzing it, winnowing it, and testing it, you come closer to When Cecilia Menjívar (2000) wanted to learn a clear and convincing model of the phenom- about the experiences of Salvadoran immigrants in enon you are studying. . . . San Francisco, she felt in-depth interviews would be a useful technique, along with personal observa- The continuous nature of qualitative inter- tions. Before she was done, she had discovered a viewing means that the questioning is rede- much more complex system of social processes and signed throughout the project. qualitative interview Contrasted with survey (1995: 46– 47) interviewing, the qualitative interview is based on a set of topics to be discussed in depth rather than Unlike a survey, a qualitative interview based on the use of standardized questions. is an interaction between an interviewer and a respondent in which the interviewer has a general plan of inquiry, including the topics to be covered, but not a set of questions that must be asked with particular words and in a particular order. At the
Conducting Qualitative Field Research ■ 319 Pencils and Photos in the Hands of Research Subjects How would you go about studying the life conditions of Peruvian children to sketch these‘new’initiatives in their pictures on their Indians living in the Amazon rain forest? With minimal telecom- own,without prompts,is noteworthy. munications infrastructure and a slow ferry-based postal service in the vast region,a mail or telephone survey wouldn’t be the best approach.It (2006: 322) might occur to you to conduct in-depth interviews in which you would work from an outline of topics to be covered.Arvind Singhal and Eliza- The photographs taken by the adult women were equally beth Rattine-Flaherty (2006) opted for a very different approach,which revealing.Several drew attention to the patriarchal social structure.As put the subjects of study more in control of the research and allowed the authors report: for important but unexpected discoveries.They derived their inspiration from the work of the renowned Brazilian educator,Paulo Freire,who Several photographs depicted the subservient position of the once set out to measure exploitation among street children.Instead of Amazonian women relative to men,a situation that Minga Peru interviewing them,he gave them cameras and asked them to bring back seeks to address.For instance,Adela’s picture shows a middle- photographs of exploitation.As Singhal and Rattine-Flaherty report: aged Amazonian woman and her husband sitting on their porch and having a conversation.The woman,sporting a forlorn One child took a photo of a nail on a wall.It made no sense to expression,sits with her legs crossed while her husband stares adults,but other children were in strong agreement.The ensuing directly into the camera,squatting with his arms and feet spread discussions showed that many young boys of that neighborhood in an open position.Especially noticeable is the physical distance worked in the shoe-shine business.Their clients were mainly in the of about 10 feet that separates the woman and the man.When city,not in the barrio where they lived.As their shoe- shine boxes Adela was asked why she took the picture and why were the were too heavy for them to carry,these boys rented a nail on a wall man and woman sitting so far apart,she noted:“The woman (usually in a shop),where they could hang their boxes for the night. is sitting at one side of the house and he is on the other and To them,that nail on the wall represented“exploitation.”The“nail this was not anything unusual.”Upon probing,we learned that on the wall”photograph spurred widespread discussions in the Amazonian men determine how close the couple sits.If they are Peruvian barrio about other forms of institutionalized exploitation, sitting closer,and if the man has his arm around his partner,it is including ways to overcome them. his decision to do so.This authority also applies to initiation of sex:The man determines if and when sex will happen. (2006: 314) (2006: 323–24) Singhal and Rattine-Flaherty’s research involved gauging the qual- ity of life in the Peruvian Amazon and assessing the impact of programs This research not only illustrates some unusual data-collection launched by a Peruvian NGO,Minga Peru.To view society through the techniques,it also represents the spirit of participatory action research, eyes of children,the researchers set up drawing sessions with colored discussed earlier in this chapter.With a very different setting and pencils.In the spirit of reciprocity,one of the authors sketched pictures purpose,Pat O’Connor (2006) asked Irish adolescents to write essays of snowmen and jack-o’-lanterns that were a part of her growing up in about themselves and about Ireland,including drawings,poems,and the Midwest.In addition to depicting life in their villages and their close songs,looking for evidence of the impact of globalization in Ireland. relationship with the natural environment,the children’s sketches often Both studies demonstrate that qualitative field research can involve a featured examples of social change being brought about by the NGO’s lot more than just observing and interviewing. developmental programs. Sources: Arvind Singhal and Elizabeth Rattine-Flaherty,“Pencils and Photos as Tools These include sketches of chicken coops,fish farms,and agro- of Communicative Research and Praxis:Analyzing Minga Peru’s Quest for Social forestry projects.These enterprises,all launched by Minga Peru, Justice in the Amazon,”International Communication Gazette 68,no.4 (2006): began in the Peruvian Amazon only in the past few years.For 313–30;Pat O’Connor,“Globalization,Individualization and Gender in Adolescents’ Texts,”International Journal of Social Research Methodology 9,no.4 (October 2006): 261–77.
320 ■ Chapter 10: Qualitative Field Research same time, the qualitative interviewer, like the in for the excitement. Properly done, field research survey interviewer, must be fully familiar with the interviewing enables you to find out. questions to be asked. This allows the interview to proceed smoothly and naturally. Although you may set out to conduct inter- views with a reasonably clear idea of what you A qualitative interview is essentially a con- want to ask, one of the special strengths of field versation in which the interviewer establishes a research is its flexibility. In particular, the answers general direction for the conversation and pursues evoked by your initial questions should shape your specific topics raised by the respondent. Ideally, the subsequent ones. It doesn’t work merely to ask respondent does most of the talking. If you’re talk- preestablished questions and record the answers. ing more than 5 percent of the time, that’s probably Instead, you need to ask a question, listen care- too much. fully to the answer, interpret its meaning for your general inquiry, and then frame another question Steinar Kvale (1996: 3–5) offers two metaphors either to dig into the earlier answer or to redirect for interviewing: the interviewer as a “miner” or the person’s attention to an area more relevant to as a “traveler.” The first model assumes that the your inquiry. In short, you need to be able to listen, subject possesses specific information and that the think, and talk almost at the same time. interviewer’s job is to dig it out. By contrast, in the second model, the interviewer The discussion of probes in Chapter 9 provides a useful guide to getting answers in more depth wanders through the landscape and enters into without biasing later answers. More generally, field conversations with the people encountered. interviewers need the skills involved in being a The traveler explores the many domains of the good listener. Be more interested than interesting. country, as unknown territory or with maps, Learn to say things like “How is that?” “In what roaming freely around the territory. . . . The in- ways?” “How do you mean that?” “What would terviewer wanders along with the local inhabi- be an example of that?” Learn to look and listen tants, asks questions that lead the subjects to expectantly, and let the person you’re interviewing tell their own stories of their lived world. fill in the silence. Asking questions and noting answers is a natu- At the same time, you can’t afford to be a to- ral human process, and it seems simple enough to tally passive receiver. You’ll go into your interviews add it to your bag of tricks as a field researcher. Be with some general (or specific) questions you want a little cautious, however. Wording questions is a answered and some topics you want addressed. At tricky business. All too often, the way we ask ques- times you’ll need the skill of subtly directing the tions subtly biases the answers we get. Sometimes flow of conversation. we put our respondent under pressure to look good. Sometimes we put the question in a particu- There’s something we can learn in this regard lar context that omits altogether the most relevant from the martial arts. The aikido master never answers. resists an opponent’s blow but instead accepts it, joins with it, and then subtly redirects it in a more Suppose, for example, that you want to find appropriate direction. Field interviewing requires out why a group of students is rioting and pillaging an analogous skill. Instead of trying to halt your on campus. You might be tempted to focus your respondent’s line of discussion, learn to take what questioning on how students feel about the dean’s he or she has just said and branch that comment recent ruling that requires students always to carry back in the direction appropriate to your purposes. The Practice of Social Research with them on campus. Most people love to talk to anyone who’s really (Makes sense to me.) Although you may collect a interested. Stopping their line of conversation tells great deal of information about students’ attitudes them that you are not interested; asking them to toward the infamous ruling, they may be rioting for elaborate in a particular direction tells them that some other reason. Perhaps most are simply joining you are.
Conducting Qualitative Field Research ■ 321 Consider this hypothetical example in which Herbert and Riene Rubin offer several ways you’re interested in why college students chose to control a “guided conversation,” including the their majors. following: YOU: What are you majoring in? If you can limit the number of main topics, it is easier to maintain a conversational flow RESP: Engineering. from one topic to another. Transitions should be smooth and logical. “We have been talking YOU: I see. How did you come to choose about mothers, now let’s talk about fathers,” engineering? sounds abrupt. A smoother transition might be, “You mentioned your mother did not care RESP: I have an uncle who was voted the best how you performed in school—was your father engineer in Arizona in 1981. more involved?” The more abrupt the transi- tion, the more it sounds like the interviewer YOU: Gee, that’s great. has an agenda that he or she wants to get through, rather than wanting to hear what the RESP: Yeah. He was the engineer in charge of interviewee has to say. developing the new civic center in Tucson. It was written up in most of the engineering (1995: 123) journals. Because field research interviewing is so much YOU: I see. Did you talk to him about your be- like normal conversation, researchers must keep coming an engineer? reminding themselves that they are not having a normal conversation. In normal conversations, RESP: Yeah. He said that he got into engineer- each of us wants to come across as an interesting, ing by accident. He needed a job when he worthwhile person. If you watch yourself the next graduated from high school, so he went to time you chat with someone you don’t know too work as a laborer on a construction job. He well, you’ll probably find that much of your atten- spent eight years working his way up from tion is spent on thinking up interesting things to the bottom, until he decided to go to college say—contributions to the conversation that will and come back nearer the top. make a good impression. Often, we don’t really hear each other, because we’re too busy thinking of YOU: So is your main interest civil engineering, what we’ll say next. As an interviewer, the desire like your uncle, or are you more interested to appear interesting is counterproductive. The in some other branch of engineering? interviewer needs to make the other person seem interesting, by being interested—and by listening RESP: Actually, I’m leaning more toward electri- more than talking. (Do this in ordinary conversa- cal engineering—computers, in particular. tions, and people will actually regard you as a great I started messing around with microcom- conversationalist.) puters when I was in high school, and my long-term plan is . . . John Lofland and his colleagues (2006: 69–70) suggest that researchers should adopt the role of Notice how the interview begins by wandering the “socially acceptable incompetent” when inter- off into a story about the respondent’s uncle. The viewing. That is, offer yourself as someone who first attempt to focus things back on the student’s does not understand the situation you find yourself own choice of major (“Did you talk to your uncle in and must be helped to grasp even the most basic . . . ?”) fails. The second attempt (“So is your and obvious aspects of that situation: “A naturalistic main interest . . . ?”) succeeds. Now the student is investigator, almost by definition, is one who does providing the kind of information you’re looking not understand. She or he is ‘ignorant’ and needs for. It’s important for field researchers to develop the ability to “control” conversations in this fash- ion. At the same time, of course, you need to be on the alert for “distractions” that point to unexpect- edly important aspects of your research interest.
322 ■ Chapter 10: Qualitative Field Research to be ‘taught.’ This role of watcher and asker of Focus Groups questions is the quintessential student role” (Lofland et al. 2006: 69). Although our discussions of field research so far have focused on studying people in the process Interviewing needs to be an integral part of of living their lives, researchers sometimes bring the entire field research process. Later, I’ll stress people into the laboratory for qualitative inter- the need to review your observational notes every viewing and observation. The focus group method, night—making sense out of what you’ve ob- which is also called group interviewing, is essen- served, getting a clearer feel for the situation you’re tially a qualitative method. It is based on struc- studying, and finding out what you should pay tured, semistructured, or unstructured interviews. more attention to in further observations. In the It allows the researcher/interviewer to question same fashion, you’ll need to review your notes on several individuals systematically and simultane- interviews, recording especially effective questions ously. This data-collection technique is frequently and detecting all those questions you should have used in political and market research but is used asked but didn’t. Start asking such questions the for other purposes as well. In Silent Racism, for next time you interview. If you’ve recorded the example, Barbara Trepagnier (2006) used focus interviews, replay them as a useful preparation for groups to examine the persistence of racism among future interviews. “well-meaning white people.” Steinar Kvale (1996: 88) details seven stages in In a hypothetical market-research example, the complete interviewing process: imagine that you’re thinking about introducing a new product. Let’s suppose that you’ve invented a 1. Thematizing: Clarifying the purpose of the inter- new computer that not only does word processing, views and the concepts to be explored spreadsheets, data analysis, and the like but also contains a fax machine, AM/FM/TV tuner, CD and 2. Designing: Laying out the process through DVD player/recorder, microwave oven, denture which you’ll accomplish your purpose, includ- cleaner, and coffeemaker. To highlight its computing ing a consideration of the ethical dimension and coffee-making features, you’re thinking of calling it “The Compulator.” You figure the new 3. Interviewing: Doing the actual interviews computer will sell for about $28,000, and you 4. Transcribing: Creating a written text of the want to know whether people are likely to buy it. Your prospects might be well served by focus interviews groups. 5. Analyzing: Determining the meaning of gath- In a focus group, typically 5 to 15 people are ered materials in relation to the purpose of the brought together in a private, comfortable envi- study ronment to engage in a guided discussion of some 6. Verifying: Checking the reliability and validity of topic—in this case, the acceptability and salability the materials of the Compulator. The subjects are selected on the 7. Reporting: Telling others what you’ve learned basis of relevance to the topic under study. Given the likely cost of the Compulator, your focus group As with all other aspects of field research, participants would probably be limited to upper- interviewing improves with practice. Fortunately, income groups, for example. Other, similar consid- it’s something you can practice any time you want. erations might figure into the selection. Practice on your friends. Participants in focus groups are not likely to be focus group A group of subjects interviewed chosen through rigorous probability-sampling together, prompting a discussion. The technique is methods. This means that the participants do not frequently used by market researchers, who ask a statistically represent any meaningful population. group of consumers to evaluate a product or discuss a type of commodity, for example.
Conducting Qualitative Field Research ■ 323 However, the purpose of the study is to explore 5. Groups are difficult to assemble. rather than to describe or explain in any definitive sense. Nevertheless, typically more than one focus 6. The discussion must be conducted in a condu- group is convened in a given study because of the cive environment. serious danger that a single group of 7 to 12 people (1988: 48) will be too atypical to offer any generalizable insights. As we’ve seen, the group interview presents several advantages, but it also has its challenges. William Gamson (1992) used focus groups to In a focus group interview, much more than in examine how U.S. citizens frame their views of po- any other type of interview, the interviewer has to litical issues. Having picked four issues—affirmative develop the skills of a moderator. Controlling the action, nuclear power, troubled industries, and dynamic within the group is a major challenge. the Arab-Israeli conflict—Gamson undertook a Letting one interviewee dominate the focus group content analysis of press coverage to get an idea of interview reduces the likelihood that the other the media context within which we think and talk subjects will express themselves. This can generate about politics. Then the focus groups were con- the problem of group conformity or groupthink, vened for a firsthand observation of the process of which is the tendency for people in a group to people discussing issues with their friends. conform with the opinions and decisions of the most outspoken members of the group. This danger Richard Krueger points to five advantages of is compounded by the possibility that only one or focus groups: two people sometimes dominate the conversation. Interviewers need to be aware of this phenom- 1. The technique is a socially oriented research enon and try to get everyone to participate fully method capturing real-life data in a social on all the issues brought in the interview. Adding environment. to the challenge, of course, is that the interviewer must resist overdirecting the interview and the 2. It has flexibility. interviewees, thus bringing her or his own views into play. 3. It has high face validity. Although focus group research differs from 4. It has speedy results. other forms of qualitative field research, it further illustrates the possibilities for doing social research 5. It is low in cost. face-to-face with those we wish to understand. In (1988: 47) addition, David Morgan (1993) suggests that focus groups are an excellent device for generating ques- In addition to these advantages, group dynam- tionnaire items for a subsequent survey. ics frequently bring out aspects of the topic that would not have been anticipated by the researcher Because they center on a particular topic and and would not have emerged from interviews with take relatively little time, focus groups are typically individuals. In a side conversation, for example, a regarded as an “in-depth” research technique. How- couple of the participants might start joking about ever, Carolina Överlien, Karin Aronsson, and the results of leaving out one letter from a prod- Margareta Hydén (2005) have used the technique uct’s name. This realization might save the manu- successfully for extended discussions of sexuality, facturer great embarrassment later on. among Swedish teenagers in a youth detention home. Krueger also notes some disadvantages of the focus group method, however: Like other social research techniques, focus groups are adapting to new communication modali- 1. Focus groups afford the researcher less control ties. George Silverman (2005), for example, offers a than individual interviews. discussion of telephone and online focus groups. 2. Data are difficult to analyze. 3. Moderators require special skills. 4. Difference between groups can be troublesome.
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 534
- 535
- 536
- 537
- 538
- 539
- 540
- 541
- 542
- 543
- 544
- 545
- 546
- 547
- 548
- 549
- 550
- 551
- 552
- 553
- 554
- 555
- 556
- 557
- 558
- 559
- 560
- 561
- 562
- 563
- 564
- 565
- 566
- 567
- 568
- 569
- 570
- 571
- 572
- 573
- 574
- 575
- 576
- 577
- 578
- 579
- 580
- 581
- 582
- 583
- 584
- 585
- 586
- 587
- 588
- 589
- 590
- 591
- 592
- 593
- 594
- 595
- 596
- 597
- 598
- 599
- 600
- 601
- 602
- 603
- 604
- 605
- 606
- 607
- 608
- 609
- 610
- 611
- 612
- 613
- 614
- 615
- 616
- 617
- 618
- 619
- 620
- 621
- 622
- 623
- 624
- 625
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 625
Pages: