Comparison of the Different Survey Methods 275 and setting a deadline for participation did increase the Census Bureau and interviewers can be trained response rates. The years ahead will see many ex- to deal nith the concept, it's extremely difficult to periments aimed at improving the effectiveness of communicate in a self-administered questionnaire. online surveys. This advantage of inteniew surveys pertains gener- ally to all complicated contingency questions. Comparison of the Different Survey Methods With interviews, you can conduct a survey based on a sample of addresses or phone numbers Now that we've seen several ways to collect survey rather than on names. An inteniewer can arrive at data, let's take a moment to compare them directly. an assigned address or call the assigned number, in- troduce the survey, and even-following instruc- Self-administered questionnaires are generally tions-choose the appropriate person at that cheaper and quicker than face-to-face interview address to respond to the survey. In contrast, self- surveys. These considerations are likely to be im- administered questionnaires addressed to \"occu- portant for an unfunded student wishing to under- pant\" receive a notoriously low response. take a survey for a term paper or thesis. Moreover, if you use the self-administered mail format it costs Finally, as we've seen, interviewers questioning no more to conduct a national survey than a local respondents face-to-face can make important ob- one of the same sample size. In contrast a national servations aside from responses to questions asked interview survey (either face-to-face or by tele- in the interviewo In a household interview, they phone) would cost far more than a local one. Also, may note the characteristics of the neighborhood, mail surveys typically require a small staff: You the dwelling unit. and so forth. They can also note could conduct a reasonable mail survey by your- characteristics of the respondents or the quality of self. although you shouldn't underestimate the their interaction nith the respondents-whether work involved. Further, respondents are sometimes the respondent had difficulty conununicating, was reluctant to report controversial or deviant atti- hostile, seemed to be lying, and so on. tudes or behaviors in interviews but are vvilling to respond to an anonymous self-administered The chief advantages of telephone surveys over questionnaire. those conducted face-to-face center primarily on time and money. Telephone interviews are much Interview surveys also offer many advantages. cheaper and can be mounted and executed quickly_ For example, they generally produce fewer incom- Also, interviewers are safer when intervievving plete questionnaires. Although respondents may people living in high-crime areas. Moreover, the skip questions in a self-administered questionnaire, impact of the inteniewers on responses is some- interviewers are trained not to do so. In CATI sur- what lessened when the respondents can't see veys, the computer offers a further check on this. In- them. As only one indicator of the popularity of terview surveys, moreover, have typically achieved telephone interviewing, when Johnny Blair and his higher completion rates than self-administered colleagues (1995) compiled a bibliography on questionnaires have. sample designs for telephone interviews, they listed over 200 items. Although self-administered questionnaires may be more effective for sensitive issues, interview Online surveys have many of the strengths and surveys are definitely more effective for compli- weaknesses of mail surveys. Once the available cated ones. Prime examples include the enumera- software has been further developed, they will likely tion of household members and the determination be substantially cheaper. An important weakness, of whether a given address corresponds to more however, lies in the difficulty of assuring that re- than one housing unit. Although the concept of spondents to an online survey will be representa- housing unit has been refined and standardized by tive of some more general population. Clearly, each survey method has its place in social research. Ultimately, you must balance the advantages and disadvantages of the different
276 Chapter 9: Survey Research methods in relation to your research needs and easy to apply the same definitions uniformly to all your resources\" subjects. The survey researcher is bound to this requirement by having to ask exactly the same Strengths and Weaknesses questions of all subjects and having to impute the of Survey Research same intent to all respondents giving a particular response. Regardless of the specific method used, surveys- like other modes of observation in social research- Survey research also has several weaknesses. have special strengths and weaknesses. You should First, the requirement of standardization often keep these in mind when determining whether a seems to result in the fitting of round pegs into survey is appropriate for your research goals. square holes. Standardized questionnaire items of- ten represent the least common denominator in as- Surveys are particularly useful in describing the sessing people's attitudes, orientations, circum- characteristics of a large population. A carefully se- stances, and experiences. By designing questions lected probability sample in combination vvith a that will be at least minimally appropriate to all re- standardized questionnaire offers the possibility of spondents, you may miss what is most appropriate making refined descriptive assertions about a stu- to many respondents. In this sense, surveys often dent body, a city, a nation, or any other large popu- appear superficial in their coverage of complex lation. Surveys determine unemployment rates, topics. Although this problem can be partly offset voting intentions, and the like with uncanny accu- by sophisticated analyses, it is inherent in survey racy. Although the examination of official docu- research\" ments-sud1 as marriage, birth, or death records- can provide equal accuracy for a few topics, no Sinularly, survey research can seldom deal with other method of observation can provide this gen- the context of social life. Although questionnaires eral capability\" can provide information in this area, the survey re- searcher rarely develops the feel for the total life Surveys-especially self-administered ones- situation in which respondents are thinking and make large samples feasible. Surveys of 2,000 re- acting that, say, the participant observer can (see spondents are not unusual. A large number of cases Chapter 10). is very important for both descriptive and explana- tory analyses, especially wherever several variables In many ways, surveys are inflexible. Studies are to be analyzed simultaneously. involving direct observation can be modified as field conditions warrant, but surveys typically require In one sense, surveys are flexible. Many ques- that an initial study design remain unchanged tions can be asked on a given topic, giving you throughout As a field researcher, for example, you considerable flexibility in your analyses. Whereas can become aware of an important new variable an experimental design may require you to commit operating in the phenomenon you're studying and yourself in advance to a particular operational begin making careful observations of it. The survey definition of a concept, surveys let you develop op- researcher would probably be unaware of the new erational definitions from actual observations. variable's importance and could do nothing about it in any event. Finally, standardized questionnaires have an important strength in regard to measurement gen- Finally, surveys are subject to the artificiality erally\" Earlier chapters have discussed the ambigu- mentioned earlier in connection with experiments. ous nature of most concepts: They have no ulti- Finding out that a person gives conservative an- mately real meanings. One person's religiosity is swers in a questionnaire does not necessarily mean quite different from another's. Although you must the person is conservative; finding out that a per- be able to define concepts in those ways most son gives prejudiced answers in a questionnaire relevant to your research goals, you may not find it does not necessarily mean the person is prejudiced. This shortcoming is especially salient in the realm of action. Surveys cannot measure social action;
Secondary Analysis 277 they can only collect self-reports of recalled past ac- through either interviewing or self-administered tion or of prospective or hypothetical action. questionnaires. As you've gathered, surveys are usually major undertakings. It's not unusual for a The problem of artificiality has two aspects. large-scale survey to take several months or even First, the topic of study may not be amenable to more than a year to progress from conceptualiza- measurement through questionnaires. Second, the tion to data in hand. (Smaller-scale surveys can, of act of studying that topic-an attitude, for ex- course, be done more quickly.) Through a method ample-may affect it A survey respondent may called secol1dmy analysis, however, researchers can have given no thought to whether the governor pursue their particular social research interests- should be impeached until asked for his or her analyzing survey data from, say, a national sample opinion by an interviewer. He or she may, at that of 2,000 respondents-while avoiding the enor- point, form an opinion on the matter. mous expenditure of time and money such a sur- vey entails. Survey research is generally weak on validity and strong on reliability\" In comparison with field Secondary analysis is a form of research in research, for example, the artificiality of the survey which the data collected and processed by one re- format puts a strain on validity\" As an illustration, searcher are reanalyzed-often for a different pur- people's opinions on issues seldom take the form of pose-by another. Beginning in the 1960s, survey strongly agreeing, agreeing, disagreeing, or strongly researchers became aware of the potential value disagreeing with a specific statement\" Their survey that lay in archiving survey data for analysis by responses in such cases must be regarded as ap- scholars who had nothing to do with the survey proximate indicators of what the researchers had in design and data collection. Even when one re- mind when they framed the questions. This com- searcher had conducted a survey and analyzed the ment, however, needs to be held in the context of data, those same data could be further analyzed by earlier discussions of the ambiguity of validity itself. others who had slightly different interests\" Thus, if To say something is a valid or an invalid measure you were interested in the relationship between assumes the existence of a \"real\" definition of political \\iews and attitudes toward gender equal- what's being measured, and many scholars now ity, you could examine that research question reject that assumption. through the analysis of any data set that happened to contain questions relating to those t\\'vo variables. Reliability is a clearer matter. By presenting all subjects with a standardized stimulus, survey re- The initial data archives were very much like search goes a long way toward eliminating unrelia- book libraries, with a couple of differences. First, bility in observations made by the researcher. More- instead of books, the data archives contained data over, careful wording of the questions can also sets: first as punched cards, then as magnetic tapes. significantly reduce the subject's own unreliability. Today they're typically contained on computer disks, CD-ROMs, or online servers. Second, whereas As with all methods of observation, a full you're expected to return books to a conventional awareness of the inherent or probable weaknesses library, you can keep the data obtained from a data of survey research can partially resolve them in archive. some cases. Ultimately, though, researchers are on the safest ground when they can employ several research methods in studying a given topic Secondary Analysis secondary analysis A form of research in which the data collected and processed by one researcher As a mode of observation, survey research involves are reanalyzed-often for a different purpose-by the folloV\\ing steps: (1) questionnaire construction, another. This is especially appropriate in the case of (2) sample selection, and (3) data collection, survey data. Data archives are repositories or Ii· braries for the storage and distribution of data for secondary analysis\"
278 Chapter 9: Survey Research The best-known current example of secondary online repository for thousands of polls conducted analysis is the General Social Survey (GSS)\" The in the United States and 70 other nations. A paid National Opinion Research Center (NORC) at the subscription allows users to obtain specific data re- University of Chicago conducts this major national sults from studies they specify, rather than obtaining survey, currently every other year, to collect data whole studies\" Although the cost of the subscription on a large number of social science variables\" These may be too steep for the average student, you might surveys are conducted precisely for the purpose check to see if your school's library has subscribed. of making data available to scholars at little or no cost and are supported by a combination of private Because secondary analysis has typically in- and government funding\" Recall that the GSS volved obtaining a data set and undertaking an ex- was created by James A Davis in 1972; it is cur- tensive analysis, I would like you to consider an- rently directed by Davis, Tom W. Smith, and Peter other approach as welL Often you can do limited analyses by investing just a little time\" Let's say v\" Marsden\" Their considerable ongoing efforts you're writing a term paper about the impact of re- ligion in contemporary American life. You want make an unusual contribution to social science re- to comment on the role of the Roman Catholic search and to education in social science\" You can church in the debate over abortion\" Although you learn more about the GSS at http://webapp,icpsr might get away with an offhand, unsubstantiated .umich\"edu/GSS/\" assertion, imagine how much more powerful your paper would be with this addition: Numerous other resources are available for identifying and acquiring survey data for secondary L Go to the General Social Survey website (see analysis\" The Roper Center for Public Opinion Figure 9-7): Research (http://www\"ropercenteLuconn.edu/) at http://webapp.icpsLumicludu/GSS/ the University of Connecticut is one excellent resource. The center also publishes the journal Pub- 2. Click \"Subject\" under \"Codebook Indexes,\" lic Perspective, on public opinion polling. Polling the then go to \"Abortion\"\" This will show you the Nations (http://www\"pollingthenationHom/) is an GSSDIRS General Social SurJey 1972 - 2000 Cumulative Codebook Best used w;th Netscape Communicator 4,x and above or MIcrosoft Internet Explorer 4,;( and abOVe FIGURE 9-7 The General Social Survey Codebook Website
Secondary Analysis 279 SDA Tables Program (Selected Study: GSS 1972-2000 Cumulative Datafile) Help: General! Recoding Variables REQUIRED Variable names to specify Row:! RACDIF2 OPTIONAL Variable names to specify IColumn: DEGREE I =SCeloecntitorn~oF=li:-;:-lt-er-;(s-;)I:17Y\"\"E===A==R\"\"(2=\"O===O\"=O:=) ===d-_ _--, Example: age (18-50) gender(1) IWeight: No Weight 0 0Percentaging: ~Column Row Total o 0 0Other options Statistics Suppress table Question text Ii21 0Color coding Show T-statistic I II IRun the Table Clear Fields FIGURE 9-8 necessarily agree, as shown in Table\" \" \"\" Moreover, Entering tile Table Request this might be just the beginning of an analysis that looks a bit more deeply into the matter, as de- questions asked about attitudes on abortion scribed in Chapter 14, on quantitative analysis\" (e\"g .. ABAl'\\Y asks if a women should be per- mitted a legal abortion for any reason). The key advantage of secondary analysis is that 3 Click the \"Analyze\" button\" it's cheaper and faster than doing original surveys, 4 Select \"Frequencies or crosstabulation\" and and, depending on who did the original survey, click the \"Start\" button\" 5 Enter ABAl'\\Y as the dependent (\"Row\") Vart8bres variable and RELIG as the independent (\"Col- umn\") variable\" In the \"Selection Filter\" field, Role Name Label Ran enter \"YEAR(2000)\" to limit the analysis to Row abany ABORTION IF WOMAN WANTS FOR ANY REASON 1- that year's survey. as shown in Figure 9-8. 6\" Press \"Run the Table\" to run the requested COlumn rell9 RS RELIGIOUS PREFERENCE 1-1 analysis Filler year(2000) GSS YEAR FOR THIS RESPONDENT 1972- The results of your analysis, shown in Figure 9-9, may suq)J'ise YOlL Whereas Catholics Cells contain: 1 23 4 5 are less supportive of abortion (33.3 percent) than -Column percent PROTESTANT CATHOLIC JEWISH NONE OTHER Jews (76\" 1 percent) and those with no religion (SPECIFy) (57.3 percent) are, they do not differ substantially ·N of cases from American Protestants (352 percent)\" 352 I··~~33\"3!:~~~ +:~~. 1:YES 337 Imagine a term paper that says, \"Whereas 142 the Roman Catholic Church has taken a strong, abany 2:NO '~;~~1 -~a! 23,9 42.7 23\"8 official position on abortion, many Catholics do not 11 105 5 COL TOTAL 100\"0 100.0 100.0 100.0 100.0 958 426 46 246 21 FIGURE 9-9 Impact of Religion on Attitude toward Abortion
280 Chapter 9: Survey Research you may benefit from the work of topflight profes- Topics Appropriate for Survey Research sionals. The ease of secondary analysis has also en- hanced the possibility of meta-analysis, in which a o Survey research is especially appropriate for researcher brings together a body of past research making descriptive studies of large populations; on a particular topic To gain confidence in your survey data may be used for explanatory pur- understanding of the relationship between religion poses as well. and abortion, for example, you could go beyond the GSS to analyze similar data collected in dozens o Questionnaires provide a method of collecting or even hundreds of other studies. data by (1) asking people questions or (2) ask- ing them to agree or disagree with statements There are disadvantages inherent in secondary representing different points of view. Questions analysis, howeveL The key problem involves the may be open-ended (respondents supply their recurrent question of valiclity. When one researcher own answers) or closed-ended (they select collects data for one particular purpose, you have from a list of provided answers), no assurance that those data will be appropriate for your research interests, Typically, you'll find that Guidelines for Asking Questions the original researcher asked a question that \"comes close\" to measuring what you're interested o Items in a questionnaire should follow several in, but you'll wish the question had been asked just guidelines: (1) The form of the items should be a little differently-or that another, related ques- appropriate to the project; (2) the items must tion had also been asked, Your question, then, is be clear and precise; (3) the items should ask whether the question that was asked provides a only about one thing (i.e., double-barreled valid measure of the variable you want to analyze. questions should be avoided); (4) respondents Nevertheless, secondary analysis can be inunensely must be competent to answer the item; (5) re- useful. Moreover, it illustrates once again the range spondents must be \"villing to answer the item; of possibilities available in finding the answers to (6) questions should be relevant to the respon- questions about social life. Although no single dent; (7) items should ordinarily be short; method unlocks all puzzles, there is no limit to the (8) negative terms should be avoided so as not ways you can find out about things. And when you to confuse respondents; (9) the items should be zero in on an issue from several independent direc- worded to avoid biasing responses, tions, you gain that much more expertise, Questionnaire Construction I've discussed secondary analysis in this chapter on survey research because it's the type of analysis o The format of a questionnaire can influence the most associated with the technique, However, quality of data collected. there is no reason that the reanalysis of social re- search data needs to be limited to those collected in o A clear format for contingency questions is nec- surveys. Nigel Fielding (2004), for example, has ex- essary to ensure that the respondents answer amined the possibilities for the archiving and re- all the questions intended for them. analysis of qualitative data as well. o The matrix question is an efficient format for MAIN POINTS presenting several items sharing the same re- sponse categories. Introduction o The order of items in a questionnaire can influ- o Survey research, a popular social research ence the responses given. method, is the administration of questionnaires to a sample of respondents selected from some o Clear instructions are important for getting ap- population. propriate responses in a questionnaire. o Questionnaires should be pretested before be- ing administered to the study sample, o Questionnaires may be administered in three basic ways: through self-administered question- naires, face-to-face interviews, or telephone surveys.
Key Tenus 281 Self-Administered Questionnaires the possibility of anonymity and privacy to o It's generally advisable to plan follow-up mail- encourage candid responses on sensitive issues. o The advantages of an interview survey over a ings in the case of self-administered question- self-administered questionnaire are fewer in- naires, sending new questionnaires to those re- complete questionnaires and fewer misunder- spondents who fail to respond to the initial stood questions, generally higher return rates, appeal. Properly monitoring questionnaire re- and greater flexibility in terms of sampling and turns will provide a good guide to when a special observations. follow-up mailing is appropriate. o The principal advantages of telephone surveys over face-to-face interviews are the savings in Interview Surveys cost and time., Telephone interviewers have o The essential characteristic of interviewers is more safety than in-person interviewers do, and they may have a smaller effect on the that they be neutral; their presence in the data- interview itself. collection process must have no effect on the o Online surveys have many of the strengths and responses given to questionnaire items, weaknesses of mail surveys. Although they're o Interviewers must be carefully trained to be cheaper to conduct, ensuring that the respon- familiar 'with the questionnaire, to follow the dents represent a more general population can question wording and question order exactly, be difficult and to record responses exactly as they are given, Strengths and Weaknesses of Survey o Interviewers can use probes to elicit an elabora- Research tion on an incomplete or ambiguous response. o Survey research in general offers advantages in Probes should be neutral. Ideally, all interview- ers should use the same probes, terms of economy, the amount of data that can be collected, and the chance to sample a large Telephone Surveys population. The standardization of the data col- o Telephone surveys can be cheaper and more lected represents another special strength of survey research. efficient than face-to-face interviews, and they o Survey research has several weaknesses: It is can permit greater control over data collection. somewhat artificiaL potentially superficiaL and The development of computer-assisted tele- relatively inflexible. Using surveys to gain a full phone interviewing (CATI) is especially sense of social processes in their natural settings promising. is difficult In generaL survey research is comparatively weak on validity and strong on New Technologies and Survey Research reliability. o New technologies offer additional opportunities Secondary Analysis for social researchers. They include various o Secondary analysis provides social researchers kinds of computer-assisted data collection and analysis as well as the chance to conduct sur- with an important option for \"collecting\" data veys by fax or over the Internet The latter two cheaply and easily but at a potential cost in methods, however, must be used with caution validity. because respondents may not be representative of the intended population. KEY TERMS Comparison of the Different Survey The following terms are defined in context in the Methods chapter and at the bottom of the page where the term o The advantages of a self-administered is introduced, as well as in the comprehensive glossary at the back of the book. questionnaire over an interview survey are economy, speed, lack of interviewer bias, and
282 Chapter 9: Survey Research bias probe ADDITIONAl. READINGS closed-ended questions questionnaire contingency question respondent Babbie, Earl. 1990. Survey Research lvlerhods. interview response rate Belmont, CA: Wadsworth A comprehensive open-ended questions secondary analysis overview of survey methods. (You thought I'd say it was lousy?) This textbook, although over- REVIEW QUESTIONS AND EXERCISES lapping the present one somewhat, covers as- pects of survey techniques omitted here. I. For each of the following open-ended questions, construct a closed-ended question that could be Bradburn, Norman M., and Seymour Sudman. used in a questionnaire. 1988. Polls and Surveys. Understanding What They Tell Us. San Francisco: Jossey-Bass. These vet- a. What was your family's total income last eran survey researchers answer questions about year? their craft the general public commonly ask. b. How do you feel about the space shuttle Couper, Mick p, Michael W. Traugott, and Mark J. program? Lamias. 2001. \"Web Survey Design and Admin- istration.\" Public Opinion Quarterly 65: 230-53. e. How important is religion in your life? An experimental study to determine which of three different web survey designs elicited the d. What was your main reason for attending best results. college? Dillman, Don A. 1999. iv/ail and Telephone Surveys.: e. What do you feel is the biggest problem fac- The Tailored Design lv/er/zod. 2nd ed. New York: ing your community? Wiley. An excellent review of the methodologi- cal literature on mail and telephone surveys. 2. Construct a set of contingency questions for use Dillman makes many good suggestions for im- in a self-administered questionnaire that would proving response rates. (This classic book was solicit the following information: originally published in 1978 under the name Alail and Telep/zone Surveys The Toral DesiglZ a. Is the respondent employed? Mer/zod.) b. If unemployed, is the respondent looking Elder, Glen H., Jr., Eliza K. Pavalko, and Elizabeth for work? c.. CHpp. 1993. Working with Archival Data: Smdy- e. If the unemployed respondent is not look- ing for work, is he or she retired, a student, ing Lives. Newbury Park, CA: Sage. This book or a homemaker? discusses the possibilities and techniques for us- ing existing data archives in the United States, d. If the respondent is looking for work, how especially those providing longitudinal data. long has he or she been looking? Feick Lawrence F. 1989. \"Latent Class Analysis of 3. Find a questionnaire printed in a magazine or Survey Questions that Include Don't Know newspaper (for a reader survey, for example). Responses.\" Public Opinion Quarterly 53: 525-47. Consider at least five of the questions in it \"Don't know\" can mean a variety of things, as and critique each one either positively or this analysis indicates. negatively. Fowler, Floyd ], Jr. 1995. Improving Survey QZlesriolls.: 4 Look at your appearance right now. Identify as- Design and Evaluation. Thousand Oaks, CA: Sage. pects of your appearance that might create a A comprehensive discussion of questionnaire problem if you were interviewing a general construction, including many suggestions for cross section of the pu blie. pretesting questions. This book discusses the logic of obtaining information through survey 5. Locate a survey being conducted on the web. questions and gives numerous guidelines for be- Briefly describe the survey and discuss its ing effective. It also offers several examples of strengths and weaknesses. questions you might use. Groves, Robert M. 1990. \"Theories and Methods of Telephone Surveys.\" Pp. 221-40 in Annual Re- !'iel!' of Sociology (voL 16), edited by W Richard Scott 'and Judith Blake. Palo Alto, CA: Annual
Online Study Resources 283 Reviews. An attempt to place telephone surveys QlIarterly 53: 495-524. The authors asked six in the context of sociological and psychological target questions in a telephone survey of I, I 00 theories and to address the various kinds of er- respondents, varying the questions immediately rors common to this research method. preceding the target questions. They found sub- stantial differences. Holbrook Allyson L. Melanie c.. Green, and Jon A. Williams, Robin M., Jr. 1989. \"The American Sol- dier: An Assessment, Several Wars Later. \" Public Krosnick. 2003. \"Telephone versus Face-to-Face Opinion Quarterly 53: 155-74. One of the classic Interviewing of National Probability Samples studies in the history of survey research is re- with Long Questionnaires: Comparisons of Re- viewed by one of its authors . spondent SatisfiCing and Social Desirability Re- sponse Bias.\" Public Opinion Quanerly 67: 79-125. SPSS EXERCISES A detailed examination of the differences to be expected from these two methods of conducting See the booklet that accompanies your text for exer- interview surveys. cises using SPSS (Statistical Package for the Social Sci- ences). There are exercises offered for each chapter, Miller, Delbert. 1991 . Handbook of Research Design and and you'll also find a detailed primer on using SPSS. Social Measurelilem. Newbury Park, CA: Sage. A Online Study Resources powerful reference work, this book, especially Part 6, cites and describes a wide variety of oper- Sociology.~ Now'\"; Research Methods ational measures used in earlier social research. In several cases, the questionnaire formats used L Before you do your final review of the chapter, are presented. Though the quality of these illus- take the SociologyNow: Research lvferhods diagnos- trations is uneven, they provide excellent exam- tic quiz to help identify the areas on which you ples of possible variations. should concentrate. You'll find information on this online tool. as well as instructions on how jvloore, David W. 2002. \"Measuring New Types of to access all of its great resources, in the front of Question-Order Effects: Additive and Subtrac- the book.. tive.\" Public Opinion Quarrerly 66: 80-91. An ex- tensive examination of the various ways that 2. As you review, take advantage of the Sociology question wording can affect responses. No\\\\'.: Researciz Metizods customized study plan, based on your quiz results . Use this study plan Sheatsley, Paul F. 1983. \"Questionnaire Construc- with its interactive exercises and other re- tion and Item Writing.\" Pp. 195-230 in Hand- sources to master the materiaL b,lok of SlllT\".\\' Researciz, edited by Peter H. Rossi. 3.. When you're finished with your review, take the posttest to confirm that you're ready to James D. Wright. and Andy B. Anderson. New move on to the next chapter. York: Academic Press. An excellent examination of the topic by an expert in the field. WEBSITE FOR THE PRACTICE OF SOCIAL RESEARCH 11 TH EDITION Smith, Eric R. A N, and Peverill Squire. 1990. \"The Effects of Prestige Names in Question Wording.\" Go to your book's website at http://sociology Pliblic Opinion QlIanerly 54: 97-116. Not only do .wadsworth.com/ babbie_practice II e for tools to prestigious names affect the overall responses aid you in studying for your exams. You'll find Tilto- given to survey questionnaires, they also affect rial QlIizzes with feedback Interner Exercises, Flashcards, such things as the correlation between educa- and Cizaprer Tiltorials, as well as E'([ended Projecrs, In/o- tion and the number of \"don't know\" answers\" hac College Ediriol1 search terms, Social Researciz in Cyberspace, GSS Dara, H0b Links, and primers for us- Swafford, MichaeL 1992. \"Soviet Survey Research: The 1970's vs . the 1990's.\" AAPOR News 19 (3): 3-4. The author contrasts the general repres- sion of survey research during his first visit in 1973-1974 with the renewed use of the method in more recent times . He notes, for example, that the Soviet government commissioned a na- tional survey to determine public opinion on the possible reunification of Germany. Tourangeau, Roger, Kenneth A. Rasinski. Norman Bradburn, and Roy D'Andrade. 1989. \"Carry- over Effects in Altitude Surveys . \" Public Opinion
284 Chapter 9: Survey Research ing various data-analysis software such as SPSS and publication, provide opportunities to learn about sur- NVivo. vey research. WEB LINKS FOR THIS CHAPTER UC Berkeley Survey Research Center http://srcweb.berkeley.edu/ Please realize that the Internet is an evolv- ing entity, subject to change. Nevertheless, Polling the Nations these few websites should be fairly stable. http://wwvv.pollingthenations.com Also, check your book's website for even more Web Links. These websites, current at the time of this book's Web Survey Methodology http://vvww.websm.org/
Qualitative Field Research Introduction Institutional Ethnography Participatory Action Topics Appropriate Research to Field Research Conducting Qualitative Field Special Considerations Research in Qualitative Field Research Preparing for the Field The Various Roles Qualitative Interviewing of the Observer Focus Groups Relations to Subjects Recording Observations Some Qualitative Field Research Ethics Research Paradigms in Qualitative Field Research Naturalism Strengths and Weaknesses Ethnomethodology of Qualitative Field Research Grounded Theory Case Studies and Validity the Extended Case Method Reliability Sociology@Now\": Research Methods Use this online tool to help you make the grade on your next exam. After reading this chapter, go to the \"Online Study Resources\" at the end of the chapter for instructions on how to benefit from SocioiogyNolV Research Methods.
286 Chapter 10: Qualitative Field Research Introduction for example, a field researcher may note the \"pater- nalistic demeanor\" of leaders at a political rally or Several chapters ago, I suggested that you've been the \"defensive evasions\" of a public official at a pub- doing social research all your life, This idea should lic hearing without trying to express either the pa- become even clearer as we turn to what probably ternalism or the defensiveness as a numerical quan- seems like the most obvious method of making ob- tity or degree, Although field research can be used servations: qualitative field research, In a sense, we to collect quantitative data-for example, noting do field research whenever we observe or partici- the number of interactions of various specified pate in social behavior and try to understand it types within a field setting-ty-pically, field research whether in a college classroom, in a doctor's wait- 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. it's Such research is at once very old and very new a theory-generating activity as welL As a field re- in social science, stretching at least from the nine- searcher. you'll seldom approach your task with teenth century studies of preliterate societies, precisely defined hypotheses to be tested, More through firsthand examinations of urban commu- typically, you'll attempt to make sense out of an nity life in the \"Chicago School\" of the 1930s and ongoing process that cannot be predicted in ad- 1940s, to contemporary observations of chat-room vance-making initial observations, developing interactions on the web, Many of the techniques tentative general conclusions that suggest particular discussed in this chapter have been used by social types of further observations, making those obser- researchers for centuries, Within the social sci- vations and thereby revising your conclusions, and ences, anthropologists are especially associated so forth, In short, the alternation of induction and with this method and have contributed to its devel- deduction discussed in Part 1 of this book is per- opment as a scientific technique. Moreover, some- haps nowhere more evident and essential than in thing similar to this method is employed by many good field research, For expository purposes, how- people who might not strictly speaking, be re- ever. this chapter focuses primarily on some of the garded as social science researchers, Newspaper re- theoretical foundations of field research and on porters are one example; welfare department case techniques of data collection, Chapter 13 discusses 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 to 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 qualirativejleld research to distin- understanding of it. As such, this mode of obser- guish this type of observational method from meth- vation is especially, though not exclusively, appro- ods designed to produce data appropriate for quan- priate to research topics and social studies that ap- titative (statistical) analysis, Thus, surveys provide pear to defy simple quantification. Field researchers data from which to calculate the percentage unem- ployed in a population, mean incomes, and so forth, Field research more typically yields qualitative data: observations not easily reduced to numbers, Thus,
Topics Appropriate to Field Research 287 may recognize several nuances of attitude and be- 8. Settlelllellfs.: Small-scale \"societies\" such as vil- havior that might escape researchers using other lages, ghettos, and neighborhoods, as opposed methods. to large societies such as nations, which are difficult to study Field research is especially appropriate to the study of those attitudes and behaviors best under- 9, SociallVorlds. Ambiguous social entities 'with stood within their natural setting, as opposed to the vague boundaries and populations, such as somewhat artificial settings of experiments and \"the sports world\" and \"Wall Street\" surveys. For example, field research provides a su- perior method for studying the dynamics of reli- 10, Lifestyles or subculrures How large numbers of gious conversion at a revival meeting, just as a sta- people adjust to life in groups such as a \"rul- tistical analysis of membership rolls would be a ing class\" or an \"urban underclass\" better way of discovering whether men or women were more likely to convert, In all these social settings, field research can reveal things that would not otherwise be apparent. Finally, field research is well suited to the study Here's a concrete example. of social processes over time, Thus, the field re- searcher might be in a position to examine the One issue I'm particularly interested in (Babbie rumblings and final explosion of a riot as events ac- 1985) is the nature of responsibility for public tually occur rather than afterward in a reconstruc- matters: Who's responsible for making the things tion of the events. that we share work? Who's responsible for keeping public spaces-parks, malls, buildings, and so Other good places to apply field research meth- on-clean? Who's responsible for seeing that ods include campus demonstrations, courtroom broken street signs get fixed? Or, if a strong wind proceedings, labor negotiations, public hearings, or knocks over garbage cans and rolls them around similar events taking place within a relatively lim- the street, who's responsible for getting them out ited area and time, Several such observations must 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 Sodal Settings (1995: 101-13), ments in our society that assign responsibility for John and Lyn Lofland discuss several elements of these activities. Government custodians are respon- social life appropriate to field research: sible for keeping public places clean, Transportation department employees are responsible for the L PraCTices: VariOllS kinds of behavior, such as street signs, and perhaps the police are responsible talking or reading a book for the garbage cans rolling around on a ,vindy day, And when these responsibilities are not fulfilled, 2. Episodes.' A variety of events sllch as divorce, we tend to look for someone to blame. crime, and illness What fascinates me is the extent to which the 3, Encoullfers: 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 tak- 4, Roles.: The analysis of the pOSitions people oc- ing responsibility. It's my notion that it has become cupy and the behavior associated with those unacceptable for someone like you or me to take positions: occupations, family roles, ethnic personal responsibility for public matters that groups haven't been assigned to us. 5, Relationships; Behavior appropriate to pairs or Let me illustrate what I mean, If you were sets of roles: mother-son relationships, walking through a public park and you threw friendships, and the like down a bunch of trash, you'd discover that your action was unacceptable to those around you. 6. Groups' Small groups, such as friendship People would glare at you, grumble to each other; cliques, athletic teams, and work groups 7, Organizations' Formal organizations, such as hospitals or schools
288 Chapter 10: Qualitative Field Research perhaps someone would say something to you damaged and long-unused newspaper box from about it Whatever the form, you'd be subjected to the bus stop, where it had been a problem for definite, negative sanctions for littering. Now here's months, when the police arrived, having been the irony. If you were walking through that same summoned by a neighbor. Another student de- park, carne across a bunch of trash that someone cided to clean out a clogged storm drain on his else had dropped, and cleaned it up, it's likely that street and found himself being yelled at by a neigh- your action would also be unacceptable to those bor who insisted that the mess should be left for around you. You'd probably face negative sanctions the street cleaners. Everyone who picked up litter for cleaning it up. was sneered at laughed at and generally put down. One young man was picking up litter scat- When I first began discussing this pattern with tered around a trash can when a passerby sneered, students, most felt the notion was absurd. Al- \"Clumsy!\" It became clear to us that there are only though we would be negatively sanctioned for lit- three acceptable explanations for picking up litter tering, cleaning up a public place would obviously in a public place: bring positive sanctions: People would be pleased with us for doing it Certainly, all my students said L You did it and got caught-somebody forced they would be pleased if someone cleaned up a you to clean up your mess. public place. It seemed likely that everyone else would be pleased, too, if we asked them how they 2. You did it and felt guilty. would react to someone's cleaning up litter in a public place or othervvise taking personal responsi- 3. You're stealing litter. bility for fixing some social problem. In the normal course of things, it's simply not To settle the issue, I suggested that my students acceptable for people to take responsibility for start fixing the public problems they came across public things. in the course of their everyday activities. As they did so, I asked them to note the answers to two Clearly, we could not have discovered the na- questions: ture and strength of agreements about taking personal responsibility for public things except L How did they feel while they were fixing a through field research. Social norms suggest that public problem they had not been assigned taking responsibility is a good thing, sometimes re- responsibility for? ferred to as good citizenship. Asking people what they thought about taking responsibility would 2. How did others around them react? have produced a solid consensus that it was good. Only going out into life, doing it and watching My students picked up litter. fixed street signs, what happened gave us an accurate picture. put knocked-over traffic cones back in place, cleaned and decorated communal lounges in their As an interesting footnote to this story, my stu- dorms, trimmed trees that blocked visibility at in- dents and I found that whenever people could get tersections, repaired public playground equipment past their initial reactions and discover that the stu- cleaned public restrooms, and took care of a hun- dents were simply taking responsibility for fixing dred other public problems that weren't \"their things for the sake of having them work, the pass- responsibility. \" ersby tended to assist. Although there are some very strong agreements making it \"unsafe\" to take Most reported feeling very uncomfortable do- responsibility for public things, the willingness of ing whatever they did. They felt foolish, goody- one person to rise above those agreements seemed goody, conspicuous, and all the other feelings that to make it safe for others to do so, and they did. keep us from performing these activities normally. In almost every case, their personal feelings of Field research is not to be confused with jour- discomfort were increased by the reactions of nalism. Social scientists and journalists may use those around them. One student was removing a similar techniques, but they have quite a different relationship to data. For instance, individual
SpeCial Considerations in Qualitative Field Research 289 interviewing is a common technique in journal- The Various Roles of the Observer ism and sociology; nevertheless, sociologists are not simply concerned vvith reporting about a In field research, observers can play any of several subject's attitude, belief. or experience. A soci- roles, including participating in what they want to ologist's goal is to treat an interview as data that observe (this was the situation of the students who need to be analyzed to understand social life more fixed public things). In this chapter. I've used the generally. term field research rather than the frequently used term participant observariol1, because field researchers Two important aspects of qualitative research need not always participate in what they're study- need to be stressed. First. a wide range of studies ing, though they usually will study it directly at the fall under the umbrella \"qualitative field research.\" 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 ap- proaches to basic questions such as \"What is data?\" The researcher may plan a role that entails \"How should we collect data?\" and \"How should varying degrees of \"participantness\"-that is, we analyze data?\" Second, we should remember the degree of actual participation in daily life. that the questions we want to answer in our re- At one extreme is the full participant who search determine the types of methods we need to goes about ordinary life in a role or set of use. A question such as \"How do women constmct roles constructed in the setting. At the other their everyday lives in order to perform their roles extreme is the complete observer, who engages as mothers, partners, and breadwinners?\" could be not at all in social interaction and may even addressed by in-depth interviews and direct obser- shun involvement in the world being studied. vations. The assessment of advertising campaigns And, of course, all possible complementary might profit from focus group discussions. In most mixes along the continuum are available to cases, we'll find that researchers have alternate the researcher. methods to choose from. (1995: 60) In summary, then, field research offers the ad- vantage of probing social life in its natural habitat. The complete participant. in this sense, may be Although some things can be studied adequately a genuine participant in what he or she is studying through questionnaires or in the laboratory, others (for example, a participant in a campus demonstra- cannoL And direct observation in the field lets re- tion) or may pretend to be a genuine participant. In searchers observe subtle communications and other any event. whenever you act as the complete par- events that might not be anticipated or measured ticipant you must let people see you only as a par- othenvise. ticipant. not as a researcher. For instance, if you're using this technique to study a group made up of Special Considerations uneducated and inarticulate people, it would not in Qualitative Field Research be appropriate for you to talk and act like a univer- sity professor or student. There are specific things to take into account in every research method, and qualitative field re- This type of research introduces an ethical is- search is no exception. When you use field re- sue, one on which social researchers themselves are search methods, you're confronted with decisions divided. Is it ethical to deceive the people you're about the role you'll playas an obsen'er and studying in the hope that they will confide in you your relations with the people you're observing. as they will not confide in an identified researcher? Let's examine some of the issues involved in these Do the potential benefits to be gained from the re- decisions. search offset such considerations? Although many professional associations have addressed this issue, the norms to be followed remain somewhat am- biguous when applied to specific situations.
290 Chapter 10: Qualitative Field Research Related to this ethical consideration is a sci- Because of these several considerations, ethical entific one. No researcher deceives his or her sub- and scientific, the field researcher frequently jects solely for the purpose of deception. Rather, it's chooses a different role from that of complete par- done in the belief that the data will be more valid ticipant. You could participate fully with the group and reliable, that the subjects vviII be more natural under study but make it clear that you were also and honest if they do not know the researcher is undertaking research. As a member of the volley- doing a research project. If the people being studied ball team, for example, you might use your position know they're being studied, they might modify to launch a study in the sociology of sports, letting their behavior in a variety of ways. This is known your teammates know what you're doing. There as the problem of reactivity. are dangers in this role also, however. The people being studied may shift much of their attention to First they might expel the researcher. Second, the research project rather than focus on the natu- they might modify their speech and behavior to ap- ral social process, making the process being ob- pear more \"respectable\" than would otherwise be served no longer typical. Or, conversely, you your- the case. Third, the social process itself might be self may come to identify too much with the radically changed. Students making plans to burn interests and viewpoints of the participants. You down the university administration building, for may begin to \"go native\" and lose much of your example, might give up the plan altogether once scientific detachment. they learn that one of their group is a social scien- tist conducting a research project. At the other extreme, the complete observer studies a social process without becoming a part of On the other side of the coin, if you're a com- it in any way. Quite possibly, because of the re- plete participant, you may affect ,vhat you're study- searcher'S unobtrusiveness, the subjects of study ing. Suppose, for example, that you're asked for might not realize they're being studied. Sitting at a your ideas about what the group should do next. bus stop to observe jaywalking at a nearby intersec- No matter what you say, you vvill affect the process tion is one example. Although the complete ob- in some fashion. If the group follows your sugges- server is less likely to affect what's being studied tion, your influence on the process is obvious. If and less likely to \"go native\" than the complete the group decides not to follow your suggestion, participant, she or he is also less likely to develop a the process whereby the suggestion is rejected may full appreciation of what's being studied. Observa- affect what happens next. Finally, if you indicate tions may be more sketchy and transitory. that you just don't know what should be done next, you may be adding to a general feeling of un- Fred Davis (1973) characterizes the extreme certainty and indecisiveness in the group. roles that observers might playas \"the Martian\" and \"the Convert.\" The latter involves delving more Ultimately, anything the participant-observer and more deeply into the phenomenon under does or does not do ,vill have some effect on study, running the risk of \"going native.\" We'll what's being observed; it's simply inevitable. More examine this risk further in the next section. seriously, there is no complete protection against this effect. though sensitivity to the issue may pro- To appreciate the \"Martian\" approach, imagine vide a partial protection. (This influence, called the that you were sent to observe some newfound Hawthorne effect, was discussed more fully in life on Mars. Probably you would feel yourself in- Chapter 8.) escapably separate from the Martians. Some social scientists adopt this degree of separation when reactivity The problem that the subjects of social observing cultures or social classes different from research may react to the fact of being studied, thus their own. altering their behavior from what it would have been normally Marshall and Rossman (1995: 60-61) also note that the researcher can vary the amount of time spent in the setting being observed: You can be a full-tin1e presence on the scene or just show up
Special Considerations in Qualitative Field Research 291 now and then. Moreover, you can focus your at- \"insider understanding.\" Ultimately, you will not be tention on a lin1ited aspect of the social setting or able to fully understand the thoughts and actions seek to observe all of it-framing an appropriate of the cult members unless you can adopt their role to match your aims. points of view as true-at least temporarily. To fully appreciate the phenomenon you've set out Different situations ultimately require different to study, you need to believe that Jesus is coming roles for the researcher. Unfortunately, there are no Thursday night. clear guidelines for making this choice-you must rely on your understanding of the situation and Adopting an alien point of view is an uncom- your own good judgment. In making your decision, fortable prospect for most people. It can be hard however, you must be guided by both methodolog- enough merely to learn about views that seem ical and ethical considerations. Because these often strange to you; you may sometimes find it hard just conflict, your decision will frequently be difficult, to tolerate certain views. But to take them on as and you may find sometin1es that your role limits your own can be ten times worse. Robert Bellah your study. (1970, 1974) has offered the term symbolic realism to indicate the need for social researchers to treat Relations to Subjects the beliefs they study as worthy of respect rather than as objects of ridicule. If you seriously entertain Having introduced the different roles field re- this prospect, you may appreciate why William searchers might play in connection \\vith their ob- Shaffir and Robert Stebbins (1991: 1) conclude servations, we now focus more specifically on how that \"fieldwork must certainly rank with the more researchers may relate to the subjects of their study disagreeable activities that humanity has fashioned and to the subjects' points of view. for itself.\" We've already noted the possibility of pretend- There is, of course, a danger in adopting the ing to occupy social statuses we don't really occupy. points of view of the people you're studying. When Consider now how you would think and feel in you abandon your objectivity in favor of adopting such a situation. such views, you lose the possibility of seeing and understanding the phenomenon within frames of Suppose you've decided to study a religious cult reference unavailable to your subjects. On the one that has enrolled many people in your neighbor- hand, accepting the belief that the world will end hood. You might study the group by joining it or Thursday night allows you to appreciate aspects of pretending to join it. Take a moment to ask yourself that belief available only to believers; stepping out- what the difference is between \"really\" joining side that view, however, makes it possible for you and \"pretending\" to join. The main difference is to consider some reasons why people might adopt whether or not you actually take on the beliefs, at- such a view. You may discover that some did so as a titudes, and other points of view shared by the consequence of personal trauma (such as unem- \"real\" members. If the cult members believe that ployment or divorce), whereas others were brought Jesus will come next Thursday night to destroy the into the fold through their participation in particu- world and save the members of the cult, do you be- lar social networks (for example, their whole bowl- lieve it or do you simply pretend to believe it? ing team joined the cult). Notice that the cult members might disagree with those \"objective\" ex- Traditionally, social scientists have tended to planations, and you might not come up with them emphasize the importance of \"objectivity\" in such to the extent that you had operated legitimately matters. In this example, that injunction would within the group's views. be to avoid getting swept up in the beliefs of the group. Without denying the advantages associated Anthropologists sometimes use the term emic with such objectivity, social scientists today also perspective in reference to taking on the point of recognize the benefits gained by immersing them- view of those being studied. In contrast, the eric selves in the points of view they're studying, perspective maintains a distance from the native what Lofland and Lofland (1995: 61) refer to as
292 Chapter 10: Qualitative Field Research point of view in the interest of achieving more you become deeply involved in the lives of the objectivity. people you're studying, you're likely to be moved by their personal problems and crises. Imagine, The apparent dilemma here is that both of for example, that one of the cult members becomes these postures offer important advantages but also ill and needs a ride to the hospitaL Should you seem mutually exclusive. In fact, it is possible to as- provide transportation? Sure. Suppose someone sume both postures. Sometimes you can simply wants to borrow money to buy a stereo. Should shift vievvpoints at will. When appropriate, you can you loan it? Probably not. Suppose they need the fully assume the beliefs of the cult; later, you can money for food? step outside those beliefs (more accurately, you can step inside the viewpoints associated with so- There are no black-and-white rules for resolv- cial science). As you become more adept at this ing situations such as these, but you should realize kind of research, you may come to hold contradic- that you'll need to deal with them regardless of tory viewpoints simultaneously, rather than whether or not you reveal that you're a researcher. switching back and forth. Such problems do not tend to arise in other types of research-surveys and experiments, for example- During my study of trance channeling-in but they are part and parcel of field research. which people allow spirits to occupy their bodies and speak through them-I found I could partici- This discussion of the field researcl1er's relations pate fully in channeling sessions without becoming to subjects flies in the face of the conventional view alienated from conventional social science. Rather of \"scientific objectivity.\" Before concluding this than \"believing\" in the reality of channeling, I section, let's take the issue one step further. found it possible to suspend beliefs in that realm: neither believing it to be genuine (like most of the In the conventional view of science, there are other participants) nor disbelieving it (like most sci- implicit differences of power and status separating entists). Put differently, I was open to either possibil- the researcher from the subjects of research. When ity. Notice how this differs from our normal need to we discussed experimental designs in Chapter 8, for \"know\" whether such things are legitimate or not. example, it was obvious who was in charge: the ex- perimenter, who organized things and told the sub- Social researchers often refer to the concerns jects what to do. Often the experimenter was the just discussed as a matter of reflexivity, in the sense only person who even knew what the research was of things acting on themselves. Thus, your own really about. Something sin1ilar might be said about characteristics can affect what you see and how survey research. The person running the survey you interpret it. The issue is broader than that, designs the questions, decides who will be selected however, and applies to the subjects as well as to for questioning, and analyzes the data collected. the researcher. Imagine yourself interviewing a homeless person (1) on the street, (2) in a home- SOciologists often look at these sorts of relation- less shelter, or (3) in a social welfare office. The re- ships as power or status relationships. In experi- search setting could affect the person's responses. In mental and survey designs, the researcher clearly other words, you might get different results de- has more power and a higher status than the pending on where you conducted the interview. people being studied do. The researchers have a Moreover, you might act differently as a researcher special knowledge that the subjects don't enjoy. in those different settings. If you reflect on this is- They're not so crude as to say they're superior to sue, you'll be able to identify other aspects of the their subjects, but there is a sense in which that's research encounter that complicate the task of implicitly assumed. (Notice that there is a similar, \"simply observing what's so.\" in1plicit assumption about the writers and readers of textbooks.) The problem we've just been discussing could be seen as psychologicaL occurring mostly inside In field research, such assumptions can be the researchers' or subjects' heads. There is a corre- problematic. When the early European anthropol- sponding problem at a social leveL however. When ogists set out to study what were originally called \"primitive\" societies, there was no question but that
Some Qualitative Field Research Paradigms 293 the anthropologists knew best. Whereas the natives reported by the researcher as it \"really is\" (Gubrium \"believed\" in witchcraft, for example, the anthro- and Holstein 1997). This tradition started in the pologists \"knew\" it wasn't really true. Whereas the 1930s and 1940s at the University of Chicago's natives said some of their rituals would appease sociology department, whose faculty and students the gods, the anthropologists explained that the fanned out across the city to observe and under- \"real\" functions of these rituals were the creation stand local neighborhoods and communities. The of social identity, the establishment of group researchers of that era and their research approach solidarity, and so on. are now often referred to as the Chicago SchooL The more social researchers have gone into the One of the earliest and best-known studies that field to study their fellow humans face-to-face, illustrates this research tradition is William Foote however, the more they have become conscious of Whyte's ethnography of Cornerville, an Italian these implicit assumptions about researcher superi- American neighborhood, in his book Street Comer ority, and the more they have considered alterna- Society (1943). An ethnography is a study that fo- tives. As we turn now to the various paradigms of cuses on detailed and accurate description rather field research, we'll see some of the ways in which than explanation. Like other naturalists, Whyte be- that ongoing concern has worked itself out. lieved that in order to learn fully about social life on the streets, he needed to become more of an in- Some Qualitative Field sider. He made contact with \"Doc,\" his key infor- Research Paradigms mant, who appeared to be one of the street-gang leaders. Doc let Whyte enter his world, and Whyte Although I've described field research as simply go- got to participate in the activities of the people of ing where the action is and observing it, there are Comerville. His study offered something that sur- actually many different approaches to this research veys could not: a richly detailed picture of life method. This section examines several field re- among the Italian immigrants of Cornerville. search paradigms: naturalism, ethnomethodology, grounded theory, case studies and the extended An in1portant feature of Whyte'S study is that case method, institutional etlmography, and partic- he reported the reality of the people of Cornerville ipatory action research. Although this survey won't on their terms. The naturalist approach is based on exhaust the variations on the method, it should telling \"their\" stories the way they \"really are,\" not give you a broad appreciation of the possibilities. the way the ethnographer understands \"them.\" The narratives collected by Whyte are taken at face value It's important to recognize that there are no as the social \"truth\" of the Cornerville residents. specific methods attached to these paradigms. You could do ethnomethodology or institutional Forty-five years later, David Snow and Leon ethnography by analyzing court hearings or con- Anderson (1987) conducted exploratory field ducting group interviews, for example. The impor- research into the lives of homeless people in tant distinctions of this section are epistemological, Austin, Texas. Their major task was to understand having to do with what data mean, regardless of how the homeless construct and negotiate their how they were collected. identity while knowing that the society they live in attaches a stigma to homelessness. Snow and Anderson believed that to achieve this goaL the Naturalism naturalism An approach to field research based on the assumption that an objective social reality exists Naturalism is an old tradition in qualitative re- and can be observed and reported accurately. search. The earliest field researchers operated on the positivist assumption that social reality was ethnography A report on social life that focuses \"out there,\" ready to be naturally observed and on detailed and accurate description rather than eXlJlanation.
294 Chapter 10: Qualitative Field Research collection of data had to arise naturallyo Like Mitchell describes a variety of survivalist indi- Whyte in Sn'eer Corner Sociay, they found some key viduals and groups, seeking to understand their rea- informants whom they followed in their everyday soning, their plans, and the threat they may pose journeys, such as at their day-labor pickup sites or for the rest of uSo He found the survivalists disillu- under bridgeso Snow and Anderson chose to mem- sioned \"lith and uncertain about the future of U.S. orize the conversations they participated in or the society, but most were more interested in creating \"talks\" that homeless people had with each other. alternative lives and cultures for themselves than in At the end of the day, the two researchers debriefed blowing anyone up. That's not to suggest none of and wrote detailed field notes about all the \"talks\" the survivalists pose a threat, but Mitchell's exami- they encountered. They also taped in-depth inter- nation moves beyond the McVeighs, Koreshes, and views with their key informants. Weavers to draw a broader picture of the whole phenomenon. Snow and Anderson reported \"hanging out\" with homeless people over the course of 12 months Whereas this chapter aims at introducing you for a total of 405 hours in 24 different settings. Out to some of the different approaches available to you of these rich data, they identified three related pat- in qualitative field research, please realize that this terns in homeless people's conversations. First, the discussion of ethnography merely sketches some of homeless showed an attempt to \"distance\" them- the many avenues social researchers have estab- selves from other homeless people, from the low- lished. If you're interested in this general approach, status job they currently had, or from the Salvation you might want to explore the idea of vimwl Army they depended on. Second, they \"embraced\" ethnography, which uses ethnographic techniques their street-life identity, their group membership for inquiry into cyberspace. Or, in a different direc- or a certain belief about why they are homeless. tion, azaoerhllography intentionally assumes a per- Third, they told \"fictive stories\" that always con- sonal stance, breaking with the general proscription trasted with their everyday life. For example, they against the researcher getting involved at that would often say that they were making much more level. You can learn more about these variants on money than they really were, or even that they ethnography by searching the web or your campus were \"going to be rich.\" libraryo A later section of this chapter will examine illstiwriollal ethnography, which links individuals and Richard Mitchell (2002) offers another, timely organizations. illustration of the power of ethnographic reporting. Recent U.S. history has raised the specter of vio- Ethnomethod%gy lence from secretive survivalist groups, dramatized by the 1992 siege at Ruby Ridge, Idaho, which left Ethnomethodology, which I introduced as a re- the wife and son of the white supremacist Randy search paradigm in Chapter 2, is a unique approach Weaver dead; the 1993 shootout with David to qualitative field research. It has its roots in the Koresh and his Branch Davidians in Waco, Texas; philosophical tradition of phenomenology, which and Timothy McVeigh's 1995 bombing, which can explain why ethnomethodologists are skeptical left 168 dead under the rubble of the nine-story about the way people report their experience of re- Murrah Federal Building in Oklahoma City. ality (Gubrium and Holstein 1997) . Alfred Schutz (1967, 1970), who introduced phenomenology, ethnornethodo!ogy An approach to the study of argued that reality was socially constructed rather social life that focuses on the discovery of implicit, than being \"out there\" for us to observe. People usually unspoken assumptions and agreements; this describe their world not \"as it is\" but \"as they make method often involves the intentional breaking of sense of it.\" Thus, phenomenologists would argue agreements as a way of revealing their existenceo that Whyte's street-corner men were describing their gang life as it made sense to them. Their
Some Qualitative Field Research Paradigms \" 295 reports, however, would not tell us how and why it (E) How I am in regard of what? My health, made sense to them. For this reason, researchers my finances, my school work, my peace of cannot rely on their subjects' stories to depict social mind, my 0 •• ? realities accurately. (S) (Red in the face and suddenly out of Whereas traclitional ethnographers believed in control.) Look I was just trying to be polite. immersing themselves in a particular culture and Frankly, I don't give a damn how you are. reporting their informants' stories as if they repre- sented reality, phenomenologists see a need to By setting aside or \"bracketing\" their expecta- \"make sense\" out of the informants' perceptions of tions from these everyday conversations, the exper- the world. Followirlg in this traclition, some field re- imenters made visible the subtleties of mundane searchers have felt the need to devise techniques interactions. For example, although \"How are that reveal how people make sense of their every- you?\" has many possible meanings, none of us day world. As we saw in Chapter 2, the sociologist have any trouble knowing what it means in casual Harold Garfinkel suggested that researchers break interactions, as the unsuspecting subject revealed the rules so that people's taken-for-granted expecta- in his final comment. tions would become apparent. This is the technique that Garfinkel called \"ethnomethodology.\" Ethnomethodologists, then, are not simply in- terested in subjects' perceptions of the worldo In Garfinkel became known for engaging his stu- these cases, we could imagine that the subjects may dents to perform a series of what he called \"breach- have thought that the experimenters were rude, ing experiments\" designed to break away from the stupid, or arrogant. The conversation itself, not the ordinary (Heritage 1984). For instance, Garfinkel informants, is the object of ethnomethodological (1967) asked his students to do a \"conversation studies. In general, in ethnomethodology the focus clarification experiment.\" Students were told to en- centers on the \"underlying patterns\" of interactions gage in an orclinary conversation with an acquain- that regulate our everyday lives. tance or a friend and to ask for clarification about any of this person's statements. Through this tech- Ethnomethodologists believe that researchers nique, they uncovered elements of conversation who use a naturalistic analysis \"[lose] the ability that are normally taken for granted. Here are two to analyze the commonsense world and its culture examples of what Garfinkel's students reported if [they use] analytical tools and insights that are (1967:42): themselves part of the world or culture being studied\" (Gubrium and Holstein 1997: 43). Case 1 D. L Wieder provides an excellent example of how The subject was telling the experimenter, different a naturalistic approach is from an ethno- methodological approach (Gubrium and Holstein a member of the subject's car pool, about 1997). In his study, Language and Social Reality: The having had a flat tire while going to work the Case oj Telling the Convict Code (1988), Wieder started previous day. to approach convicts in a halfway house in a tradi· tional ethnographic style: He was going to become I had a flat tire. an insider by befriencling the inmates and by con- (E) What do you mean, you had a flat tire? ducting participant observations, He took careful She appeared momentarily stunned. Then notes and recorded interactions among irunates she answered in a hostile way: \"What do you and between inmates and staff. His first concern mean, 'What do you mean?' A flat tire is a was to describe the life of the convicts of the flat tire. That is what I meant. Nothing speciaL halfway house the way it \"really was\" for them. What a crazy question.\" Wieder'S observations allowed him to report on a \"convict code\" that he thought was the source of Case 6 the deviant behavior expressed by the inmates The victim waved his hand cheerily. (S) How are you?
296 Chapter 10: Qualitative Field Research toward the staff. This code, which consisted of a from an analysis of the patterns, themes, and com- series of rules such as \"Don't kiss ass,\" \"Don't mon categories discovered in observational data. snitch,\" and \"Don't trust the staff,\" was followed by The first major presentation of this method can be the inmates who interfered with the staff members' found in Glaser and Strauss's book, The DiscovelY of attempts to help them make the transition between Grounded Theory (1967). Grounded theory can be prison and the community. described as an approach that attempts to combine a naturalist approach with a positivist concern for a It became obvious to Wieder that the code was \"systematic set of procedures\" in doing qualitative more than an explanation for the convicts' deviant research. behavior; it was a \"method of moral persuasion and justification\" (Wieder 1988: 175). At this point Strauss and Juliet Corbin (1998: 43-46) he changed his naturalistic approach to an ethno- have suggested that grounded theory allows the methodological one. Whereas naturalistic field re- researcher to be scientific and creative at the same searchers aim to understand social life as the par- time, as long as the researcher follows these ticipants understand it, ethnomethodologists are guidelines: more intent on identifying the methods through which understanding occurs. In the case of the o Think Comparatively: The authors suggest convict code, Wieder came to see that convicts that it is essential to compare numerous inci- used the code to make sense of their own interac- dents as a way of avoiding the biases that can tions with other convicts and with the staff. The arise from interpretations of initial observations. ethnography of the halfway house thus shifted to an ethnography of the code. For instance, the con- o Obtain Multiple Viewpoints: In part this victs would say, \"You know I won't snitch,\" refer- refers to the different points of view of partici- ring to the code as a way to justify their refusal pants in the events under study, but Strauss to answer Wieder's question (p. 168). According and Corbin suggest that different observational to Wieder, the code \"operated as a device for techniques may also provide a variety of stopping or changing the topic of conversation\" viewpoints. (p. 175). Even the staff would refer to the code to justify their reluctance to help the convicts. Al- o Periodically Step Back: As data accumulate, though the code was something that constrained you'll begin to frame interpretations about behavior, it also functioned as a tool for the control what is going on, and it's important to keep of interactions. checking your data against those interpreta- tions. As Strauss and Corbin (1998: 45) say, Grounded Theory \"The data themselves do not lie.\" Grounded theory originated from the collaboration o Maintain an Attitude of Skeptism: As you of Barney Glaser and Anselm Strauss, sociologists begin to interpret the data, you should regard who brought together two main traditions of re- all those interpretations as provisionaL using search, positivism and interactionism. Essentially, new observations to test those interpretations, grounded theory is the attempt to derive theories not just confirm them. grounded theory An inductive approach to the o Follow the Research Procedures: Grounded study of social life that attempts to generate a theory theory allows for flexibility in data collection as from the constant comparing of unfolding observa- theories evolve, but Strauss and Corbin (1998: tions. This is very different from hypothesis testing, 46) stress that three techniques are essential: in which theory is used to generate hypotheses to be \"making comparisons, asking questions, and tested through observations. sampling.\" 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
Some Qualitative Field Research Paradigms 297 of data, grounded theorists are quite open to the Shopping Romania use of qualitative studies in conjunction with quan- titative ones. Here are two examples of the in1ple- Much has been written about large-scale changes mentation of this approach. caused by the shift from socialism to capitalism in the former USSR and its Eastern European allies. Studying Academic Change Patrick Jobes and his colleagues (1997) wanted to learn about the transition on a smaller scale among Clifton Conrad's (1978) study of academic change average Romanians. They focused on the task of in universities is an early example of the grounded shopping. theory approach. Conrad hoped to uncover the major sources of changes in academic curricula and Noting that shopping is normally thought at the same time understand the process of change. of as a routine, relatively rational activity, the re- Using the grounded theory idea of theoretical sam- searchers suggested that it could become a social pling-whereby groups or institutions are selected problem in a radically changing economy. They on the basis of their theoretical relevance-Conrad used the Grounded Theory Method to examine chose four universities for the purpose of his study. Romanian shopping as a social problem, looking In two, the main vehicle of change was the formal for the ways in which ordinary people solved the curriculum committee; in the other two, the ve- problem. hicle of change was an ad hoc group. Their first task was to learn something about Conrad explained, step by step, the advantage how Romanians perceived and understood the task of using the grounded theory approach in building of shopping. The researchers-participants in a his theory of academic change. He described the social problems class-began by interviewing 40 process of systematically coding data in order to shoppers and asking whether they had eXlJerienced create categories that must \"emerge\" from the data problems in connection with shopping and what and then assessing the fitness of these categories actions they had taken to cope with those problems. with each other. Going continuously from data to theory and theory to data allowed him to reassess Once the initial interviews were completed, the validity of his initial conclusions about aca- the researchers reviewed their data, looking for cat- demic change. egories of responses- the shoppers' most common problems and solutions. One of the most common For instance, it first seemed that academic problems was a lack of money. This led to the re- change was mainly caused by an administrator searchers' first working hypothesis: The \"socio- who was pushing for it. By reexamining the data economic position of shoppers would be associated and looking for more plausible explanations, \\,'lith how they perceived problems and sought so- Conrad found the pressure of interest groups a lutions\" (1997: 133). This and other hypotheses more convincing source of change. The emergence helped the researchers focus their attention on of these interest groups actually allowed the ad- more-specific variables in subsequent interviewing. ministrator to become an agent of change. As they continued, they also sought to inter- Assessing how data from each of the two types view other types of shoppers. When they inter- of universities fit with the other helped refine the viewed students, for example, they discovered that theory building. Conrad concluded that changes in different types of shoppers were concerned with university curricula are based on the following pro- different kinds of goods, which in turn affected the cess: Conflict and interest groups emerge because problems faced and the solutions tried. of internal and external social structural forces; they push for administrative intervention and rec- As the researchers developed additional hy- ommendation to make changes in the current aca- potheses in response to the continued interviewing, demic program; these changes are then made by they also began to develop a more or less standard- the most powerful decision-making body. ized set of questions to ask shoppers. Initially, all the questions were open-ended, but they eventu- ally developed closed-ended items as welL
298 Chapter 10: Qualitative Field Research This study illustrates the key, inductive prin- basis for the development of more generaL nomo- ciples of grounded theory: Data are collected in the thetic theories. absence of hypotheses. The initial data are used to determine the key variables as perceived by those Michael Burawoy and his colleagues (1991) being studied, and hypotheses about relationships have suggested a somewhat different relationship among the variables are similarly derived from the between case studies and theory. For them, the data collected. Continuing data collection yields extended case method has the purpose of dis- refined understanding and, in turn, sharpens the covering flaws in, and then modifying, existing so- focus of data collection itself. cial theories. This approach differs importantly from some of the others already discussed. Case Studies and the Extended Case Method Whereas the grounded theorists seek to enter the field \\vith no preconceptions about what Social researchers often speak of case studies, they'll find, Burawoy suggests just the opposite: to which focus attention on one or a few instances try \"to layout as coherently as possible what we of some social phenomenon, such as a village, a expect to find in our site before entry\" (Burawoy family, or a juvenile gang. As Charles Ragin and et aL 1991: 9). Burawoy sees the extended case Howard Becker (1992) point out, there is little con- method as a way to rebuild or improve theory in- sensus on what may constitute a \"case,\" and the stead of approving or rejecting it. Thus, he looks term is used broadly. The case being studied, for for all the ways in which observations conflict example, might be a period of time rather than a \\vith existing theories and what he calls \"theoreti- particular group of people. The limitation of atten- cal gaps and silences\" (1991: 10). This orientation tion to a particular instance of something is the to field research implies that kno\\ving the liter- essential characteristic of the case study. ature beforehand is actually a must for Burawoy and his colleagues, whereas grounded theorists The chief purpose of case studies may be would worry that knowing what others have descriptive, as when an anthropologist describes concluded might bias their observations the culture of a preliterate tribe. Or the in-depth and theories. study of a particular case can yield explanatory insights, as when the community researchers Rob- To illustrate the extended case method, I'll use ert and Helen Lynd (1929, 1937) and W. Lloyd two examples of studies by Burawoy's students. Warner (1949) sought to understand the struc- ture and process of social stratification in small- Teacher-Student Negotiations town USA. Leslie Hurst (1991) set out to study the patterns of Case study researchers may seek only an interaction between teachers and students of a jun- idiographic understanding of the particular case ior high school. She went into the field armed \\vith under examination, or-as we've seen \\vith existing, contradictory theories about the \"official\" grounded theory-case studies can form the functions of the schooL Some theories suggested that the purpose of schools was to promote social case study The in-depth examination of a single mobility, whereas others suggested that schools instance of some social phenomenon, such as a vil- mainly reproduced the status quo in the form of a lage, a family, or a juvenile gang. stratified division of labor. The official roles as- extended case method A technique developed by signed to teachers and students could be inter- Michael Burawoy in which case study observations preted in terms of either view. are used to discover flaws in and to improve existing social theories. Hurst was struck, however, by the contrast be- tween these theories and the types of interactions she observed in the classroom. In her own experi- ences as a student, teachers had total rights over the minds, bodies, and souls of their pupik She
Some Qualitative Field Research Paradigms 299 observed something quite different at a school in purpose on the school experience, missing its a lower-middle-class neighborhood in Berkeley, day-to-day reality. California-Emerald Junior High SchooL where she volunteered as a tutor. She had access to the (1991: 186) classroom of Mr. Henry (an eighth-grade English teacher) as well as other teachers' classrooms, the In summary, what emerges from Hurst's study lunchroom, and English Department meetings. She is an attempt to improve the traditional sociological wrote field notes based on the negotiations be- understanding of education by adding the idea that tween students and teachers. She explained the na- classroom, schooL and family have separate func- ture of the student-teacher negotiations she \\vit- tions, which in turn can explain the emergence of nessed by focusing on the separation of functions \"negotiated order\" in the classroom. among the schooL the teacher, and the family. The Fight against AIDS In Hurst's observation, the school fulfilled the function of controlling its students' \"bodies\"-for Katherine Fox (1991) set out to study an agency example, by regulating their general movements whose goal was to fight the AIDS epidemic by and activities \\vithin the school. The students' bringing condoms and bleach for cleaning needles \"minds\" were to be shaped by the teacher, whereas to intravenous drug users. It's a good example of students' families were held responsible for their finding the limitations of well-used models of theo- \"souls\"; that is, families were expected to socialize retical explanation in the realm of understanding students regarding personal values, attitudes, deviance-specifically, the \"treatment model\" that sense of property, and sense of decorum. When predicted that drug users would come to the clinic students don't come to school \\vith these values in and ask for treatment. Fox's interactions \\vith out- hand, the teadler, according to Hurst, \"must first reach workers-most of whom were part of the negotiate \\vith the students some compromise on community of drug addicts or former prostitutes- how the students will conduct themselves and contradicted that model. on what \\viII be considered classroom decorum\" (1991: 185). To begin, it was necessary to understand the drug users' subculture and use that knowledge to Hurst explained that the constant bargaining devise more realistic policies and programs. The between teachers and students is an expression of target users had to be convinced, for example, that the separation between \"the body,\" which is the tlle program workers could be trusted, that they school's concern, and \"the soul\" as family domain. were really interested only in providing bleach and The teachers, who had limited sanctioning power condoms. The target users needed to be sure they to control their students' minds in the classroom, were not going to be arrested. were using forms of negotiations \\vith students so that they could \"control ... the student's body and Fox's field research didn't stop \\vith an exami- sense of property\" (1991: 185), or as Hurst defines nation of the drug users. She also studied the it, \"babysit\" the student's body and soul. agency workers, discovering that the outreach pro- gram meant different things to the research direc- Hurst says she differs from the traditional socio- tors and the outreach workers. Some of the volun- logical perspectives as follows: teers who were actually providing the bleach and condoms were frustrated about the minor changes I do not approach schools \\vith a futuristic eye. they felt they could make. Many thought the pro- I do not see the school in terms of training, so- gram was just a bandage on the AIDS and drug- cializing, or slotting people into future hierar- abuse problems. Some resented having to take chies. To approach schools in this manner is to field notes. Directors, on the other hand, needed miss the negotiated, chaotic aspects of the reports and field notes so that they could validate classroom and educational experience. A futur- their research in the eyes of the federal and state ist perspective tends to impose an order and agencies that financed the project. Fox's study showed how the AIDS research project developed
300 Chapter 10: Qualitative Field Research the bureaucratic inertia typical of established or- researcher can reveal aspects of society that would ganizations: Its goal became that of sustaining itself. have been missed by an inquiry that began with the official purposes of institutions. Both of these studies illustrate how the ex- tended case method can operate. The researcher This approach links the \"microlevel\" of every- enters the field with full knowledge of existing the- day personal experiences with the \"macrolevel\" of ories but aims to uncover contradictions that re- institutions. As M. L Campbell puts it, quire the modification of those theories. Institutional etlmography, like other forms of One criticism of the case study method is the ethnography, relies on interviewing, observa- limited generalizability of what may be observed in tions and documents as data. Institutional a single instance of some phenomenon. This risk is ethnography departs from other ethnographic reduced, however, when more than one case is approaches by treating those data not as the studied in depth: the comparative case study method. topic or object of interest, but as \"entry\" into You can find examples of this in the discussion of the social relations of the setting. The idea is to comparative and historical methods in Chapter 11 tap into people's expertise. of this book. (199& 57) Institutional Ethnography Here are two examples of this approach. Institutional ethnography is an approach origi- nally developed by Dorothy Smith (1978) to better Mothering Schooling and Child Development understand women's everyday experiences by dis- covering the power relations that shape those expe- Our first example of institutional ethnography riences. Today this methodology has been extended is a study by Alison Griffith (1995), who collected to the ideologies that shape the experiences of any data with Dorothy Smith on the relationship oppressed subjects. among mothering, schooling, and children's developmenL Griffith started by interviewing Smith and other sociologists believe that if re- mothers from three cities of southern Ontario searchers ask women or other members of subordi- about their everyday work of creating a relation- nated groups about \"how things work,\" they can ship between their families and the schooL This discover the institutional practices that shape their was the starting point for other interviews with realities (M. L Campbell 1998; D. Smith 1978). parents, teachers, school administrators, social The goal of such inquiry is to uncover forms of op- workers, school psychologists, and central office pression that more traditional types of research of- administrators. ten overlook. In her findings, Griffith explained how the dis- Dorothy Smith's methodology is similar to eth- course about mothering had shifted its focus over nomethodology in the sense that the subjects time from a mother-child interaction to \"child- themselves are not the focus of the inquiry. The in- centered\" recommendations. She saw a distinct stitutional ethnographer starts with the personal similarity in the discourse used by schools, the me- experiences of individuals but proceeds to uncover dia (magazines and television programs), the state, the institutional power relations that structure and child development professionals. and govern those experiences. In this process, the Teachers and child development professionals institutional ethnography A research technique saw the role of mothers in terms of a necessary col- in which the personal experiences of individuals laboration between mothers and schools for the are used to reveal power relationships and other child to succeed not only in school but also in life. characteristics of the institutions within which they Because of unequal resources, all mothers do not operate. participate in this discourse of \"good\" child devel- opment the same way. Griffith found that working- class mothers were perceived as weaker than
Some Qualitative Field Research Paradigms 301 middle-class mothers in the \"stimulation\" effort of This approach began in Third World research schooling. Griffith argues that this child develop- development, but it spread quickly to Europe and ment discourse, embedded in the school institution, North America (Gaventa 1991). It comes from a perpetuates the reproduction of class by making vivid critique of classical social science research. Ac- middle-class ideals for family-school relations the cording to the PAR paradigm, traditional research is norm for everyone. perceived as an \"elitist model\" (Whyte, Green- wood, and Lazes 1991) that reduces the \"subjects\" Compulsory Heterosexuality of research to \"objects\" of research. According to many advocates of the PAR perspective, the distinc- The second illustration of institutional ethnography tion between the researcher and the researched is taken from Didi Khayatt's (1995) study of the in- should disappear.. They argue that the subjects who stitutionalization of compulsory heterosexuality in will be affected by research should also be respon- schools and its effects on lesbian students. In 1990, sible for its design. Khayatt began her research by interviewing 12 Toronto lesbians, 15 to 24 years of age. Beginning Implicit in this approach is the belief that with the young women's viewpoint, she expanded research functions not only as a means of knowl- her inquiry to other students, teachers, guidance edge production but also as a \"tool for the educa- counselors, and administrators. tion and development of consciousness as well as mobilization for action\" (Gaventa 1991: 121-22). Khayatt found that the school's administrative Advocates of participatory action research equate practices generated a compulsory heterosexuality, access to information with power and argue that which produced a sense of marginality and vulner- this power has been kept in the hands of the domi- ability among lesbian students. For example, the nant class, sex, ethnicity, or nation. Once people school didn't punish harassment and name-calling see themselves as researchers, they automatically against gay students. The issue of homosexuality regain power over knowledge. was excluded from the curriculum lest it appear to students as an alternative to heterosexuality. Examples of this approach include research on community power structures, corporate research, In both of the studies I've described, the in- and \"right-to-know\" movements (Whyte, Green- quiry began with the women's standpoint-moth- wood, and Lazes 1991). Here are two examples of ers and lesbian students. However, instead of em- corporate research that used a PAR approach. phasizing the subjects' viewpoints, both analyses focused on the power relations that shaped these The Xerox Corporation women's experiences and reality. A participatory action research project took place at Participatory Action Research the Xerox corporation at the instigation of leaders of both management and the union. Management's Our final field research paradigm takes us further goal was to lower costs so that the company could along in our earlier discussion of the status and thrive in an increasingly competitive market. The power relationships linking researchers to the sub- union suggested a somewhat broader scope: im- jects of their research. Within the participatory proving the quality of working life while lowering a<.'tion research paradigm (PAR), the researcher'S manufacturing costs and increasing productivity. function is to serve as a resource to those being studied-typically, disadvantaged groups-as an participatory action research An approach to opportunity for them to act effectively in their own social research in which the people being studied interest. The disadvantaged subjects define their are given control over the purpose and procedures problems, define the remedies desired, and take the of the research; intended as a counter to the implicit lead in designing the research that will help them view that researchers are superior to those they realize their aims. study.
302 Chapter 10: Qualitative Field Research Company managers began by focusing atten- affect their own lives. Bernita Quoss, Margaret tion on shop-level problems; they were less con- Cooney, and Terri Longhurst (2000) report a re- cerned with labor contracts or problematic mana- search project involving welfare policy in Wyo- gerial policies. At the time, management had a plan ming. University students, many of them welfare to start an \"outsourcing\" program that would lay recipients, undertook research and lobbying efforts off 180 workers, and the union had begun mobiliz- aimed at getting Wyoming to accept postsecondary ing to oppose the plan. Peter Lazes, a consultant education as \"work\" under the state's new welfare hired by Xerox, spent the first month convincing regulations. management and the union to create a \"cost study team\" (CST) that included workers in the wire har- This project began against the backdrop of the ness department. 1996 Personal Responsibility and Work Opportu- nity Act (PRWORA), which Eight full-time workers were assigned to the CST for six months. Their task was to study the eliminated education waivers that had been possibilities of making changes that would save available under the previous welfare law, the the company $3.2 million and keep the 180 jobs. 1988 Family Support Act (FSA). These waivers The team had access to all financial information had permitted eligible participants in the cash and was authorized to call on anyone within the assistance AFDC program to attend college as company. This strategy allowed workers to make an alternative to work training requirements. suggestions outside the realm usually available to Empirical studies of welfare participants who them. According to Whyte and his colleagues, \"re- received these waivers have provided evidence shaping the box enabled the CST to call upon man- that education, in general, is the most effective agement to explain and justify all staff services\" way to stay out of poverty and achieve self- (1991: 27). Because of the changes suggested by sufficiency. the CST and implemented by management, the company saved the targeted $3.2 million. (QUOSS. Cooney. and Longhurst 2000: 47) Management was so pleased by this result that The students began by establishing an organi- it expanded the wire harness CST project to three zation called Empower and by making presenta- other departments that were threatened by compe- tions on campus to enlist broad student and tition. Once again, management was happy about faculty support. They compiled existing research the money saved by the teams of workers. relevant to the issue and established relationships with members of the state legislature. By the The Xerox case study is an interesting example time the 1997 legislative session opened, the stu- of participatory action research because it shows dents were actively engaged in the process of how the production of knowledge does not always modifying state welfare laws to offset the shift in have to be an elitist enterprise. The \"experts\" do not federal policy. necessarily have to be the professionals. According to Whyte and his colleagues, \"At Xerox, participa- The students prepared and distributed fact tory action research created and guided a powerful sheets and other research reports that would be process of organizational learning-a process relevant to the legislators' deliberations. They at- whereby leaders of labor and management learned tended committee meetings and lobbied legislators from each other and from the consultant/facilitator, on a one-to-one basis. When erroneous or mislead- while he learned from them\" (1991: 30). ing data were introduced into the discussions, the student-researchers were on hand to point out the PAR and Welfare Policy errors and offer corrections. Participatory action research often involves poor Ultimately, they were successfuL Welfare people, as they are typically less able than other recipients in Wyoming were allowed to pursue groups to influence the policies and actions that postsecondary education as an effective route out of poverty.
Conducting Qualitative Field Research 303 Blocking aDemolition to activists in the future. This indicates that there can be many different forms of participatory action In another example of researchers being directly research. involved in what they study, John Lofland (2003) detailed the demolition of a historic building in At the same time, this is a valuable study for Davis, California, and community attempts to block a study of research methods, because Lofland, the demolition. One thing that makes the book es- as the author of research methods textbooks, is pecially unusual is its reliance on photographs and particularly sensitive to the methodological aspects facsimile news articles and government documents of the study. as raw data for the analysis (and for the reader): what Lofland refers to as \"documentary sociology.\" The depth and intensity of my involvement is a two-edged sword. On the one edge, my As Lofland explains, he was involved in the is- involvement provided me with a view closer sue first as an active participant, joining with other than that of some other people. I was one type community members in the attempt to block dem- of \"insider.\" This means I could gather data of olition of the Hotel Aggie (also known as the \"Ter- certain sorts that were not available to the less minal Building\" and \"Terminal Hotel\"). Built in involved. 1924 in a town of around a thousand inhabitants, the hotel fell victim to population growth and ur- On the other edge, my partisanship clearly ban development. Lofland says his role as re- poses the threat of bias. I have always been searcher 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 carillot be the final judge. Each reader will have to form her or his Before that, I was only and simply an involved own assessment. I can hope, however, that the citizen. Along with many other people, I was \"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 so few photographs in this book taken by me is the case with some other methods of repre- before that date, but many after that date. I had senting what happened. then begun seriously to document what was going on with a camera and field notes. (Lof/and 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 Chap- day, it was my practice to indicate to everyone I ter 2, all our thoughts occur within, and are shaped encountered that I was \"writing a book\" about by, 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 en- (Lof/and 2003: 20) rich the observation of social life. Recall the discussion of informed consent in Conducting Qualitative Chapter 3, a method of protecting research sub- Field Research jects. In this case, as Lofland notes elsewhere, ex- plicit consent was not necessarily needed here, be- So far in this chapter we've examined the kinds of cause the behavior being studied was public. Still, topics appropriate to qualitative field research, spe- his instincts as a social researcher were to ensure cial considerations in doing this kind of research, that he treat subjects appropriately. and a sampling of paradigms that direct different types of research efforts. Along the way we've seen One of Lofland's purposes was to study this failed attempt to secure \"historic preservation\" sta- tus for a building, thus providing useful information
304 Chapter 10: Qualitative Field Research some examples that illustrate field research in ac- probably a mixture of fact and point of view. Mem- tion. To round out the picture, we turn now to bers of the political group in our example (as well specific ideas and techniques for conducting field as members of opposing political groups) would be research, beginning with how researchers prepare unlikely to provide completely unbiased informa- for work in the field. tion. Before making your first contact with the stu- dent group, then, you should already be quite fa- Preparing for the Field miliar with it, and you should understand its general philosophical context. Suppose for the moment that YOll've decided to undertake field research on a campus political or- There are many ways to establish your initial ganization. Let's assume further that you're not a contact with the people you plan to study. How member of that group, that you do not know a you do it will depend, in part, on the role you in- great deal about it, and that you'll identify yourself tend to play. Especially if you decide to take on the to the participants as a researcher. This section vvill role of complete participant, you must find a way use this example and others to discuss some of the to develop an identity with the people to be stud- ways you might prepare yourself before undertak- ied. If you wish to study dishwashers in a restau- ing direct observations. rant, the most direct method would be to get a job as a dishwasher. In the case of the student political As is true of all research methods, you would group, you might simply join the group. be well advised to begin with a search of the rele- vant literature, filling in your knowledge of the Many of the social processes appropriate to subject and learning what others have said about field research are open enough to make your it. (Library research is discussed at length in contact with the people to be studied rather Appendix A.) simple and straightforward. If you wish to observe a mass demonstration, just be there. If you wish In the next phase of your research, you might to observe patterns in jaywalking, hang around wish to discuss the student political group with busy streets. others who have already studied it or with anyone else likely to be familiar with it. In particular, you Whenever you wish to make more formal con- might find it useful to discuss the group with one tact with the people, identifying yourself as a re- or more informants (discussed in Chapter 7). searcher, you must establish a rapport with them. Perhaps you have a friend who is a member, or You might contact a participant with whom you you can meet someone who is. This aspect of your feel comfortable and gain that person's assistance. preparation is likely to be more effective if your In studying a formal group, you might approach relationship with the informant extends beyond the groups' leaders, or you might find that one of your research role. In dealing with members of the your informants can introduce you. group as informants, you should take care that your initial discussions do not compromise or Although you'll probably have many options limit later aspects of your research. Keep in mind in making your initial contact with the group, that the impression you make on the informant, realize that your choice can influence your subse- the role you establish for yourself, may carryover quent observations. Suppose, for example, that into your later effort. For example, creating the you're studying a university and begin with high- initial impression that you may be an undercover level administrators. This choice is likely to have FBI agent is unlikely to facilitate later observations a couple of important consequences. First, your of the group. initial impressions of the university will be shaped to some extent by the adrninistrators' views, You should also be wary about the information which will be quite different from those of students you get from informants. Although they may have or faculty. This initial impression may influence more direct, personal knowledge of the subject the way you observe and interpret events subse- under study than you do, what they \"know\" is quently-especially if you're unaware of the influence.
Conducting Qualitative Field Research 305 Second, if the administrators approve of your When Cecilia Menjivar (2000) wanted to learn research project and encourage students and about the experiences of Salvadoran immigrants in faculty to cooperate with you, the latter groups will San Francisco, she felt in-depth interviews would probably look on you as somehow aligned with the be a useful technique, along with personal obser- administration, which can affect what they say to vations. Before she was done, she had discovered you. For example, faculty members might be reluc- a much more complex system of social processes tant to tell you about plans to organize through the and structures than we would have imagined. Al- teamsters' union. though it was important for new immigrants to have a support structure of family members already In making a direct, formal contact with the in the United States, Menjivar found that her inter- people you want to study, you'll be required to give viewees were often reluctant to call on relatives for them some explanation of the purpose of your help, for several reasons. On the one hand, they study. Here again, you face an ethical dilemma. might jeopardize those family members who were Telling them the complete purpose of your research here illegally and living in poverty. At the same might eliminate their cooperation altogether or time, asking for help would put them in debt to significantly affect their behavior. On the other those helping them out. Menjivar also discovered hand, giving only what you believe would be an that Salvadoran gender norms put women immi- acceptable explanation may involve outright decep- grants in an especially difficult situation, because tion. Your decisions in this and other matters will they were largely prohibited from seeking the help probably be largely determined by the purpose of of men they weren't related to, lest they seem to your study, the nature of what you're studying, the obligate themselves sexually. These are the kinds of observations you wish to use, and similar factors, discoveries that can emerge from open-ended, in- but ethical considerations must be taken into ac- depth interviewing. count as well. We've already discussed interviewing in Chap- Previous field research offers no fixed rule- ter 9, and much of what was said there applies to methodological or ethical-to follow in this regard. qualitative field interviewing. The interviewing Your appearance as a researcher, regardless of your you'll do in connection with field observation, stated purpose, may result in a warm welcome however, is different enough to demand a separate from people who are flattered that a scientist finds treatment. In surveys, questionnaires are rigidly them important enough to study. Or, it may result structured; however, less-structured interviews are in your being totally ostracized or worse. It proba- more appropriate to field research. Herbert and bly wouldn't be a good idea, for example, to burst Riene Rubin (1995: 43) describe the distinction as into a meeting of an organized crime syndicate and follows: \"Qualitative interviewing design is flexible, announce that you're writing a term paper on or- iterative, and continuous, rather than prepared ganized crime. in advance and locked in stone.\" They elaborate in this way: Qualitative Interviewing Design in qualitative interviewing is iterative. In part, field research is a matter of going where the That means that each time you repeat the basic action is and simply watching and listening. As the process of gathering information, analyzing it, baseball legend Yogi Berra said, \"You can see a lot winnowing it, and testing it, you come closer to just by observing\"-provided that you're paying at- a clear and convincing model of the phenome- tention. At the same time, as I've already indicated, non you are studying.... field research can involve more active inquiry. Sometimes it's appropriate to ask people questions The continuous nature of qualitative inter- and record their answers. Your on-the-spot observa- vievving means that the questioning is re- tions of a full-blown riot will lack something if you designed throughout the project. don't know why people are rioting. Ask somebody. (Rubin and Rubill 1995:46-47)
306 Chapter 10: Qualitative Field Research Unlike a survey, a qualitative interview is particular context that omits altogether the most an interaction between an interviewer and a relevant answers. respondent in which the interviewer has a general plan of inquiry, including the topics to be covered, Suppose, for example, that you want to find but not a set of questions that must be asked with out why a group of students is rioting and pillaging particular words and in a particular order. At the on campus. You might be tempted to focus your same time, the qualitative interviewer, like the sur- questioning on how students feel about the dean's vey interviewer, must be fully familiar with the recent ruling that requires students always to carry questions to be asked. This allows the interview to The Practice ofSodal Research with them on campus. proceed smoothly and naturally. (Makes sense to me.) Although you may collect a great deal of information about students' attitudes A qualitative interview is essentially a conver- toward the infamous ruling, they may be rioting sation in which the interviewer establishes a gen- for some other reason. Perhaps most are simply eral direction for the conversation and pursues joining in for the excitement. Properly done, field specific topics raised by the respondent. Ideally, the research interviewing enables you to find out. respondent does most of the talking. If you're talk- ing more than 5 percent of the time, that's probably Although you may set out to conduct inter- too much. views with a reasonably clear idea of what you want to ask, one of the special strengths of field re- Steinar Kvale (1996: 3-5) offers two metaphors search is its flexibility. In particular, the answers for interviewing: the interviewer as a \"miner\" or as evoked by your initial questions should shape your a \"traveler.\" The first model assumes that the sub- subsequent ones. It doesn't work merely to ask ject possesses specific information and that the in- preestablished questions and record the answers. terviewer's job is to dig it out. By contrast, in the Instead, you need to ask a question, listen carefully second model, the interviewer to the answer, interpret its meaning for your gen- eral inquiry, and then frame another question wanders through the landscape and enters into either to dig into the earlier answer or to redirect conversations with the people encountered. the person's attention to an area more relevant to The traveler explores the many domains of the your inquiry. In short, you need to be able to listen, country, as unknown territory or with maps, think, and talk almost at the same time. roaming freely around the territory.... The in- terviewer wanders along with the local inhabi- The discussion of probes in Chapter 9 provides tants, asks questions that lead the subjects to a useful guide to getting answers in more depth tell their own stories of their lived world. without biasing later answers. More generally, field interviewers need the skills involved in being a Asking questions and noting answers is a natu- good listener. Be more interested than interesting. ral human process, and it seems simple enough to Learn to say things like \"How is that?\" \"In what add it to your bag of tricks as a field researcher. ways?\" \"How do you mean that?\" \"What would be Be a little cautious, however. Wording questions an example of that?\" Learn to look and listen ex- is a tricky business. All too often, the way we ask pectantly, and let the person you're interviewing fill questions subtly biases the answers we get. Some- in the silence. times we put our respondent under pressure to look good. Sometimes we put the question in a At the same time, you can't afford to be a to- tally passive receiver. You'll go into your interviews qualitative interview Contrasted with survey in- with some general (or specific) questions you want terviewing, the qualitative interview is based on a answered and some topics you want addressed. At set of topics to be discussed in depth rather than times you'll need the skill of subtly directing the based on the use of standardized questions. flow of conversation. There's something we can learn in this connec- tion from the martial arts. The aikido master never resists an opponent's blow but instead accepts it,
Conducting Qualitative Field Research 307 joins with it, and then subtly redirects it in a more first attempt to focus things back on the student's appropriate direction. Field interviewing requires own choice of major (\"Did you talk to your an analogous skill. Instead of trying to halt your uncle ... 1\") fails. The second attempt (\"So is respondent's line of discussion, learn to take what your main interest ... ?\") succeeds. Now the stu- he or she has just said and branch that comment dent is providing the kind of information you're back in the direction appropriate to your purposes. looking for. It's important for field researchers to Most people love to talk to anyone who's really develop the ability to \"control\" conversations in interested. Stopping their line of conversation tells this fashion. them that you are not interested; asking them to elaborate in a particular direction tells them that Herbert and Riene Rubin offer several ways you are. to control a \"guided conversation,\" such as the following: Consider this hypothetical example in which you're interested in why college students chose If you can limit the number of main topics, it is their majors. easier to maintain a conversational flow from one topic to another. Transitions should be You: What are you majoring in? smooth and logical. \"We have been talking REsp: Engineering. about mothers, now let's talk about fathers,\" sounds abrupt. A smoother transition might be, You: I see. How did you come to choose \"You mentioned your mother did not care how engineering? you performed in school-was your father more involved?\" The more abrupt the transi- RESP: I have an uncle who was voted the best engi- tion, the more it sounds like the interviewer neer in Arizona in 1981. has an agenda that he or she wants to get through, rather than wanting to hear what the You: Gee, that's great. interviewee has to say. RESP: Yeah. He was the engineer in charge of de- (1995.,123) veloping the new civic center in Tucson. It was written up in most of the engineering Because field research interviewing is so much journals. like normal conversation, researchers must keep reminding themselves that they are not having a You: I see. Did you talk to him about your becom- normal conversation. In normal conversations, ing an engineer? each of us wants to come across as an interesting, worthwhile person. If you watch yourself the next REsp: Yeah. He said that he got into engineering time you chat with someone you don't know too by accident. He needed a job when he well, you'll probably find that much of your atten- graduated from high schooL so he went to tion is spent on thinking up interesting things to work as a laborer on a construction job. He say-contributions to the conversation that will spent eight years working his way up from make a good impression. Often, we don't really the bottom, until he decided to go to college hear each other, because we're too busy thinking of and come back nearer the top. what we'll say next. As an interviewer, the desire to appear interesting is counterproductive. The You: So is your main interest civil engineering, interviewer needs to make the other person seem like your uncle, or are you more interested interesting, by being interested-and by listening in some other branch of engineering? more than talking. (Do this in ordinary conversa- tions, and people will actually regard you as a great RESP: Actually, I'm leaning more toward electrical conversationalist. ) engineering-computers, in particular. I started messing around with microcomput- John and Lyn Lofland (1995: 56-57) suggest ers when I was in high school, and my that investigators adopt the role of the \"socially long-term plan is ... Notice how the interview begins by wandering off into a story about the respondent's uncle. The
308 Chapter 10: Qualitative Field Research acceptable incompetent\" when interviewing. That As with all other aspects of field research, inter- is, offer yourself as someone who does not under- viewing improves vvith practice. Fortunately, it's stand the situation you find yourself in and must something you can practice any time you want. be helped to grasp even the most basic and obvious Practice on your friends. aspects of that situation: \"A naturalistic investigator, almost by definition, is one who does not under- Focus Groups stand. She or he is 'ignorant' and needs to be 'taught.' This role of watcher and asker of questions Although our discussions of field research so far is the quintessential student role\" (Lofland and have focused on studying people in the process Lofland 1995: 56). of living their lives, researchers sometimes bring people into the laboratory for qualitative Interviewing needs to be an integral part of the intervielving and observation. The focus group entire field research process. Later, I'll stress the method, which is also called group interviewing, is need to review your observational notes every essentially a qualitative method. It is based on night-making sense out of what you've observed, structured, semistructured, or unstructured inter- getting a clearer feel for the situation you're study- views. It allows the researcher/interviewer to ing, and finding out what you should pay more question several individuals systematically and attention to in further observations. In the same sinmltaneously. Focus group data teclmique is fashion, you'll need to review your notes on inter- typically used in market research but not views, recording especially effective questions and exclusively. detecting all those questions you should have asked but didn't. Start asking such questions the next Imagine that you're thinking about introducing time you interview. If you've recorded the inter- a new product. Let's suppose that you've invented views, replay them as a useful preparation for fu- a new computer that not only does word process- ture interviews. ing, spreadsheets, data analysis, and the like but also contains a fax machine, AM/FM/TV tuner, Steinar Kvale (1996: 88) details seven stages in CD and DVD player/recorder, microwave oven, the complete intervievving process: denture cleaner, and coffeemaker. To highlight its computing and coffee-making features, you're 1. Thematizing: clarifying the purpose of the inter- thinking of calling it \"The Compulator.\" You figure views and the concepts to be explored the new computer will sell for about $28,000, and you want to know whether people are likely 2. Designing: laying out the process through which to buy it. Your prospects might be well served by you'll accomplish your purpose, including a focus groups. consideration of the ethical dimension In a focus group, typically 12 to 15 people are 3. /nten!ielVillg.: doing the actual interviews brought together in a room to engage in a guided discussion of some topic-in this case, the accept- 4. Transcribing: creating a written text of the ability and salability of The Compulator. The sub- interviews jects are selected on the basis of relevance to the topic under study. Given the likely cost of The 5. Analyzing: determining the meaning of gathered Compulator, your focus group participants would materials in relation to the purpose of the study probably be limited to upper-income groups, for example. Other, similar considerations might figure 6. Verifying: checking the reliability and validity of into the selection. the materials Participants in focus groups are not likely to be 7. Reporting: telling others what you've learned chosen through rigorous, probability sampling methods. This means that the participants do not focus group A group of subjects interviewed to- statistically represent any meaningful population. gether, prompting a discussion. The technique is fre- quently used by market researchers, who ask a group of consumers to evaluate a product or discuss a type of commodity, for example.
Conduding Qualitative Field Research 309 However, the purpose of the study is to explore 6. The discussion must be conducted in a con- rather than to describe or explain in any definitive ducive environment. sense. Nevertheless, typically more than one focus group is convened in a given study because of (198&44-45) the serious danger that a single group of 7 to 12 people will be too atypical to offer any general- As we've seen, the group interview presents izable insights. several advantages, but it also has its challenges. In a focus group interview, much more than in any William Gamson (1992) has used focus groups other type of interview, the interviewer has to de· to examine how U.S. citizens frame their views of velop the skills of a moderator. Controlling the dy- political issues. Having picked four issues-affirma- namic within the group is a major challenge. Let- tive action, nuclear power, troubled industries, and ting one interviewee dominate the focus group the Arab-Israeli conflict-Gamson undertook a interview reduces the likelihood that the other sub- content analysis of press coverage to get an idea of jects will el\\.1Jress themselves. This can generate the the media context within which we think and talk problem of group conformity or groupthink, which about politics. Then the focus groups were con- is the tendency for people in a group to conform vened for a firsthand observation of the process of with opinions and decisions of the most outspoken people discussing issues with their friends. members of the group. Interviewers need to be aware of this phenomenon and try to get everyone Richard Krueger points to five advantages of to participate fully on all the issues brought in the focus groups: interview. Adding to the challenge, of course, is that the interviewer must resist overdirecting the 1. The technique is a socially oriented research interview and the interviewees, thus bringing her method capturing real-life data in a social or his own views into play. envirorunent. Although focus group research differs from 2. It has flexibility. other forms of qualitative field research, it further illustrates the possibilities for doing social research 3. It has high face validity. face-to-face with those we wish to understand. In addition, David Morgan (1993) suggests that focus 4. It has speedy results. groups are an excellent device for generating ques- tionnaire items for a subsequent survey. 5. It is low in cost. Like other social research techniques, focus (1988: 47) groups are adapting to new communication modal- ities. George Silverman (2005), for example, offers In addition to these advantages, group dynam- a discussion of telephone and online focus groups. ics frequently bring out aspects of the topic that would not have been anticipated by the researcher Recording Observations and would not have emerged from interviews with individuals. In a side conversation, for example, a The greatest advantage of the field research method couple of the participants might start joking about is the presence of an observing, thinking researcher the results of leaving out one letter from a prod- on the scene of the action. Even tape recorders and uct's name. This realization might save the manu- cameras cannot capture all the relevant aspects of facturer great embarrassment later on. social processes. Consequently, in both direct ob- servation and interviewing, it is vital to make full Krueger also notes some disadvantages of the and accurate notes of what goes on. If possible, focus group method, however: take notes on your observations as you observe. When that's not feasible, write down your notes 1. Focus groups afford the researcher less control as soon as possible afterward. than individual interviews. 2. Data are difficult to analyze. 3. Moderators require special skills. 4. Difference between groups can be troublesome. 5.. Groups are difficult to assemble.
310 Chapter 10: Qualitative Field Research Thursday August26, 12:00-1:00 all used to working like awell-oiled machine and they say,\"Oh,here is the woman,something different\"and sometimes they can be hor- R: What is challenging for women directors on adaily experience, on a ribe' they can resist your directing and they can,they can sabotage daily life? you, by taking along time to light, or to move sets, or to do some- thing ,..and during thattime you're wasting time,and that goes on J Surviving\" areport,and the report goes to the front [368J office,and, you know, and so on and so on and so on and so forth. And people upstairs R OK Could you develop alittle bit on that? [I need to work on my in- don't know what the circumstances are,and they are not aboutto terview schedule so that my interviewee answers with more elabo- fire acinematographer that is on their show for ever and ever .... nor ration without having to probe.] do they wantto know thatthis guy is areal bastard,and making your life ahorroLThey don't want to know that, so therefore, they go Yeah,1 mean it's all abouttrying to get,you know,in,trying to getthe off, because she's awoman let's not hire any more women, since he job,and try,you know,to do agreat job so that you are invited back has problems with womenJou know, so,there is that aspect. to the next thing\"And particularly since they are so many, you know, difficulties in women directing\" It makes it twice as hard to gain into [I need to review the literature on institutional discrimination. lt this position where you do an incredible job, because \"...you can't just seems that the challenges that Joy is facing are not amatter of aparticu- do an average job, you have to [347] do this job that just knocks your lar individual. She is in adouble bind situation where whether she com- socks off all the time,and sometimes you don't getthe opportunity plains or not,she will not be treated equal to men.Time seems to be one to do that, because either you don't have agood producer or you quantifiable measurement of how well she does her job and, as observed have so many pressures that you can't see straight or your script is in other professions, the fact that she is awoman is perceived as ahandi- lousy,and you have to make asilk purse out ofsow's haiLYou know, cap\" Review literature on women in high management position.! need to you have alot ofextra strikes against you than the average guy who keep asking about the dynamics between my interviewees and the has similar problems, because you are awoman and they look at it, crewmembers on the setThe cinematographer has the highest status on and women are more visible than men jn unique positions\" the set under the director.. Explore other interviews about reasons for conflict between them] [It seems that Joy is talking about the particularities ofthe film in- dustryJhere are not that many opportunities and in order to keep work- [Methods (note to myselfforthe next interviews): try to avoid ing, she needs to build acertain reputation.!t is only by continuing to di- phone interviews unless specific request from the interviewee.!t is rect that she can maintain or improve her reputation\"She thinks that it is difficult to assess how the interviewee feels with the questions\" Need even harder for women but does not explain it] body language because Ibecome more nervous about the interview process] R Hum . what about on the set did you experience, did it feel \"\",did people make it clear that you were awoman,and you felttreated Note: Anumber in brackets represents aword that was inaudible from the interview. It differently? [I am trying to get herto speak about more specific and is the number that appeared on the transcribing machine, with each interview starting more personal experiences without leading her answer] at count O.The numbers help the researcher locate apassage quickly when he or she reviews the interview Yeah,oh yeah,1 mean alot ofwomen have commiserated about, you know when you have to walk on the set for the first time, they're In your notes, include both your empirical ob- group leader (an observation), that you think this servations and your interpretations of them. In represents an attempt by Person X to take over lead- other words, record what you \"know\" has hap- ership of the group (an interpretation), and that you pened and what you \"think\" has happened. Be sure think you heard the leader comment to that effect in to identify these different kinds of notes for what response to the opposition (a tentative observation), they are\" For example, you might note that Person X spoke out in opposition to a proposal made by a Of course, you cannot hope to observe every- thing; nor can you record everything you do
Conduding Qualitative Field Research 311 observe, Just as your observations will represent a didn't have any trouble answering any of those sample of all possible observations, your notes vvill questions, how sure are you of your answers? represent a sample of your observations. The idea, Would you be vvilling to bet a hundred dollars that of course, is to record the most pertinent ones, a panel of impartial judges would observe what \"Interview Transcript Annotated vvith Researcher you recall? Memos\" provides an example given by Sandrine Zerbib from an in-depth interview vvith a woman Even if you pride yourself on having a photo- film director. graphic memory, it's a good idea to take notes ei- ther during the observation or as soon afterward as Some of the most important observations can possible, If you take notes during observation, do it be anticipated before you begin the study; others unobtrusively, because people are likely to behave vvill become apparent as your observations pro- differently if they see you taking dovvn everything gress. Sometimes you can make note taking easier they sayar do. by preparing standardized recording forms in ad- vance. In a study of jaywalking, for example, you Second, it's usually a good idea to take notes might anticipate the characteristics of pedestrians in stages. In the first stage, you may need to take that are most likely to be useful for analysis-age, sketchy notes (words and phrases) in order to gender. social class, ethnicity, and so forth-and keep abreast of what's happening, Then go off by prepare a form in which observations of these yourself and rewrite your notes in more detaiL If variables can be recorded easily. Alternatively, you you do this soon after the events you've observed, might develop a symbolic shorthand in advance to the sketchy notes should allow you to recall most speed up recording. For studying audience partici- of the details, The longer you delay, the less pation at a mass meeting, you might want to con- likely you'll be able to recall things accurately struct a numbered grid representing the different and fully. sections of the meeting room; then you could record the location of participants easily, quickly, and I know this method sounds logical, but it takes accurately, self-discipline to put it into practice. Careful obser- vation and note taking can be tiring, especially if it None of this advance preparation should involves excitement or tension and if it extends limit your recording of unanticipated events and over a long period, If you've just spent eight hours aspects of the situation, Quite the contrary, speedy observing and making notes on how people have handling of anticipated observations can give you been coping vvith a disastrous flood, your first de- more freedom to observe the unanticipated. sire afterward vvililikely be to get some sleep, dry clothes, or a drink You may need to take some in- You're already familiar vvith the process of tak- spiration from newspaper reporters who undergo ing notes, just as you already have at least informal the same sorts of hardships, then write their stories experience vvith field research in general. Like good to meet their deadlines. field research, however, good note taking requires careful and deliberate attention and involves Third, you'll inevitably wonder how much you specific skills, Some guidelines follow. (You can should record, Is it really worth the effort to write learn more from John and Lyn Lofland's Al1alyzing out all the details you can recall right after the ob- Sodal Settings [1995: 91-96].) servational session? The general guideline is yes. Generally, in field research you can't be really sure First, don't trust your memory any more of what's important and what's unimportant until than you have to; it's untrustworthy, To illustrate you've had a chance to review and analyze a great this point, try this experiment. Recall the last three volume of information, so you should record even or four movies you saw that you really liked. things that don't seem important at the outset. Now, name five of the actors or actresses, Who They may turn out to be significant after all. Also, had the longest hair? Who was the most likely to the act of recording the details of something start conversations? Who was the most likely to \"unimportant\" may jog your memory on some- make suggestions that others followed? Now, if you thing that is important.
312 ll! Chapter 10: Qualitative Field Research Realize that most of your field notes will not be iii Is it ethical to develop a calculated stance to- reflected in your final report on the project. Put ward other humans, that is, to be strategic in more harshly, most of your notes will be \"wasted.\" your relations? But take heart: Even the richest gold ore yields only about 30 grams of gold per metric ton, mean- iii Is it ethical to take sides or to avoid taking sides ing that 99.997 percent of the ore is wasted. Yet. in a factionalized situation? that 30 grams of gold can be hammered out to cover an area 18 feet square-the equivalent of iii Is it ethical to \"pay\" people with trade-offs for about 685 book pages. So take a ton of notes, and access to their lives and minds? plan to select and use only the gold. iii Is it ethical to \"use\" people as allies or infor- Like other aspects of field research (and all re- mants in order to gain entree to other people search for that matter), proficiency comes with or to elusive understandings? practice. The nice thing about field research is you can begin practicing now and can continue practic- Planning and conducting field research in a respon- ing in almost any situation. You don't have to be sible way requires attending to these and other eth- engaged in an organized research project to prac- ical concerns. tice observation and recording, You might start by volunteering to take the minutes at committee Strengths and Weaknesses meetings, for example. Or just pick a sunny day of Qualitative FieJd Research on campus, find a shady spot. and try observing and recording some specific characteristics of the Like all research methods, qualitative field research people who pass by. You can do the same thing at has distinctive strengths and weaknesses. As I've al- a shopping mall or on a busy street corner. Re- ready indicated, field research is especially effective member that observing and recording are profes- for studying subtle nuances in attitudes and behav- sional skills and, like all worthwhile skills, they im- iors and for examining social processes over time. prove with practice. As such, the chief strength of this method lies in the depth of understanding it permits. Whereas Research Ethics other research methods may be challenged as in Qualitative Field Research \"superficiaL\" this charge is seldom lodged against field research. As I've noted repeatedly, all forms of social research raise ethical issues, By bringing researchers into di- Flexibility is another advantage of field re- rect and often intimate contact with their subjects, search. As discussed earlier, you can modify your field research raises ethical concerns in a particu- field research design at any time. Moreover. you're larly dramatic way, Here are some of the issues always prepared to engage in field research, mentioned by John and Lyn Lofland (1995: 63): whenever the occasion should arise, whereas you could not as easily initiate a surveyor an iii Is it ethical to talk to people when they do not experiment. know you will be recording their words? Field research can be relatively inexpensive iii Is it ethical to get information for your own as well. Other social scientific research methods purposes from people you hate? may require costly equipment or an expensive research staff, but field research typically can be iii Is it ethical to see a severe need for help and undertaken by one researcher with a notebook and not respond to it directly? a pencil. This is not to say that field research is never expensive, The nature of the research proj- iii Is it ethical to be in a setting or situation but not ect, for example, may require a large number of commit yourself wholeheartedly to it? trained observers. Expensive recording equipment may be needed, Or you may wish to undertake
Strengths and Weaknesses of Qualitative Field Research 313 participant observation of interactions in pricey upon others for activities of daily living and Paris nightclubs. may consequently become a burden. Field research has several weaknesses as well. -Fear of alienation-from significant oth- First, being qualitative rather than quantitative, it is ers and health care givers, thereby creating not an appropriate means for arriving at statistical helplessness and hopelessness. descriptions of a large population, Observing casual political discussions in Laundromats, for example, -Fear of contagion-that cancer is trans- would not yield trustworthy estimates of the future missible and/or inherited. voting behavior of the total electorate. Neverthe- less, the study could provide important insights -Fear of losing one's dignity-losing con- into how political attitudes are formed. trol of all bodily functions and being totally vulnerable. To assess field research further. let's focus on the issues of validity and reliability. Recall that va- (Garam 1980 2167) idity and reliability are both qualities of measure- ments, Validity concerns whether measurements Observations and conceptualizations such as actually measure what they're supposed to rather these are valuable in their own right. In addition, than something else. Reliability, on the other hand, they can provide the basis for further research- is a matter of dependability: If you made the same both qualitative and quantitative. measurement again and again, would you get the same result? Let's see how field research stacks up Now listen to what Joseph Howell has to say in these respects. about \"toughness\" as a fundamental ingredient of life on Clay Street. a white, working-class neigh- Validity borhood in Washington, D.c.: Field research seems to provide measures with Most of the people on Clay Street saw them- greater validity than do survey and experimental selves as fighters in both the figurative and lit- measurements, which are often criticized as su- eral sense. They considered themselves strong, perficial and not really valid. Let's review a couple independent people who would not let them- of field research examples to see why this is so. selves be pushed around. For Bobbi. being a fighter meant battling the welfare department \"Being there\" is a powerful technique for gain- and cussing out social workers and doctors ing insights into the nature of human affairs in all upon occasion. It meant spiking Barry's beer their rich complexity. Listen, for example, to what with sleeping pills and bashing him over the this nurse reports about the impediments to pa- head with a broom, For Barry it meant telling tients' coping with cancer: off his boss and refusing to hang the door, an act that led to his being fired, It meant going Common fears that may impede the coping through the ritual of a duel vl'ith AI, It meant process for the person vvith cancer can include pushing Bubba around and at times getting the following: rough vvith Bobbi. -Fear of death-for the patient, and the June and Sam had less to fight about, implications his or her death will have for though if pressed they both hinted that they, significant others, too, would fight. Being a fighter led Ted into near conflict with Peg's brothers, Les into -Fear of incapacitation-because cancer conflict with Lonnie, Arlene into conflict with can be a chronic disease with acute episodes Phyllis at the bowling alley, etc that may result in periodic stressful periods, the variability of the person's ability to cope and (1973, 292) constantly adjust may require a dependency Even without having heard the episodes How- ell refers to in this passage, you have the distinct impression that Clay Street is a tough place to live in. That \"toughness\" comes through far more
314 Chapter 10: Qualitative Field Research powerfully through these field observations than it of comparative evaluations: identifying who is would in a set of statistics on the median number of more conservative than who, for example. Even fistfights occurring during a specified period. if you and I had different political orientations, we would probably agree pretty much in ranking These examples point to the superior validity of the relative conservatism of the members of a field research, as compared with surveys and ex- group. periments. The kinds of comprehensive measure- ments available to the field researcher tap a depth As we've seen, field research is a potentially of meaning in concepts such as common fears of powerful tool for social scientists, one that provides cancer patients and \"toughness\" (or such as liberal a useful balance to the strengths and weaknesses of and conservative) that are generally unavailable experiments and surveys. The remaining chapters to surveys and experiments. Instead of specifying of Part 3 present additional modes of observation concepts, field researchers commonly give detailed available to social researchers. illustrations. Reliability MAIN POINTS Field research, however, can pose problems of reli- Introduction ability. Suppose you were to characterize your best o Field research involves the direct observation friend's political orientations according to every- thing you know about him or her. Your assessment of social phenomena in their natural settings. of your friend's politics would appear to have con- Typically field research is qualitative rather siderable validity; certainly it's unlikely to be su- than quantitative. perficial. We couldn't be sure, however, that an- o In field research, observation, data processing, other observer would characterize your friend's and analysis are interwoven, cyclical processes. politics the same way you did, even with the same amount of observation. Topics Appropriate to Field Research o Field research is especially appropriate to topics Although they are in-depth, field research measurements are also often very personal. How and processes that are not easily quantifiable, I judge your friend's political orientation depends that are best studied in natural settings, or that greatly on my own, just as your judgment depends change over time. Among these topics are prac- on your political orientation. Conceivably, then, tices, episodes, encounters, roles, relationships, you could describe your friend as middle-of-the- groups, organizations, settlements, social road, although I might feel that I've been observing worlds, and lifestyles or subcultures. a fire-breathing radical. Special Considerations in Qualitative As I suggested earlier, researchers who use Field Research qualitative techniques are conscious of this issue o Among the special considerations involved in and take pains to address it. Individual researchers often sort out their ovvn biases and points of view, field research are the various possible roles of and the communal nature of science means that the observer and the researcher'S relations with their colleagues will help them in that regard. subjects. As a field researcher, you must decide Nevertheless, it's prudent to be wary of purely whether to observe as an outsider or as a par- descriptive measurements in field research-your ticipant, whether or not to identify yourself as a own, or someone else's. If a researcher reports researcher, and how to negotiate your relation- that the members of a club are fairly conservative, ships with subjects. such a judgment is unavoidably linked to the researcher's own politics. You can be more trusting
Additional Readings 315 Some Qualitative Field Research Paradigms KEY TERMS o Field research can be guided by anyone of The following terms are defined in context in the several paradigms, such as naturalism, ethno- chapter and at the bottom of the page where the term methodology, grounded theory, case studies is introduced, as well as in the comprehensive glossary and the extended case method, institutional at the back of the book. ethnography, and participatory action research. case study institutional ethnography Conducting Qualitative Field Research ethnography naturalism o Preparing for the field involves doing back- ethnomethodology participatory action extended case method ground research, determining how to make focus group research contact with subjects, and resolving issues grounded theory qualitative interview of what your relationship to your subjects reactivity will be. o Field researchers often conduct in-depth inter- REVIEW QUESTIONS AND EXERCISES views that are much less structured than those conducted in survey research. Qualitative inter- L Think of some group or activity you participate vievving is more of a guided conversation than a in or are very familiar with. In two or three search for specific information. Effective inter- paragraphs, describe how an outsider might ef- viewing involves skills of active listening and fectively go about studying that group or activ- the ability to direct conversations unobtrusively. ity. What should he or she read, what contacts o To create a focus group, researchers bring sub- should be made, and so on? jects together and observe their interactions as they explore a specific topic 2. Choose any two of the paradigms discussed in o Whenever possible, field observations should this chapter. Then describe how your hypotheti- be recorded as they are made; otherwise, cal study from Exercise 1 might be conducted if they should be recorded as soon afterward as you followed each. Compare and contrast the possible. way these paradigms might work in the context o Among the advantages of field research are the your study. depth of understanding it can provide, its flexi- bility, and (usually) its inexpensiveness. 3. To explore the strengths and weaknesses of ex- periments, surveys, and field research, choose Research Ethics in Qualitative a general research area (e.g., prejudice, political Field Research orientation, education) and write brief descrip- o Conducting field research responsibly involves tions of studies in that area that could be con- ducted using each of these three methods. In confronting several ethical issues that arise from each case, explain why the chosen method is the researcher'S direct contact with subjects. the most appropriate for the study you describe. Strengths and Weaknesses of Qualitative 4. Return to the example you devised in response Field Research to Exercise 1 and list five ethical issues that you o Compared with surveys and e;.,:periments, field can imagine having to confront if you were to undertake your study. research measurements generally have more validity but less reliability. Also, field research is 5. Using InfoTrac College Edition, find a research generally not appropriate for arriving at statisti- report using the Grounded Theory Method. cal descriptions of large populations. Summarize the study design and main findings. ADDITIONAL READINGS Adler, Patricia A, and Peter Adler. 2003. \"The Promise and Pitfalls of Going into the Field.\" Contexts 2 (2): 41-47. An excellent report on
316 \" Chapter 10: Qualitative Field Research some of the complexities of field research, with contains a wealth of references to field research tips on distinguishing good from not-so-good illustrations. ethnography. Long, Andrew E, and Mary Godfrey. 2004. \"An Burawoy, MichaeL 1998. \"The Extended Case Evaluation Tool to Assess the Quality of Qualita- Method.\" Sociological TlzeOlY 16 (1): 4-33. Here's tive Research Studies.\" Imernational Journal ofSo- a discussion of this paradigm by its creator. cial Research Methods 7 (2): 181-96. The criteria for evaluating the quality of quantitative stud- Denzin, Norman Ie and Yvonna S. Lincoln, eds. ies, such as surveys, have been established for 1994. Handbook of Qualitative Research. Thousand some time, but evaluating qualitative studies has Oaks, CA: Sage. This handbook is an extensive been more elusive This article details an evalu- collection of articles covering issues regarding ation tool for the latter purpose. the wide field of qualitative research . This book has also been published as three volumes: VoL L Morgan, David L, ed . 1993. Successful FoClls Groups. The Landscape of Qualitative Research. Theories and Advancing the State ofthe Art. Newbury Park, CA: Issues; VoL 2. Strategies of Qualitative Inquiry; and Sage. This collection of articles on the uses of VoL 3. Interpreting Qualirative Materials. focus groups points to many aspects not nor- mally considered. Gans, Herbert J. 1999. \"Participant Observation in the Era of 'Ethnography.'\" ]oumalofContempo- Shaffir, William B., and Robert A. Stebbins, eds. rary Ethnography 28 (5): 540-48. A thoughtful 1991. Experiencing Fieldwork.: An Inside View of discussion of recent developments and problems Qualitative Research. Newbury Park, CA: Sage. in participant observation. Several field research practitioners discuss the nature of the craft and recall experiences in the Gubrium, Jaber E, and James A. Holstein. 1997. field. Here's an opportunity to gain a \"feel\" for The New Language of Qualitative Method. New the method as well as learn some techniques. York: Oxford University Press. This book pro- vides the necessary foundations for understand- Shostak, Arthur, ed. 1977. Our Sociological Eye.: ing some of the main approaches or traditions Personal ESsays on Society and Culture. Port Wash- in qualitative field research. ington, NY: Alfred. An orgy of social scientific introspection, this delightful collection of first- Johnson, Jeffrey C 1990. Selecting Ethnographic Infor- person research accounts offers concrete, inside mants. Newbury Park, CA: Sage. The author dis- views of the thinking process in sociological re- cusses the various strategies that apply to the search, especially field research. task of sampling in field research. Silverman, David. 1999. Doing Qualitative Research: A Kelle, Udo, ed. 1995. Compllter-Aided Qualirative Data Practical Handbook. Thousand Oaks, CA: Sage. Analysis: Theory, lviethods, and Practice. Thousand This book focuses on the process of collecting Oaks, CA: Sage. An international group of and interpreting qualitative data. scholars report on their experiences with a variety of computer programs used in the analy- Strauss, Anselm, and Juliet Corbin. 1998. Basics of sis of qualitative data. Qualitative Research.' Techniques and Procedures for Developing Grounded TheOly. Thousand Oaks, CA: Kvale, Steinar. 1996. IlIterViews: An Imroductioll to Sage. This is a very important book to read be- Qualitative Research Interviewing. Thousand Oaks, fore data collection and during data analysis if CA: Sage. An in-depth presentation of in-depth you choose to take a grounded theory approach. interviewing. Besides presenting techniques, Kvale places interviewing in the context of post- Uwe, Flick. 1998. An Introduction to Qualitative Re- modernism and other philosophical systems. search. Thousand Oaks, CA: Sage. This book pro- vides a good entrance to the large field of quali- Lofland, John, and Lyn Lofland. 1995. Allalyzing So- tative research. cial Settings. 3rd ed. Belmont, CA: Wadsworth. An unexcelled presentation of field research SPSS EXERCISES methods from beginning to end. This eminently readable book draws the links between the See the booklet that accompanies your text for exer- logic of scientific inquiry and the nitty-gritty cises using SPSS (Statistical Package for the Social practicalities of observing, communicating, Sciences). There are exercises offered for each chapter, recording, filing, reporting, and everything else and you'll also find a detailed primer on using SPSS. involved in field research. In addition, the book
Online Study Resources 317 Online Study Resources ing various data-analysis software such as SPSS and NVivo. Sociology~ Now'\": Research Methods WEB LINKS FOR THIS CHAPTER 1. Before you do your final review of the chapter, take the SociologyNow Research Methods diagnos- Please realize that the Internet is an evolv- tic quiz to help identify the areas on which you ing entity, subject to change. Nevertheless, should concentrate. You'll find information on these few websites should be fairly stable. this online tooL as well as instructions on how Also, check your book's website for even more lVeb to access all of its great resources, in the front of Links. These websites, current at the time of this book's the book. publication, provide opportunities to learn about qual- itative field research. 2. As you review, take advantage of the Sociology Now Research Methods customized study plan, Rahmat M. Samil(-Ibrahlm, Online Grounded based on your quiz results. Use this study plan Theory Articles with its interactive exercises and other re- http://wIVw.vlsm.org/rms46/citations-gtm2.htm! sources to master the materiaL As the name suggests, this website contains hot links to numerous scholarly articles using or discussing the 3. When you're finished with your review, take Grounded Theory Method. the posttest to confirm that you're ready to move on to the next chapter. Goshen College: WEB Links to Participatory Action Research Sites WEBSITE FOR THE PRACTICE http://www.goshen.edu/soan/soan96p.htm OF SOCIAL RESEARCH 11 TH EDITION This will show you many of the ways in which social researchers wed action and inquiry in the design and Go to your book's website at http://sociology execution of research projects. .wadsworth.com/babbie_practicelle for tools to aid you in studying for your exams. You'll find Tuto- Dr. Rita S. Y. Berry, \"Collecting Data by rial Quizzes with feedback, Internet Exercises, Flashcards, In-Depth Interviewing\" and Chapter Tutorials, as well as Extended Projects, Info- http://wwwJeeds.ac.uk/educolldocuments/ TTac College Edition search terms, Social Research in 00000 1172.htm Cyberspace, GSS Data, Web Lillks, and primers for us- Here's a lengthy and practical discussion of this data- collection technique.
Unobtrusive Research Introduction The Consequences of Globalization Content Analysis Units of Analysis Topics Appropriate Problems of Validity to Content Analysis Problems of Reliability Sampling in Content Sources of Existing Analysis Statistics Coding in Content Analysis An Illustration of Content Comparative and Historical Analysis Research Strengths and Weaknesses of Content Analysis Examples of Comparative and Historical Research Analyzing Existing Statistics Sources of Comparative Durkheim's Study and Historical Data of Suicide Analytical Techniques Sociology~Now'\": Research Methods Use this online tool to help you make the grade on your next exam. After reading this chapter, go to the \"Online Study Resources\" at the end of the chapter for instructions on how to benefit from SoaologyNow: Research Methods.
Introduction 319 Introduction In 1966, Eugene Webb and three colleagues published an ingenious little book on social research With the exception of the complete observer in (revised in 2000) that has become a classic. It fo- field research, each of the modes of observation cuses on the idea of unobtrusive or nonreactive re- discussed so far requires the researcher to intrude search. Webb and his colleagues have played freely to some degree on whatever he or she is studying. ,vith the task of learning about human behavior by This is most obvious in the case of experiments, fol- observing what people inadvertently leave behind lowed closely by survey research. Even the field re- them. Do you want to know what exhibits are the searcher, as we've seen, can change things in the most popular at a museum? You could conduct a process of studying them. poll, but people nlight tell you what they thought you wanted to hear or what might make them look At least one previous example in this book, intellectual and serious. You could stand by differ- however, was totally exempt from that danger. ent exhibits and count the viewers that came by, but Durkheim's analysis of suicide did nothing to affect people might come over to see what you were do- suicides one way or the other (see Chapter 5). His ing. Webb and his colleagues suggest you check the study is an example of unobtrusive research, or wear and tear on the floor in front of various ex- methods of studying social behavior without affect- hibits. Those that have the most-worn tiles are ing it. As you'll see, unobtrusive measures can be probably the most popular. Want to know which qualitative or quantitative. exhibits are popular >vith little kids? Look for mucus on the glass cases. To get a sense of the most popu- TIns chapter examines three types of unobtru- lar radio stations, you could arrange >vith an auto sive research methods: content analysis, analysis of mechanic to check the radio dial settings for cars existing statistics, and comparative and historical re- brought in for repair. search. In content analysis, researchers exanline a class of social artifacts that usually are written docu- The possibilities are limitless. Like a detective ments such as newspaper editorials. The Durkheim investigating a crime, the social researcher looks for study is an example of the analysis of existing statis- clues. If you stop to notice, you'll find that clues of tics. As you'll see, there are great masses of data aU social behavior are all around you. In a sense, every- around you, awaiting your use in the understand- thing you see represents the answer to some im- ing of social life. Finally, comparative and historical portant social scientific question-all you have to research, a form of research with a venerable his- do is think of the question. tory in the social sciences, is currently enjoying a resurgence of popularity. Like field research, com- Although problems of validity and reliability parative and historical research is a usually qualita- crop up in unobtrusive measures, a little ingenuity tive method, one in which the main resources for can either handle them or put them in perspective. observation and analysis are historical records. The I encourage you to look at Webb's book. It's enjoy- method's name includes the word comparative be- able reading, and it can be a source of stimulation cause social scientists-in contrast to historians and insight for social inquiry through data that al- who may simply describe a particular set of ready exist. For now, let's turn our attention to events- seek to discover common patterns that three unobtrusive methods often employed by so- recur in different times and places. cial scientists, beginning >vith content analysis. To set the stage for our examination of these unobtrusive research Methods of studying social three research methods, I want to draw your atten- behavior \\vithout affecting it. Such methods can be tion to an excellent book that should sharpen your qualitative or quantitative.. senses about the potential for unobtrusive mea- sures in general. It is, among other things, the book from whidl I take the term ul10btrusive measures.
320 ill Chapter 11: Unobtrusive Research Content Analysis social scientific research topic: The first might ad- dress national character, the second political orien- As I mentioned in the chapter introduction, tations, and the third political process. Although content analysis is the study of recorded human such topics might be studied by observing individ- communications. Among the forms suitable for ual people, content analysis provides another study are books, magazines, web pages, poems, approach. newspapers, songs, paintings, speeches, letters, e-mail messages, bulletin board postings on the An early example of content analysis is the work Internet, laws, and constitutions, as well as any ofIda R Wells. In 1891, Wells, whose parents had components or collections thereof. Shulamit Rein- been slaves, wanted to test the vvidely held assump- harz points out that feminist researchers have used tion that African American men were being lynched content analysis to study \"children's books, fairy in the South primarily for raping white women. As tales, billboards, feminist nonfiction and fiction a research method, she examined newspaper ar- books, children's art work, fashion, fat-letter post- ticles on the 728 lynchings reported during the pre- cards, Girl Scout Handbooks, works of fine art, vious ten years. In only a third of the cases were the newspaper rhetoric, clinical records, research publi- lynching victims even accused of rape, much less cations, introductory sociology textbooks, and cita- proven guilty. Primarily, they were charged with be- tions, to mention only a few\" (1992: 146-47). In ing insolent, not staying in \"their place\" (cited in another example, when William Mirola set out to Reinharz 1992: 146). discover the role of religion in the movements to establish the eight-hour working day in America, More recently, the best-selling book Megatrends his data were taken \"from Chicago's labor, religious, 2000 (Naisbitt and Aburdene 1990) used content and secular presses, from pamphlets, and from analysis to determine the major trends in modern speeches given by eight-hour proponents from U.S. life. The authors regularly monitored thou- three representative factions within the movement\" sands of local newspapers a month in order to (2003: 273). discover local and regional trends for publication in a series of quarterly reports. Their book exam- Topics Appropriate to Content Analysis ines some of the trends they observed in the nation at large. Content analysis is particularly well suited to the study of communications and to answering the Some topics are more appropriately addressed classic question of communications research: \"Who by content analysis than by any other method of says what to whom, why, how; and with what ef- inquiry Suppose that you're interested in violence fect?\" Are popular French novels more concerned on television. Maybe you suspect that the manufac- with love than novels in the United States are? Was turers of men's products are more likely to sponsor the popular British music of the 1960s more politi- violent TV shows than other kinds of sponsors are. cally cynical than the popular German music dur- Content analysis would be the best way of find- ing that period? Do political candidates who pri- ing out. marily address \"bread and butter\" issues get elected more often than those who address issues of high Briefly, here's what you'd do. First, you'd de- principle? Each of these questions addresses a velop operational definitions of the two key vari- ables in your inquiry: men's products and violence. The content analysis The study of recorded human section on coding, later in this chapter, will discuss communications, such as books, websites, paintings, some of the ways you could do thaL Ultimately, and laws. you'd need a plan that would allow you to watch TV, classify sponsors, and rate the degree of violence on particular shows. Next, you'd have to decide what to watch. Probably you'd decide (1) what stations to watch, (2) for what period, and (3) at what hours. Then, you'd stock up on beer and potato chips and start
Content Analysis ill 321 watching, classifying, and recording. Once you'd analysis. We'll then review some of the sampling completed your observations, you'd be able to ana- techniques that might be applied to them in con- lyze the data you collected and determine whether tent analysis. men's product manufacturers sponsored more blood and gore than other sponsors did. Units olAnalysis Gabriel Rossman (2002) had a somewhat dif- As I discussed in Chapter 4, determining appropri- ferent question regarding the mass media. Public ate units of analysis-the individual units that we concern over the concentration of media in fewer make descriptive and explanatory statements and fewer corporate hands has grown, so Rossman about-can be a complicated task For example, decided to ask the following question: If a newspa- if we vvish to compute the average family income, per is owned by the same conglomerate that owns the individual family is the unit of analysis. But a movie production company, can you trust that we'll have to ask individual members of families newspaper's movie reviews of its parent company's how much money they make. Thus, individuals productions? will be the units of observation, even though the in- dividual family remains the unit of analysis. Simi- You can't, according to Rossman's findings. larly, we may wish to compare crime rates of differ- Because many newspapers rate movies somewhat ent cities in terms of their size, geographic region, quantitatively (e.g., three stars out of four), he was racial composition, and other differences. Even able to perform a simple, quantitative analysis. For though the characteristics of these cities are partly each movie review, he asked two main questions: a function of the behaviors and characteristics of (1) Was the movie produced by the same company their individual residents, the cities would ulti- that owned the newspaper? and (2) What rating mately be the units of analysis. did the film receive? He found that. indeed, movies produced by the parent company received higher The complexity of this issue is often more ap- ratings than other movies did. Further, the ratings parent in content analysis than in other research given to movies by newspapers with the same par- methods, especially when the units of observation ent company were higher than the ratings those differ from the units of analysis. A few examples movies received from other newspapers. This dis- should clarify this distinction. crepancy, moreover, was strongest in the case of big-budget movies in which the parent company Let's suppose we want to find out whether had invested heavily. criminal law or civil law makes the most distinc- tions between men and women. In this instance, As a mode of observation, content analysis re- individual laws would be both the units of obser- quires a thoughtful handling of the \"what\" that is vation and the units of analysis. We might select a being communicated. The analysis of data collected sample of a state's criminal and civil laws and then in this mode, as in others, addresses the \"why\" and categorize each law by whether or not it makes a \"with what effecL\" distinction between men and women. In this fash- ion, we could determine whether criminal or civil Sampling in Content Analysis law distinguishes by gender the most. In the study of communications, as in the study Somewhat differently, we might wish to deter- of people, you often can't observe directly all you mine whether states that enact laws distinguishing would like to explore. In your study of TV violence between different racial groups are also more likely and sponsorship, for example, I'd advise against at- than other states to enact laws distinguishing be- tempting to watch everything that's broadcast. It tween men and women. Although the examination wouldn't be possible, and your brain would proba- of this question would also involve the coding of bly short-circuit before you came close to discover- individual acts of legislation, the unit of analysis in ing that for yourself. Usually, it's appropriate to this case is the individual state, not the law. sample. Let's begin by revisiting the idea of units of Or, changing topics radically, let's suppose we're interested in representationalism in painting. If we
322 Chapter 11: Unobtrusive Research wish to compare the relative popularity of repre- Paragraphs ::::: Lines sentational and nonrepresentational paintings, the individual paintings will be our units of analysis. If, = L1;;::::: on the other hand, we wish to discover whether re- presentationalism in painting is more characteristic '- of wealthy or impoverished painters, of educated or uneducated painters, of capitalist or socialist FIGURE 11-1 painters, the individual painters will be our units AFew Possible Units of Analysis for Content Analysis of analysis. products and the programs by their violence. The It's essential that this issue be clear, because program classifications would be transferred to the commercials occurring near them. Figure 11-2 sample selection depends largely on what the unit provides an example of the kind of record you of analysis is. If individual writers are the units of analysis, the sample design should select all or a might keep. sample of the writers appropriate to the research Notice that in the research design illustrated in questionoIf books are the units of analysis, we should select a sample of books, regardless of their Figure 11-2, all the commercials occurring in the authors. Bruce Berg (1989: 112-13) points out that same program break are grouped and get the same even if you plan to analyze some body of textual scores. Also, the number of violent instances re- materials, the units of analysis might be words, corded as following one commercial break is the themes, characters, paragraphs, items (such as a same as the number preceding the next break. This book or letter), concepts, semantics, or combina- simple design allows us to classify each commercial tions of theseo Figure 11-1 illustrates some of those by its sponsorship and the degree of violence asso- ciated with ie Thus, for example, the first Grunt possibilities. Aftershave commercial is coded as being a men's I'm not suggesting that sampling should be product and as having 10 instances of violence as- sociated with it. The Buttercup Bra commercial is based solely on the units of analysis. Indeed, we may often subsample-select samples of subcate- gories-for each indivi.dual unit of analysis. Thus, if writers are the units of analysis, we might (1) select a sample of writers from the total population of writers, (2) select a sample of books written by each writer selected, and (3) select portions of each se- lected book for observation and coding. Finally, let's look at a trickier example: the study of TV violence and sponsorso What's the unit of analysis for the research question \"Are the man- ufacturers of men's products more likely to sponsor violent shows than other sponsors are?\" Is it the TV show? The sponsor? The instance of violence? In the simplest study design, it would be none of these. Though you might structure your inquiry in various ways, the most straightforvvard design would be based on the commercial as the unit of analysis. You would use two kinds of observational units: the commercial and the program (the show that gets squeezed in between commercials). You would want to observe both units. You would clas- sify commercials by whether they advertised men's
Content Analysis 323 Commercial Men's Number of Instances Break Product? of Violence Sponsor Before the After the Yes No ? Commercial Commercial Break Break 1st yreutr Afters/uw-e, .; 6 4 2nd Brute-JockSh'~ .; 6 4 3rd Baid-No-More\" LotU>n- .; 4 3 .; 3 0 I yreutr Afters/uw-e, 4~ SnD\"\"f'~ TOO~ I .; 3 0 S~ .; 3 0 yotiii.nNs dea,n,ser 1 6~ .;1 0 BU! TIutmb H~s I SIU) .. Too.d..o~ 1- .; 1 0 '/ ..... .; 1 0 'W\"',-\".~...I~-h BU:! TIutmb H~s Bu:tterc.up- BrM .; 0 0I , I II FIGURE 11-2 Example of Recording Sheet for TV Violence coded as not being a men's product and as having casting logs, you won't know when the commer- no violent instances associated with it. cials are going to OCCUL Moreover, you need to ob- serve the programming as well as the commercials. In the illustration, we have four men's product As a result, you must set up a sampling design commercials with an average of 75 violent in- that will include everything you need in order to stances each. The four commercials classified as de- observe enough. finitely not men's products have an average of 1.75, and the two that might or might not be considered In designing the sample, you'd need to estab- men's products have an average of 1 violent instance lish the universe to be sampled fromo In this case, each. If this pattern of differences persisted across a which TV stations will you observe? What will be much larger number of observations, we'd proba- the period of the study-the number of days? bly conclude that manufacturers of men's products And during which hours of each day will you ob- are more likely to sponsor TV violence than other serve? Then, how many commercials do you want sponsors are. to observe and code for analysis? Watch television for a while and find out how many commercials The point of this illustration is to demonstrate occur each hour; then you can figure out how how units of analysis figure into the data collection many hours of observation you'll need (and can and analysiso You need to be clear about your unit stand). of analysis before planning your sampling strategy, but in this case you can't simply sample commer- Now you're ready to design the sample selec- cials. Unless you have access to the stations' broad- tion. As a practical matter, you wouldn't have to
324 €I Chapter 11: Unobtrusive Research sample among the different stations if you had as- Other forms of communication may also be sam- sistants-each of you could watch a different chan- nel during the same period. But let's suppose you're pled at any of the conceptual levels appropriate working alone. Your final sampling frame, from which a sample will be selected and watched, might to them. look something like this: In content analysis, we could employ any of the conventional sampling techniques discussed i~ Chapter 7. We might select a random or systematIc Jan. 7, Channel 2, 7-9 P.M. sample of French and US. novelists, of laws passed Jan. 7, Channel 4, 7-9 P.M. Jan. 7, Channel 9, 7-9 P.M. in the state of Mississippi. or of Shakespearean so- Jan. 7, Channel 2, 9-11 P.M. Jan. 7, Channel 4, 9-11 P.M. liloquies. We might select (with a random start) Jan. 7, Channel 9, 9-11 P.M. every 23rd paragraph in Tolstoy's War and Peace. Or Jan. 8, Channel 2, 7-9 P.l\"\" we rclght number all of the songs recorded by the Jan. 8, Channel 4, 7-9 P.M. Jan. 8, Channel 9, 7-9 P.M. Beatles and select a random sample of 25. Jan. 8, Channel 2, 9-11 P.M. Jan. 8, Channel 4, 9-11 P.;V1. Stratified sampling is also appropriate to con- Jan. 8, Channel 9, 9-11 P.M. Jan. 9, Channel 2, 7-9 P.M. tent analysis. To analyze the editorial policies of Jan. 9, Channel 4, 7-9 P.M. etc. U.S. newspapers, for example, we might first gr~up all newspapers by the region of the country or SIze of the community in which they are published, frequency of publication, or average circulation. Wteemamtii~ahsat mthpelne select a stratified random or sys- of newspapers for analysis. Having done so, we might select a sample of editorials from each selected newspaper, perhaps stratified chronologically. Cluster sampling is equally appropriate to con- Notice that I've made several decisions for you tent analysis. Indeed, if individual editorials are our in the illustration. First, I've assumed that channels units of analysis, then the selection of newspapers 2, 4, and 9 are the ones appropriate to your s:udy at the first stage of sampling would be a cluster I've assumed that you found the 7-11 P.M. pnme- sample. In an analysis of political speech.e:, .we time hours to be the most relevant and that two- might begin by selecting a sample of politlC1ans; hour periods will do the job. I picked January 7 out each politician would represent a cluster of of the hat for a starting date. In practice, of course, political speeches. The TV commercial study de- all these decisions should be based on your careful scribed previously is another example of cluster consideration of what would be appropriate to your sampling. particular study. . It should be repeated that sampling need not Once you have become clear about your umts end when we reach the unit of analysis. If novels of analysis and the observations best suited to those are the unit of analysis in a study, we might select units and have created a sampling frame like the a sample of novelists, a subsample of novels written one I've illustrated, sampling is simple and straight- by each selected author, and a subsample of para- forvvard. The alternative procedures available to you ~alraepchosnwteintht ionf each novel. We would then analyze the paragraphs for the purpose of are the same ones described in Chapter 7: random, systematic, stratified, and so on. describing the novels themselves. (If you haven't realized this yet, researchers speak of samples Sampling Techniques within samples as \"subsamples.\") As we've seen, in the content analysis of written Let's turn now to the coding or classification of prose, sampling may occur at any or all of several levels, including the contexts relevant to the works. the material being observed. Part 4 discusses the manipulation of such classifications to draw de- scriptive and explanatory conclusions.
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
- 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 - 603
Pages:
