SURVEY PRACTICES THAT AAPOR CONDEMNS AAPOR joins the Research Industry Coalition and the National Council on Public Polls in condemning certain misleading practices sometimes performed in the name of research. In no case are the following practices deemed legitimate or acceptable elements of professionally conducted research: 1. Requiring a monetary payment or soliciting monetary contributions from members of the public as part of a research process. This set of practices amounts to fund raising under the guise of research. It takes unfair advantage of the cooperative attitude that a majority of the public manifests when asked to take part in a legitimate information gathering process. In some cases, unwary members of the public are enticed to contribute money as a condition of gaining some future “benefit” from their participation. 2. Offering products or services for sale, or using participant contacts as a means of generating sales leads. A common practice is to gain entry or acceptance in order to make a sales pitch by initially defining the contact as being made for “research” purposes. This trades on the prestige of science, and it exploits the willingness of the public to reveal information about themselves in the public interest. In some cases, questions establish respondents' susceptibility to sales pressure or their interest in some product or service. Follow-up contacts are then made to those so identified, all under the guise of “research.” 3. Revealing the identity of individual respondents to a survey or participants in a research process without their permission. It is normal research practice to pledge anonymity or confidentiality to the public in order to secure their cooperation and frankness in responding to questions. Revealing the identity of individuals, for whatever purpose, is a violation of that pledge unless a respondent’s prior informed consent has been obtained. 4. Representing the results of a 900-number or other type of self-selected “poll” as if they were the outcome of legitimate research. 900-number and other types of write-in, call-in, and interactive polls have become increasingly common. These “polls” report the opinions of only those people who called in, and not those of the general public. AAPOR believes that any publicizing or promotion of such activities not only damages legitimate market and survey research, but can be very misleading when used to influence public policy or simply to disseminate information about the general public. 5. Conducting a so-called “push poll,” a telemarketing technique in which telephone calls are used to canvass potential voters, feeding them false or misleading “information” about a candidate under the pretense of taking a poll to see how this “information” affects voter preferences. So-called “Push polls” are not polls at all. They are a form of political telemarketing whose intent is not to measure public opinion but to manipulate — “push” — voters away from one candidate and toward the opposing candidate. Such polls defame selected candidates by spreading false or misleading information about them. The intent is to disseminate campaign propaganda under the guise of conducting a legitimate public opinion poll. Read AAPOR’s recent statement on “push” polls. As members of AAPOR, a professional organization which relies on public cooperation to gather information that is useful in formulating public policy as well as in understanding the public’s preferences for products and services, we condemn these practices in the strongest terms. FIGURE 3-1 Excerpt from the Code of Conduct of the American Association for Public Opinion Research Source:American Association for Public Opinion Research,By-Laws (2005).The entire code of conduct can be found at the link on this book’s website:http://www .cengage.com/sociology/babbie.
Two Ethical Controversies ■ 75 justice. The first principle, autonomy, requires ics. Social researchers still disagree on some general that subjects be treated with respect as autono- principles, and those who agree in principle often mous agents and affirms that those persons debate specifics. with diminished autonomy are entitled to special protection. In practice, this principle is This section briefly describes two research reflected in the process of informed consent, in projects that have provoked ethical controversy and which the risks and benefits of the research are discussion. The first project studied homosexual disclosed to the subject. The second principle, behavior in public restrooms, and the second ex- beneficence, involves maximizing possible amined obedience in a laboratory setting. benefits and good for the subject, while mini- mizing the amount of possible harm and risks Trouble in the Tearoom resulting from the research. Since the fruits of knowledge can come at a cost to those partici- As a graduate student, Laud Humphreys became pating in research, the last principle, justice, interested in the study of homosexual behavior. seeks a fair distribution of the burdens and He developed a special interest in the casual and benefits associated with research, so that cer- fleeting same-sex acts engaged in by some male tain individuals or groups do not bear dispro- nonhomosexuals. In particular, his research inter- portionate risks while others reap the benefits. est focused on homosexual acts between strangers meeting in the public restrooms in parks, called (Frankel and Siang 1999: 2–3) “tearooms” among homosexuals. The result was the publication in 1970 of Tearoom Trade. Professional Codes of Ethics What particularly interested Humphreys about Ethical issues in social research are both important the tearoom activity was that the participants and ambiguous. For this reason, most of the profes- seemed otherwise to live conventional lives as sional associations of social researchers have created “family men” and accepted members of the and published formal codes of conduct describing community. They did nothing else that might what is considered acceptable and unacceptable qualify them as homosexuals. Thus, it was impor- professional behavior. As one example, Figure 3-1 tant to them that they remain anonymous in presents a portion of the code of conduct of the their tearoom visits. How would you study some- American Association for Public Opinion Research thing like that? (AAOPR), an interdisciplinary research association in the social sciences. Most professional associations Humphreys decided to take advantage of the have such codes of ethics. See, for example, the social structure of the situation. Typically, the American Sociological Association, the American tearoom encounter involved three people: the two Psychological Association, the American Political men actually engaging in the sexual act and a look- Science Association, and so forth. You can find out, called the “watchqueen.” Humphreys began many of these on the associations’ websites. In showing up at public restrooms, offering to serve addition, the Association of Internet Researchers as watchqueen whenever it seemed appropriate. (AoIR) has a code of ethics accessible online. The Because the watchqueen’s payoff was the chance to excerpt presented details several pseudoresearch watch the action, Humphreys was able to conduct practices that are denounced by AAPOR and other field observations as he would in a study of political professional researchers. rallies or jaywalking behavior at intersections. Two Ethical Controversies To round out his understanding of the tearoom trade, Humphreys needed to know something As you may already have guessed, the adoption more about the people who participated. Because and publication of professional codes of conduct the men probably would not have been thrilled have not totally resolved the issue of research eth- about being interviewed, Humphreys developed a different solution. Whenever possible, he noted the license numbers of participants’ cars and tracked down their names and addresses through the
76 ■ Chapter 3: The Ethics and Politics of Social Research police. Humphreys then visited the men at their laboratory. Humphreys’ study was sociological, this homes, disguising himself enough to avoid recogni- one psychological. And whereas Humphreys exam- tion, and announced that he was conducting a ined behavior considered by many to be deviant, survey. In that fashion, he collected the personal the researcher in this study examined obedience information he couldn’t get in the restrooms. and conformity. As you can imagine, Humphreys’ research One of the most unsettling clichés to come out provoked considerable controversy both inside and of World War II was the German soldier’s common outside the social science community. Some critics excuse for atrocities: “I was only following orders.” charged Humphreys with a gross invasion of pri- From the point of view that gave rise to this com- vacy in the name of science. What men did in pub- ment, any behavior—no matter how reprehen- lic restrooms was their own business. Others were sible—could be justified if someone else could be mostly concerned about the deceit involved— assigned responsibility for it. If a superior officer Humphreys had lied to the participants by leading ordered a soldier to kill a baby, the fact of the order them to believe he was only a voyeur-participant. supposedly exempted the soldier from personal Even people who felt that the tearoom participants responsibility for the action. were fair game for observation because they used a public facility protested the follow-up survey. They Although the military tribunals that tried the felt it was unethical for Humphreys to trace the war crime cases did not accept this excuse, social participants to their homes and to interview them researchers and others have recognized the extent under false pretenses. to which this point of view pervades social life. People often seem willing to do things they know Still others justified Humphreys’ research. would be considered wrong, if they can claim that The topic, they said, was worth study. It couldn’t some higher authority ordered them to do it. Such be studied any other way, and they regarded the was the pattern of justification in the 1968 My Lai deceit as essentially harmless, noting that Hum- tragedy of Vietnam, when U.S. soldiers killed more phreys was careful not to harm his subjects by than 300 unarmed civilians—some of them young disclosing their tearoom activities. One result of children—simply because their village, My Lai, was Humphrey’s research was to challenge some of believed to be a Viet Cong stronghold. This sort of the common stereotypes about the participants in justification appears less dramatically in day-to-day anonymous sexual encounters in public places, civilian life. Few would disagree that this reliance showing them to be basically conventional in other on authority exists, yet Stanley Milgram’s study aspects of their lives. (1963, 1965) of the topic provoked considerable controversy. The tearoom trade controversy has never been resolved. It’s still debated, and it probably always To observe people’s willingness to harm others will be, because it stirs emotions and involves when following orders, Milgram brought 40 adult ethical issues people disagree about. What do you men from many different walks of life into a think? Was Humphreys ethical in doing what he laboratory setting designed to create the phenom- did? Are there parts of the research that you believe enon under study. If you had been a subject in the were acceptable and other parts that were not? experiment, you would have had something like (For more on the political and ethical context of the following experience. the “tearoom” research, see the discussion by Joan Sieber at the link on this book’s website: http:// You’ve been informed that you and another www.cengage.com/sociology/babbie.) subject are about to participate in a learning experi- ment. Through a draw of lots, you’re assigned the Observing Human Obedience job of “teacher” and your fellow subject the job of “pupil.” The pupil is led into another room and The second illustration differs from the first in strapped into a chair; an electrode is attached to many ways. Whereas Humphreys’ study involved his wrist. As the teacher, you’re seated in front of participant observation, this study took place in the an impressive electric control panel covered with dials, gauges, and switches. You notice that each
The Politics of Social Research ■ 77 switch has a label giving a different number of in the experiment. As a subject, you wouldn’t volts, ranging from 15 to 315. The switches have actually have been hurting another person, but other labels, too, some with the ominous phrases you would have been led to think you were. The “Extreme-Intensity Shock,” “Danger—Severe experiment was designed to test your willingness Shock,” and “XXX.” to follow orders to the point of presumably killing someone. The experiment runs like this. You read a list of word pairs to the learner and then test his ability Milgram’s experiments have been criticized to match them up. Because you can’t see him, a both methodologically and ethically. On the ethical light on your control panel indicates his answer. side, critics have particularly cited the effects of the Whenever the learner makes a mistake, you’re experiment on the subjects. Many seemed to have instructed by the experimenter to throw one of experienced personally about as much pain as they the switches—beginning with the mildest—and thought they were administering to someone else. administer a shock to your pupil. Through an open They pleaded with the experimenter to let them door between the two rooms, you hear your pupil’s stop giving the shocks. They became extremely up- response to the shock. Then you read another list set and nervous. Some had uncontrollable seizures. of word pairs and test him again. How do you feel about this research? Do you As the experiment progresses, you administer think the topic was important enough to justify ever more intense shocks, until your pupil screams such measures? Would debriefing the subjects be for mercy and begs for the experiment to end. sufficient to ameliorate any possible harm? Can You’re instructed to administer the next shock you think of other ways the researcher might have anyway. After a while, your pupil begins kicking examined obedience? the wall between the two rooms and continues to scream. The implacable experimenter tells you There is a wealth of discussion regarding the to give the next shock. Finally, you read a list and Milgram experiments on the web. Search for ask for the pupil’s answer—but there is no reply “Milgram experiments,” “human obedience experi- whatever, only silence from the other room. The ments,” or “Stanley Milgram.” experimenter informs you that no answer is con- sidered an error and instructs you to administer the The Politics of Social Research next higher shock. This continues up to the “XXX” shock at the end of the series. As I indicated earlier, both ethics and politics hinge on ideological points of view. What is unacceptable What do you suppose you really would have from one point of view will be acceptable from done when the pupil first began screaming? When another. Although political and ethical issues are he began kicking on the wall? Or when he became often closely intertwined, I want to distinguish totally silent and gave no indication of life? You’d between them in two ways. refuse to continue giving shocks, right? And surely the same would be true of most people. First, the ethics of social research deals mostly with the methods employed; political issues tend to So we might think—but Milgram found oth- center on the substance and use of research. Thus, erwise. Of the first 40 adult men Milgram tested, for example, some critics raise ethical objections to nobody refused to continue administering the the Milgram experiments, saying that the methods shocks until they heard the pupil begin kicking the harm the subjects. A political objection would be wall between the two rooms. Of the 40, 5 did so that obedience is not a suitable topic for study, ei- then. Two-thirds of the subjects, 26 of the 40, con- ther because (1) we should not tinker with people’s tinued doing as they were told through the entire willingness to follow orders from higher authority series—up to and including the administration of or (2), from the opposite political point of view, the highest shock. because the results of the research could be used to make people more obedient. As you’ve probably guessed, the shocks were phony, and the “pupil” was a confederate of the experimenter. Only the “teacher” was a real subject
78 ■ Chapter 3: The Ethics and Politics of Social Research The second distinction between the ethical and separated. Explanations of the status quo in society, political aspects of social research is that there are they contend, shade subtly into defenses of that no formal codes of accepted political conduct. Al- same status quo. Simple explanations of the social though some ethical norms have political aspects— functions of, say, discrimination can easily become for example, specific guidelines for not harming justifications for its continuance. By the same subjects clearly relate to Western ideas about the token, merely studying society and its ills without a protection of civil liberties—no one has developed commitment to making society more humane has a set of political norms that all social researchers been called irresponsible. accept. In Chapter 10, we’ll examine participatory action The only partial exception to the lack of politi- research, which is explicitly committed to using cal norms is the generally accepted view that a social research for purposes designed and valued researcher’s personal political orientation should by the subjects of the research. Thus, for example, not interfere with or unduly influence his or her researchers committed to improving the working scientific research. It would be considered improper conditions for workers at a factory would ask the for a researcher to use shoddy techniques or to workers to define the outcomes they would like distort or lie about his or her research as a way of to see and to have a hand in conducting social furthering the researcher’s political views. As you research relevant to achieving the desired ends. The can imagine, however, studies are often enough role of the researchers is to ensure that the workers attacked for allegedly violating this norm. have access to professional research methods. Objectivity and Ideology Quite aside from abstract disagreements about whether social science can or should be value- In Chapter 1, I suggested that social research can free, many have argued about whether particular never be totally objective, because researchers are research undertakings are value-free or whether human and therefore necessarily subjective. As a they represent an intrusion of the researcher’s own collective enterprise, science achieves the equiva- political values. Typically, researchers have denied lent of objectivity through intersubjectivity. That such intrusion, and their denials have then been is, different scientists, having different subjective challenged. Let’s look at some examples of the views, can and should arrive at the same results controversies this issue has produced. when they employ accepted research techniques. Essentially, this will happen to the extent that each Social Research and Race can set personal values and views aside for the duration of the research. Nowhere have social research and politics been more controversially intertwined than in the area The classic statement on objectivity and of racial relations. Social researchers studied the neutrality in social science is Max Weber’s lecture topic for a long time, and the products of the social “Science as a Vocation” ([1925] 1946). In this talk, research have often found their way into practical Weber coined the phrase value-free sociology and politics. A few brief references should illustrate the urged that sociology, like other sciences, needed point. to be unencumbered by personal values if it were to make a special contribution to society. Liberals In 1896, when the U.S. Supreme Court and conservatives alike could recognize the “facts” established the principle of “separate but equal” of social science, regardless of how those facts ac- as a means of reconciling the Fourteenth Amend- corded with their personal politics. ment’s guarantee of equality to African Americans with the norms of segregation, it neither asked for Most social researchers have agreed with this nor cited social research. Nonetheless, it is widely abstract ideal, but not all. Marxist and neo-Marxist believed that the Court was influenced by the writ- scholars, for example, have argued that social sci- ings of William Graham Sumner, a leading social ence and social action cannot and should not be scientist of his era. Sumner was noted for his view that the mores and folkways of a society were rela-
The Politics of Social Research ■ 79 tively impervious to legislation and social planning. in 1966 when James Coleman, a respected sociolo- His view has often been paraphrased as “stateways gist, published the results of a major national study do not make folkways.” Thus, the Court ruled of race and education. Contrary to general agree- that it could not accept the assumption that “social ment, Coleman found little difference in academic prejudices may be overcome by legislation” and performance between African American students denied the wisdom of “laws which conflict with the attending integrated schools and those attending general sentiment of the community” (Blaunstein segregated ones. Indeed, such obvious things as li- and Zangrando 1970: 308). As many a politician braries, laboratory facilities, and high expenditures has said, “You can’t legislate morality.” per student made little difference. Instead, Cole- man reported that family and neighborhood factors When the doctrine of “separate but equal” was had the most influence on academic achievement. overturned in 1954 (Brown v. Board of Education), the new Supreme Court decision was based in part Coleman’s findings were not well received by on the conclusion that segregation had a detrimen- many of the social researchers who had been active tal effect on African American children. In drawing in the civil rights movement. Some scholars criti- that conclusion, the Court cited several sociological cized Coleman’s work on methodological grounds, and psychological research reports (Blaunstein and but many others objected hotly on the grounds Zangrando 1970). that the findings would have segregationist political consequences. The controversy that raged around For the most part, social researchers in this the Coleman report was reminiscent of that pro- century have supported the cause of African voked a year earlier by Daniel Moynihan (1965) in American equality in the United States, and their his critical analysis of the African American family convictions often have been the impetus for their in the United States. Whereas some felt Moynihan research. Moreover, they’ve hoped that their was blaming the victims, others objected to his research will lead to social change. There is no tracing those problems to the legacy of slavery. doubt, for example, that Gunnar Myrdal’s classic two-volume study (1944) of race relations in the Another example of political controversy sur- United States had a significant impact on the topic rounding social research in connection with race of his research. Myrdal amassed a great deal of data concerns IQ scores. In 1969, Arthur Jensen, a Har- to show that the position of African Americans vard psychologist, was asked to prepare an article directly contradicted U.S. values of social and politi- for the Harvard Educational Review examining the cal equality. Further, Myrdal did not attempt to data on racial differences in IQ test results (Jensen hide his own point of view in the matter. (You can 1969). In the article, Jensen concluded that genetic pursue Myrdal’s landmark research further online differences between African Americans and whites by searching for “Gunnar Myrdal” or “An Ameri- accounted for the lower average IQ scores of Afri- can Dilemma.”) can Americans. Jensen became so identified with that position that he appeared on college campuses Many social researchers have become directly across the country discussing it. involved in the civil rights movement, some more radically than others. Given the broad support for Jensen’s research has been attacked on numer- ideals of equality, research conclusions supporting ous methodological bases. Critics charged that the cause of equality draw little or no criticism. much of the data on which Jensen’s conclusion To recognize how solid the general social science was based were inadequate and sloppy—there are position is in this matter, we need only examine a many IQ tests, some worse than others. Similarly, few research projects that have produced conclu- it was argued that Jensen had not taken social-en- sions disagreeing with the predominant ideological vironmental factors sufficiently into account. Other position. social researchers raised still other methodological objections. Most social researchers have— overtly, at least—supported the end of school segregation. Beyond the scientific critique, however, many Thus, an immediate and heated controversy arose condemned Jensen as a racist. Hostile crowds
80 ■ Chapter 3: The Ethics and Politics of Social Research booed him, drowning out his public presentations. biologist Alfred Kinsey and his colleagues published Ironically, Jensen’s reception by several univer- landmark studies of sexual practices of American sity audiences did not differ significantly from the men (1948) and women (1953). Kinsey’s extensive reception received by abolitionists over a century interviewing allowed him to report on frequency before, when the prevailing opinion favored leav- of sexual activity, premarital and extramarital sex, ing the institution of slavery intact. homosexual behavior, and so forth. His studies produced public outrage and efforts to close his Many social researchers limited their objections research institute at Indiana University. to the Moynihan, Coleman, and Jensen research to scientific, methodological grounds. The politi- Although today most people no longer get cal firestorms ignited by these studies, however, worked up about the Kinsey reports, Americans point out how ideology often shows up in matters tend to remain touchy about research on sex. of social research. Although the abstract model of In 1987, the National Institutes of Health (NIH), science is divorced from ideology, the practice of charged with finding ways to combat the AIDS science is not. epidemic, found they needed hard data on con- temporary sexual practices if they were to design To examine a more recent version of the con- effective anti-AIDS programs. Their request for troversy surrounding race and achievement, search research proposals resulted in a sophisticated study the web for differing points of view concerning design by Edward O. Laumann and colleagues. The “The Bell Curve”—sparked by a book with that proposed study focused on the different patterns of title by Richard J. Herrnstein and Charles Murray. sexual activity characterizing different periods of life, and it received rave reviews from the NIH and The controversies relating to research and race their consultants. are hardly over at present, as we saw in the Chap- ter 2 discussion of critical race theory. Enter Senator Jesse Helms (R-North Carolina) and Congressman William Dannemeyer (R-Cali- The Politics of Sexual Research fornia). In 1989, having learned of the Laumann study, Helms and Dannemeyer began a campaign As I indicated earlier, the Laud Humphreys study of to block the study and shift the same amount of tearoom trade raised ethical issues that researchers money to a teen celibacy program. Anne Fausto- still discuss and debate. At the same time, it seems Sterling, a biologist, sought to understand the op- clear that much of the furor raised by the research position to the Laumann study. was related to the subject matter itself. As I have written elsewhere, The surveys, Helms argued, are not really intended “to stop the spread of AIDS. The Laud Humphreys didn’t just study S-E-X but real purpose is to compile supposedly scientific observed and discussed homosexuality. And facts to support the left-wing liberal argument it wasn’t even the caring-and-committed- that homosexuality is a normal, acceptable relationships-between-two-people-who-just- life-style. . . . As long as I am able to stand happen-to-be-of-the-same-sex homosexuality on the floor of the U.S. Senate,” he added, “I but tawdry encounters between strangers am never going to yield to that sort of thing, in public toilets. Only adding the sacrifice of because it is not just another life-style; it is Christian babies could have made this more sodomy.” inflammatory for the great majority of Ameri- cans in 1970. (Fausto-Sterling 1992) (Babbie 2004: 12) Helms won a 66–34 vote in favor of his amend- ment in the U.S. Senate. Although the House of Whereas Humphreys’ research topic proved Representatives rejected the amendment, and it unusually provocative for many, much tamer was dropped in conference committee, government sexuality research has also engendered outcries of public horror. During the 1940s and 1950s, the
The Politics of Social Research ■ 81 funding for the study was put on hold. Laumann census included more city dwellers than rural and his colleagues then turned to the private sector residents. An urban America was something and obtained funding, albeit for a smaller study, new and disturbing, especially to those who from private foundations. Their research results held to the Jeffersonian belief that independent were published in 1994 as The Social Organization of farmers best protected democracy. Among Sexuality. those of this persuasion were rural, conserva- tive congressmen in the South and West. They Politics and the Census saw that reapportionment would shift power to factory-based unions and politically radical There is probably a political dimension to every immigrants concentrated in Northeastern cities. attempt to study human social behavior. Consider Conservatives in Congress blocked reappor- the matter of the U.S. decennial census, mandated tionment, complaining among other things by the Constitution. The original purpose was to that because January 1 was then census day, discover the population sizes of the various states transient agricultural workers were “incor- to determine their proper representation in the rectly” counted in cities rather than on the House of Representatives. Whereas each state gets farms to which they would return in time for two senators, large states get more representatives spring planting. (Census day was later shifted to than small ones do. So what could be simpler? Just April 1, where it has remained.) The arguments count the number of people in each state. dragged out for a decade, and Congress was not reapportioned until after the next census. From the beginning, there was nothing simple about counting heads in a dispersed, national pop- (Prewitt 2003) ulation like the United States. Even the definition of a “person” was anything but straightforward. In more recent years, concern for undercounting A slave, for example, counted as only three- the urban poor has become a political issue. The fifths of a person for purposes of the census. This big cities, which have the most to lose from the un- decreased the representation of the slaveholding dercounting, typically vote Democratic rather than Southern states, though counting slaves as whole Republican, so you can probably guess which party people might have raised the dangerously radiscal supports efforts to improve the counting and which idea that they should be allowed to vote. party is less enthusiastic. By the same token, when social scientists have argued in favor of replacing Further, the logistical problems of counting the attempt at a total enumeration of the popula- people who reside in suburban tract houses, urban tion with modern survey sampling methods (see apartments, college dorms, military barracks, Chapter 7), they have enjoyed more support from farms, cabins in the woods, and illegal housing Democrats, who would stand to gain from such units, as well as counting those who have no place a methodological shift, than from Republicans, to live, has always presented a daunting task. It’s who would stand to lose. Rather than suggesting the sort of challenge social researchers tackle with Democrats support science more than Republicans relish. However, the difficulty of finding the hard- do, this situation offers another example of how to-reach and the techniques created for doing so the political context in which we live and conduct cannot escape the political net. social research often affects that research. Kenneth Prewitt, who directed the Census Politics with a Little “p” Bureau from 1998 to 2001, describes some of the political aspects of counting heads: Social research is often confounded by political ide- ologies, but the “politics” of social research runs far Between 1910 and 1920, there was a massive deeper still. Social research in relation to contested wartime population movement from the rural, Southern states to industrial Northern cities. In 1920, for the first time in American history, the
82 ■ Chapter 3: The Ethics and Politics of Social Research social issues simply cannot remain antiseptically from Washington lured me into casually agree- objective—particularly when differing ideologies ing that I was familiar with a certain professional are pitted against each other in a field of social sci- journal. Unfortunately, the journal did not exist. I ence data. was mortified and suddenly found myself shifting domains. Without really thinking about it, I now The same is true when research is invoked in was less committed to being a friendly Mr. Chips disputes between people with conflicting interests. and more aligned with ninja-professor. I would not For instance, social researchers who have served as be fully satisfied until I, in turn, could mortify the “expert witnesses” in court would probably agree attorney, which I succeeded in doing. that the scientific ideal of a “search for truth” seems hopelessly naive in a trial or lawsuit. Although Even though the civil servants got their cost- expert witnesses technically do not represent either of-living allowance back, I have to admit I was side in court, they are, nonetheless, engaged by also concerned with how I looked in front of the only one side to appear, and their testimony tends courtroom assemblage. I tell you this anecdote to to support the side of the party who pays for their illustrate the personal “politics” of human interac- time. This doesn’t necessarily mean that these wit- tions involving presumably scientific and objective nesses will lie on behalf of their patrons, but the research. We need to realize that as human beings contenders in a lawsuit are understandably social researchers are going to act like human more likely to pay for expert testimony that beings, and we must take this into account in as- supports their case than for testimony that sessing their findings. This recognition does not attacks it. invalidate their research or provide an excuse for rejecting findings we happen to dislike, but it does Thus, as an expert witness, you appear in court need to be taken into account. only because your presumably scientific and honest judgment happens to coincide with the interests of Politics in Perspective the party paying you to testify. Once you arrive in court and swear to tell the truth, the whole truth, Although the ethical and the political dimensions of and nothing but the truth, however, you find your- research are in principle distinct, they do intersect. self in a world foreign to the ideals of objective con- Whenever politicians or the public feel that social templation. Suddenly, the norms are those of win- research is violating ethical or moral standards, ning and losing. As an expert witness, of course, all they’ll be quick to respond with remedies of their you have to lose is your respectability (and perhaps own. Moreover, the standards they defend may not the chance to earn fees as an expert witness in the be those of the research community. Even when future). Still, such stakes are high enough to create researchers support the goals of measures directed discomfort for most social researchers. at the way research is done, the means specified by regulations or legislation can hamstring research. I recall one case in federal court when I was testifying on behalf of some civil service workers Legislators show special concern for research who had had their cost-of-living allowance on children. Although the social research norms (COLA) cut on the basis of research I thought was discussed in this chapter would guard against rather shoddy. I was engaged to conduct more bringing any physical or emotional harm to chil- “scientific” research that would demonstrate the dren, some of the restrictive legislation introduced injustice worked against the civil servants (Babbie from time to time borders on the actions of one 1982: 232– 43). particular western city, which shall remain name- less. In response to concerns that a public school I took the stand, feeling pretty much like a teacher had been playing New Age music in class respected professor and textbook author. In short and encouraging students to meditate, the city order, however, I found I had moved from the council passed legislation stating that no teacher academy to the hockey rink. Tests of statistical significance and sampling error were suddenly less relevant than a slap shot. At one point, an attorney
The Politics of Social Research ■ 83 could do anything that would “affect the minds of firm, personal feelings about and that affect students”! their lives. Moreover, researchers are human beings, and their feelings often surface in their In recent years, the “politicization of science” professional lives. To think otherwise would be has become a particularly hot topic, with charges naive. flung from both sides of the political spectrum. On the one hand, renewed objections to the teach- Second, science manages to proceed in the ing of evolution have coupled with demands for midst of political controversy and hostility. Even the teaching of Intelligent Design (replacing Cre- when researchers get angry and call each other ationism). In many of these regards, science is names, or when the research community comes seen as a threat to religiously-based views, and under attack from the outside, scientific inquiry scientists are sometimes accused of an antireligious persists. Studies are done, reports are published, agenda. and new things are learned. In short, ideological disputes do not bring science to a halt, but they do On the other hand, a statement by the Union of make it more challenging—and exciting. Concerned Scientists (2005), cosigned by thou- sands of scientists, illustrates the concern that the Third, an awareness of ideological consid- concentration of political power in the hands of one erations enriches the study and practice of so- party can threaten the independent functioning of cial research methods. Many of the established scientific research: characteristics of science, such as intersubjectivity, function to cancel out or hold in check our human The United States has an impressive history of shortcomings, especially those we are unaware of. investing in scientific research and respecting Otherwise, we might look into the world and never the independence of scientists. As a result, we see anything but a reflection of our personal biases have enjoyed sustained economic progress and and beliefs. public health, as well as unequaled leader- ship within the global scientific community. Finally, whereas researchers should not let their Recent actions by political appointees, however, own values interfere with the quality and honesty threaten to undermine this legacy by prevent- of their research, this does not mean that research- ing the best available science from informing ers cannot or should not participate in public policy decisions that have serious consequences debates and express both their scientific expertise for our health, safety, and environment. and personal values. You can do scientifically excellent research on racial prejudice, all the while Across a broad range of issues—from child- being opposed to prejudice and saying so. Some hood lead poisoning and mercury emissions would argue that social scientists, because of their to climate change, reproductive health, and scientific expertise in the workings of society, have nuclear weapons—political appointees have an obligation to speak out, rather than leaving that distorted and censored scientific findings that role to politicians, journalists, and talk-show hosts. contradict established policies. In some cases, Herbert Gans (2002) writes of the need for “public they have manipulated the underlying science sociologists”: to align results with predetermined political decisions. A public sociologist is a public intellectual who applies sociological ideas and findings to social I hope you take away four main lessons from (defined broadly) issues about which sociol- this discussion. First, science is not untouched by ogy (also defined broadly) has something to politics. The intrusion of politics and related ideolo- say. Public intellectuals comment on whatever gies is not unique to social research; the natural sci- issues show up on the public agenda; public ences have experienced and continue to experience sociologists do so only on issues to which they similar intrusions. But social science is particularly can apply their sociological insights and linked to social life. Social researchers study things findings. that matter to people—things that people have
84 ■ Chapter 3: The Ethics and Politics of Social Research MAIN POINTS social research inevitably involves the political beliefs of people outside the research community. Introduction • Although most researchers agree that political ori- • In addition to technical, scientific considerations, so- entation should not unduly influence research, in cial research projects are likely to be shaped by ad- practice separating politics and ideology from the ministrative, ethical, and political considerations. conduct of research can be quite difficult. Some researchers maintain that research can and should Ethical Issues in Social Research be an instrument of social action and change. More subtly, a shared ideology can affect the way • What is ethical and unethical in research is ulti- other researchers receive one’s research. mately a matter of what a community of people • Even though the norms of science cannot force agree is right and wrong. individual researchers to give up their personal • Researchers agree that participation in research values, the intersubjective character of science provides a guard against scientific findings being should normally be voluntary. This norm, the product of bias only. however, can conflict with the scientific need for generalizability. KEY TERMS • Researchers agree that research should not harm The following terms are defined in context in the chapter and at the bottom of the page where the those who participate in it, unless they give their term is introduced, as well as in the comprehensive informed consent, thereby willingly and know- glossary at the back of the book. ingly accepting the risks of harm. anonymity debriefing • Whereas anonymity refers to the situation in which confidentiality informed consent even the researcher cannot identify specific PROPOSING SOCIAL RESEARCH: ETHICAL ISSUES information with the individuals it describes, confidentiality refers to the situation in which the If you are actually proposing a research project, you researcher promises to keep information about may be required to submit your proposal to your subjects private. The most straightforward way campus Institutional Review Board (IRB). In that case, to ensure confidentiality is to destroy identifying you will need to inform yourself as to the forms and information as soon as it’s no longer needed. procedures involved locally. The key concern here is the protection of research subjects: avoiding harm, • Many research designs involve a greater or lesser safeguarding their privacy, and the other such topics discussed in this chapter. degree of deception of subjects. Because deceiv- ing people violates common standards of ethical In this section of the proposal, you will discuss behavior, deception in research requires a strong the ethical risks involved in your study and the steps justification—and even then the justification may you will take to avoid them. Perhaps you will prepare be challenged. forms to insure that subjects are aware of and give informed consent to the risks attendant on their partici- • Social researchers have ethical obligations to the pation. The terms “anonymous” and/or “confidential” are likely to appear in your discussion. community of researchers as well as to subjects. These obligations include reporting results fully REVIEW QUESTIONS AND EXERCISES and accurately as well as disclosing errors, limita- tions, and other shortcomings in the research. 1. Consider the following real and hypothetical research situations. What is the ethical component • Professional associations in several disciplines in each example? How do you feel about it? Do you think the procedures described are ultimately publish codes of ethics to guide researchers. These acceptable or unacceptable? You might find codes are necessary and helpful, but they do not resolve all ethical questions. Two Ethical Controversies • Laud Humphreys’ study of “tearoom” encounters and Stanley Milgram’s study of obedience raise ethical issues that are debated to this day. The Politics of Social Research • Social research inevitably has a political and ideolog- ical dimension. Although science is neutral on po- litical matters, scientists are not. Moreover, much
SPSS Exercises ■ 85 discussing some of these situations with class- alcohol use is planned, the researcher decides mates useful. to keep this finding quiet. a. A psychology instructor asks students in an h. To test the extent to which people may try to introductory psychology class to complete save face by expressing attitudes on mat- questionnaires that the instructor will analyze ters they are wholly uninformed about, the and use in preparing a journal article for researcher asks for their attitudes regarding a publication. fictitious issue. b. After a field study of deviant behavior during i. A research questionnaire is circulated among a riot, law enforcement officials demand that students as part of their university registration the researcher identify those people who were packet. Although students are not told they observed looting. Rather than risk arrest as must complete the questionnaire, the hope is an accomplice after the fact, the researcher that they will believe they must—thus ensur- complies. ing a higher completion rate. c. After completing the final draft of a book j. A researcher pretends to join a radical political reporting a research project, the researcher- group in order to study it and is successfully author discovers that 25 of the 2,000 survey accepted as a member of the inner planning interviews were falsified by interviewers. To circle. What should the researcher do if the protect the bulk of the research, the author group makes plans for the following? leaves out this information and publishes the book. (1) A peaceful, though illegal, demonstration d. Researchers obtain a list of right-wing radicals they wish to study. They contact the radicals (2) The bombing of a public building during with the explanation that each has been a time it is sure to be unoccupied selected “at random” from among the general population to take a sampling of “public (3) The assassination of a public official opinion.” 2. Review the discussion of the Milgram experiment e. A college instructor who wants to test the on obedience. How would you design a study to effect of unfair berating administers an hour accomplish the same purpose while avoiding the exam to both sections of a specific course. ethical criticisms leveled at Milgram? Would your The overall performance of the two sections design be equally valid? Would it have the same is essentially the same. The grades of one sec- effect? tion are artificially lowered, however, and the instructor berates the students for perform- 3. Suppose a researcher who is personally in favor ing so badly. The instructor then administers of small families—as a response to the problem the same final exam to both sections and of overpopulation—wants to conduct a survey to discovers that the performance of the unfairly determine why some people want many children berated section is worse. The hypothesis and others don’t. What personal-involvement is confirmed, and the research report is problems would the researcher face, and how published. could she or he avoid them? What ethical issues should the researcher take into account in design- f. In a study of sexual behavior, the investiga- ing the survey? tor wants to overcome subjects’ reluctance to report what they might regard as shameful 4. Using InfoTrac College Edition, search for “in- behavior. To get past their reluctance, subjects formed content” and then narrow your search to are asked, “Everyone masturbates now “research.” Skim the resulting articles and begin and then; about how much do you to identify groups of people for whom informed masturbate?” consent may be problematic—people who may not be able to give it. Suggest some ways in which g. A researcher studying dorm life on campus the problem might be overcome. discovers that 60 percent of the residents regularly violate restrictions on alcohol con- SPSS EXERCISES sumption. Publication of this finding would probably create a furor in the campus com- See the booklet that accompanies your text for munity. Because no extensive analysis of exercises using SPSS (Statistical Package for the Social Sciences). There are exercises offered for each chapter, and you’ll also find a detailed primer on using SPSS.
86 ■ Chapter 3: The Ethics and Politics of Social Research Online Study Resources 3. When you’re finished with your review, take the posttest to confirm that you’re ready to move on If your book came with an access code card, visit to the next chapter. www.cengage.com/login to register. To purchase ac- cess, please visit www.ichapters.com. WEBSITE FOR THE PRACTICE OF SOCIAL RESEARCH 12TH EDITION 1. Before you do your final review of the chapter, take the CengageNOW pretest to help identify the Go to your book’s website at www.cengage.com/ areas on which you should concentrate. You’ll sociology/babbie for tools to aid you in studying find information on this online tool, as well as for your exams. You’ll find Tutorial Quizzes with instructions on how to access all of its great re- feedback, Internet Exercises, Flash Cards, Glossaries, and sources, in the front of the book. Essay Quizzes, as well as InfoTrac College Edition search terms, suggestions for additional reading, Web Links, 2. As you review, take advantage of the CengageNOW and primers for using data-analysis software such as personalized study plan, based on your quiz SPSS. results. Use this study plan with its interactive ex- ercises and other resources to master the material.
This page intentionally left blank
4 Research Design Posing problems properly is often more difficult than answering them. Indeed, a properly 5 Conceptualization, phrased question often seems to answer Operationalization, itself. You may have discovered the answer to and Measurement a question just in the process of making the question clear to someone else. 6 Indexes, Scales, and Typologies Part 2 deals with what should be observed; that is, Part 2 considers the posing of proper scientific 7 The Logic of Sampling questions, the structuring of inquiry. Part 3 will describe some of the specific methods of social science observation. Chapter 4 addresses the beginnings of research. It examines some of the purposes of inquiry, units of analysis, and the reasons scientists get involved in research projects. Chapter 5 deals with the specification of what it is you want to measure—the processes of conceptualiza- tion and operationalization. It looks at some of the terms that you and I use quite casually in everyday life—prejudice, liberalism, happiness, and so forth—and shows how essential it is to clarify what we really mean
PART 2 The Structuring of Inquiry: Quantitative and Qualitative by such terms when we do research. This process of Although such composite measures are constructed dur- clarification is called conceptualization. ing the analysis of data (see Part 4), the raw materials for them must be provided for in the design and execu- Once we clarify what we mean by certain terms, we tion of data collection. can then measure the referents of those terms. The pro- cess of devising steps or operations for measuring what Finally, we’ll look at how social researchers we want to study is called operationalization. Chapter 5 select people or things for observation. Chapter 7, on deals with the topic of operationalization in general, sampling, addresses the fundamental scientific issue paying special attention to the framing of questions for of generalizability. As you’ll see, we can select a few interviews and questionnaires. people or things for observation and then apply what we observe to a much larger group. For example, by To complete the introduction to measurement, Chap- surveying 2,000 U.S. citizens about whom they favor ter 6 breaks with the chronological discussion of how for president of the United States, we can accurately pre- research is conducted. In this chapter, we’ll examine dict how tens of millions will vote. In this chapter, we’ll techniques for measuring variables in quantitative examine techniques that increase the generalizability of research through the combination of several indicators: what we observe. indexes, scales, and typologies. As an example, we might ask survey respondents five different ques- What you learn in Part 2 will bring you to the verge tions about their attitudes toward gender equality and of making controlled social science observations. Part 3 then combine the answers to all five questions into a will then show you how to take that next step. composite measure of gender-based egalitarianism. 89
CHAPTER FOUR Research Design CHAPTER OVERVIEW Introduction The Time Dimension Cross-Sectional Studies Here you’ll see the wide variety of Three Purposes of Research Longitudinal Studies research designs available to social Exploration Approximating Longitudinal researchers as well as how to design Description Studies a study—that is,specifying exactly Explanation Examples of Research Strategies who or what is to be studied when, how,and for what purpose. The Logic of Nomothetic How to Design a Research Project Explanation Getting Started 90 Conceptualization Criteria for Nomothetic Causality Choice of Research Method Nomothetic Causal Analysis and Operationalization Hypothesis Testing Population and Sampling False Criteria for Nomothetic Observations Causality Data Processing Analysis Necessary and Sufficient Causes Application Research Design in Review Units of Analysis Individuals The Research Proposal Groups Elements of a Research Proposal Organizations Social Interactions The Ethics of Research Design Social Artifacts Units of Analysis in Review Faulty Reasoning about Units of Analysis: The Ecological Fallacy and Reductionism CengageNOW for Sociology Use this online tool to help you make the grade on your next exam. After reading this chapter, go to “Online Study Resources” at the end of the chapter for instructions on how to benefit from CengageNOW.
Introduction ■ 91 Introduction ographies of terrorists, case studies of terrorist organizations, case studies on types of terrorism, Science is an enterprise dedicated to “finding out.” case studies on particular terrorist incidents, and No matter what you want to find out, though, case studies of terrorism in selected regions and there will likely be a great many ways of doing it. countries” (2004: 27). Quantitative researchers, That’s true in life generally. Suppose, for example, on the other hand, have addressed terrorism in a that you want to find out whether a particular au- variety of ways, including analyses of media cover- tomobile—say, the new Burpo-Blasto—would be age, statistical modeling of terrorist events, and the a good car for you. You could, of course, buy one use of various databases relevant to the topic. As and find out that way. Or you could talk to a lot of you’ll see in this chapter, any research topic can be B-B owners or to people who considered buying approached from many different directions. Each of one but didn’t. You might check the classified ads the topics we’ll examine is relevant to both qualita- to see if there are a lot of B-Bs being sold cheap. tive and quantitative studies, though some topics You could read a consumer magazine evaluation may be more relevant to one than to the other of Burpo-Blastos. A similar situation occurs in approach. scientific inquiry. This chapter provides a general introduction Ultimately, scientific inquiry comes down to to research design, whereas the other chapters in making observations and interpreting what you’ve Part 2 elaborate on specific aspects of it. In prac- observed, the subjects of Parts 3 and 4 of this book. tice, all aspects of research design are interrelated. Before you can observe and analyze, however, you As you read through Part 2, the interrelationships need a plan. You need to determine what you’re among parts will become clearer. going to observe and analyze: why and how. That’s what research design is all about. We’ll start by briefly examining the main pur- poses of social research. Then, we’ll consider units Although the details vary according to what of analysis—the what or whom you want to study. you wish to study, you face two major tasks in any Next we’ll consider ways of handling time in social research design. First, you must specify as clearly research, or how to study a moving target that as possible what you want to find out. Second, you changes over time. must determine the best way to do it. Interestingly, if you can handle the first consideration fully, you’ll With these ideas in hand, we’ll turn to how probably handle the second in the same process. to design a research project. This overview of the As mathematicians say, a properly framed question research process serves two purposes: Besides de- contains the answer. scribing how you might go about designing a study, it provides a map of the remainder of this book. Let’s say you’re interested in conducting social research on terrorism. When Jeffrey Ross (2004) Next, we’ll look at the elements of research addressed this issue, he found the existing studies proposals. Often, you’ll need to detail your inten- used a variety of qualitative and quantitative tions before you actually conduct your research; approaches. Qualitative researchers, for example, this might be required in order to obtain funding generated original data through for a major project or perhaps to get your instruc- tor’s approval for a class project. You’ll see that the Autobiographies research proposal provides an excellent opportu- nity for you to consider all aspects of your research Incident Reports and Accounts in advance. Also, this section should help you with the end-of-chapter exercise concerning the Hostages’ Experiences with Terrorists research proposal, if you are doing that. Finally, the last section of this chapter focuses on the ethical Firsthand Accounts of Implementing Policies dimension of research design. Ross goes on to discuss some of the secondary materials used by qualitative researchers: “bi-
92 ■ Chapter 4: Research Design Three Purposes of Research group discussions. This technique is frequently used in market research; we’ll examine it further in Social research can serve many purposes. Three Chapter 10. of the most common and useful purposes are ex- ploration, description, and explanation. Although Exploratory studies are most typically done a given study can have more than one of these pur- for three purposes: (1) to satisfy the researcher’s poses—and most do—examining them separately curiosity and desire for better understanding, (2) to is useful because each has different implications for test the feasibility of undertaking a more extensive other aspects of research design. study, and (3) to develop the methods to be em- ployed in any subsequent study. Exploration A while back, for example, I became aware of Much of social research is conducted to explore the growing popularity of something called “chan- a topic, that is, to start to familiarize a researcher neling,” in which a person known as a channel or with that topic. This approach typically occurs medium enters a trance state and begins speaking when a researcher examines a new interest or with a voice that claims it originates outside the when the subject of study itself is relatively new. channel. Some of the voices say they come from a spirit world of the dead, some say they are from As an example, let’s suppose that widespread other planets, and still others say they exist in taxpayer dissatisfaction with the government erupts dimensions of reality difficult to explain in ordinary into a taxpayers’ revolt. People begin refusing to human terms. pay their taxes, and they organize themselves around that issue. You might like to learn more The channeled voices, often referred to as enti- about the movement: How widespread is it? What ties, sometimes use the metaphor of radio or televi- levels and degrees of support are there within the sion for the phenomenon they represent. “When community? How is the movement organized? you watch the news,” one told me in the course What kinds of people are active in it? An explor- of an interview, “you don’t believe Dan Rather is atory study could help you find at least approxi- really inside the television set. The same is true of mate answers to some of these questions. You me. I use this medium’s body the way Dan Rather might check figures with tax-collecting officials, uses your television set.” collect and study the literature of the movement, attend meetings, and interview leaders. The idea of channeling interested me from several perspectives, not the least of which was Exploratory studies are also appropriate for the methodological question of how to study more persistent phenomena. Suppose you’re un- scientifically something that violates so much happy with your college’s graduation requirements of what we take for granted, including scientific and want to help change them. You might study staples such as space, time, causation, and the history of such requirements at the college and individuality. meet with college officials to learn the reasons for the current standards. You could talk to several Lacking any rigorous theory or precise expecta- students to get a rough idea of their sentiments tions, I merely set out to learn more. Using some on the subject. Though this last activity would not of the techniques of qualitative field research dis- necessarily yield an accurate picture of student cussed in Chapter 10, I began amassing information opinion, it could suggest what the results of a more and forming categories for making sense of what I extensive study might be. observed. I read books and articles about the phe- nomenon and talked to people who had attended Sometimes exploratory research is pursued channeling sessions. I then attended channeling through the use of focus groups, or guided small- sessions myself, observing those who attended as well as the channel and entity. Next, I conducted personal interviews with numerous channels and entities.
Three Purposes of Research ■ 93 In most interviews, I began by asking the Exploratory studies are quite valuable in social human channels questions about how they first science research. They’re essential whenever a re- began channeling, what it was like, and why they searcher is breaking new ground, and they almost continued, as well as standard biographical ques- always yield new insights into a topic for research. tions. The channel would then go into a trance, Exploratory studies are also a source of grounded whereby the interview continued with the entity theory, as discussed in Chapter 2. speaking. “Who are you?” I might ask. “Where do you come from?” “Why are you doing this?” “How The chief shortcoming of exploratory studies can I tell if you are real or a fake?” Although I went is that they seldom provide satisfactory answers into these interview sessions with several questions to research questions, though they can hint at the prepared in advance, each of the interviews fol- answers and can suggest which research methods lowed whatever course seemed appropriate in light could provide definitive ones. The reason explor- of the answers given. atory studies are seldom definitive in themselves has to do with representativeness; that is, the This example of exploration illustrates where people you study in your exploratory research may social research often begins. Whereas research- not be typical of the larger population that inter- ers working from deductive theories have the key ests you. Once you understand representativeness, variables laid out in advance, one of my first tasks you’ll be able to know whether a given exploratory was to identify some of the possibly relevant vari- study actually answered its research problem or ables. For example, I noted a channel’s gender, age, only pointed the way toward an answer. (Repre- education, religious background, regional origins, sentativeness is discussed at length in Chapter 7.) and previous participation in things metaphysical. I chose most of these variables because they com- Description monly affect behavior. A major purpose of many social science studies is I also noted differences in the circumstances to describe situations and events. The researcher of channeling sessions. Some channels said they observes and then describes what was observed. must go into deep trances, some use light trances, Because scientific observation is careful and delib- and others remain conscious. Most sit down while erate, however, scientific descriptions are typically channeling, but others stand and walk about. Some more accurate and precise than casual ones are. channels operate under pretty ordinary conditions; others seem to require props such as dim lights, The U.S. Census is an excellent example of incense, and chanting. Many of these differences descriptive social research. The goal of the cen- became apparent to me only in the course of my sus is to describe accurately and precisely a wide initial observations. variety of characteristics of the U.S. population, as well as the populations of smaller areas such as Regarding the entities, I have been interested states and counties. Other examples of descrip- in classifying where they say they come from. tive studies are the computation of age-gender Over the course of my interviews, I’ve developed profiles of populations done by demographers, a set of questions about specific aspects of the computation of crime rates for different cities, “reality,” attempting to classify the answers they and a product-marketing survey that describes the give. Similarly, I ask each to speak about future people who use, or would use, a particular product. events. A researcher who carefully chronicles the events that take place on a labor union picket line has, or Over the course of this research, my exami- at least serves, a descriptive purpose. A researcher nation of specific topics has become increasingly who computes and reports the number of times focused as I’ve identified variables that seem worth individual legislators voted for or against organized pursuing: gender, education, and religion, for ex- labor also fulfills a descriptive purpose. ample. Note, however, that I began with a reason- ably blank slate.
94 ■ Chapter 4: Research Design Many qualitative studies aim primarily at juana legalization. Further study of gender and description. An anthropological ethnography, for political orientation might then lead to a deeper example, may try to detail the particular culture explanation of these attitudes. of some preliterate society. At the same time, such studies are seldom limited to a merely descriptive The Logic purpose. Researchers usually go on to examine of Nomothetic Explanation why the observed patterns exist and what they imply. The preceding examination of what factors might cause attitudes about legalizing marijuana il- Explanation lustrates nomothetic explanation, as discussed in Chapter 1. Recall that in this model, we try to find The third general purpose of social science re- a few factors (independent variables) that can search is to explain things. Descriptive studies account for many of the variations in a given answer questions of what, where, when, and how; phenomenon. This explanatory model stands in explanatory questions, of why. So when William contrast to the idiographic model, in which we Sanders (1994) set about describing the varieties seek a complete, in-depth understanding of a of gang violence, he also wanted to reconstruct the single case. process that brought about violent episodes among the gangs of different ethnic groups. In our example, an idiographic approach would suggest all the reasons that one person was op- Reporting the voting intentions of an electorate posed to legalization—involving what her parents, is descriptive, but reporting why some people plan teachers, and clergy told her about it; any bad to vote for Candidate A and others for Candidate experiences experimenting with it; and so forth. B is explanatory. Identifying variables that explain When we understand something idiographically, why some cities have higher crime rates than oth- we feel we really understand it. When we know ers involves explanation. A researcher who sets all the reasons why someone opposed legalizing out to know why an antiabortion demonstration marijuana, we couldn’t imagine that person having ended in a violent confrontation with police, as op- any other attitude. posed to simply describing what happened, has an explanatory purpose. In contrast, a nomothetic approach might sug- gest that overall political orientations account for Let’s look at a specific case. What factors do much of the difference of opinion about legalizing you suppose might shape people’s attitudes toward marijuana. Because this model is inherently proba- the legalization of marijuana? To answer this, you bilistic, it is more open than the idiographic model might first consider whether men and women dif- to misunderstanding and misinterpretation. Let’s fer in their opinions. An explanatory analysis of the examine what social researchers mean when they 2002 General Social Survey (GSS) data indicates say one variable (nomothetically) causes another. that 38 percent of men and 30 percent of women Then, we’ll look at what they don’t mean. said marijuana should be legalized. Criteria for Nomothetic Causality What about political orientation? The GSS data show that 55 percent of liberals said marijuana There are three main criteria for nomothetic causal should be legalized, compared with 29 percent of relationships in social research: (1) the variables moderates and 27 percent of conservatives. Fur- must be correlated, (2) the cause takes place before ther, 41 percent of Democrats, compared with the effect, and (3) the variables are nonspurious. 34 percent of Independents and 28 percent of Re- publicans, supported legalization. Given these statistics, you might begin to de- velop an explanation for attitudes toward mari-
The Logic of Nomothetic Explanation ■ 95 Correlation The third variable at work here is season or tempera- ture. Most drowning deaths occur during sum- Unless some actual relationship— or correla- mer—the peak period for ice-cream sales. tion—is found between two variables, we can’t say that a causal relationship exists. Our analysis of Here are a couple of other examples of spuri- GSS data suggested that political orientation was a ous relationships, or ones that aren’t genuine. cause of attitudes about legalizing marijuana. Had There is a negative relationship between the num- the same percentage of liberals and conservatives ber of mules and the number of Ph.D.’s in towns supported legalization, we could hardly say that and cities: the more mules, the fewer Ph.D.’s and political orientations caused the attitude. Though vice versa. Perhaps you can think of another vari- this criterion is obvious, it emphasizes the need to able that would explain this apparent relationship. base social research assertions on actual observa- The answer is rural versus urban settings. There tions rather than assumptions. are more mules (and fewer Ph.D.’s) in rural areas, whereas the opposite is true in cities. Time Order Or, consider the positive correlation between Next, we can’t say a causal relationship exists unless shoe size and math ability among schoolchildren. the cause precedes the effect in time. Notice that it Here, the third variable that explains the puzzling makes more sense to say that most children’s reli- relationship is age. Older children have bigger feet gious affiliations are caused by those of their par- and more highly developed math skills, on average, ents than to say that parents’ affiliations are caused than younger children do. See Figure 4-1 for an by those of their children—even though it would illustration of this spurious relationship. Notice that be possible for you to change your religion and for observed associations go in both directions. That your parents to follow suit. Remember, nomothetic is, as one variable occurs or changes, so does the explanation deals with “most cases” but not all. other. In our marijuana example, it would make The list goes on. Areas with many storks have sense to say that gender causes, to some extent, at- high birthrates. Those with few storks have low titudes toward legalization, whereas it would make birthrates. Do storks really deliver babies? Birth- no sense to say that opinions about marijuana rates are higher in the country than in the city; determine a person’s gender. Notice, however, that more storks live in the country than the city. The the time order connecting political orientations and third variable here is urban/rural areas. attitudes about legalization is less clear, though we sometimes reason that general orientations cause Finally, the more fire trucks that put out a fire, specific opinions. And sometimes our analyses the more damage to the structure. Can you guess involve two or more independent variables that what the third variable is? In this case, it’s the size were established at the same time: looking at the of the fire. effects of gender and race on voting behavior, for example. As we’ll see in the next chapter, the issue Thus, when social researchers say there is a of time order can be a complex matter. causal relationship between, say, education and racial tolerance, they mean (1) there is a statistical Nonspuriousness correlation An empirical relationship between two The third requirement for a causal relationship variables such that (1) changes in one are associated is that the effect cannot be explained in terms of with changes in the other or (2) particular attributes some third variable. For example, there is a cor- of one variable are associated with particular attri- relation between ice-cream sales and deaths due butes of the other. Correlation in and of itself does to drowning: the more ice cream sold, the more not constitute a causal relationship between the two drownings, and vice versa. There is, however, no variables, but it is one criterion of causality. direct link between ice cream and drowning. spurious relationship A coincidental statistical correlation between two variables, shown to be caused by some third variable.
96 ■ Chapter 4: Research Design FIGURE 4-1 An Example of a Spurious Causal Relationship. Finding an empirical correlation between two variables does not necessarily estab- lish a causal relationship. Sometimes the observed correlation is the incidental result of other causal relationships, involving other variables. correlation between the two variables, (2) a per- within the study design you are using. Often this son’s educational level occurred before their current specification will take the form of a level of statisti- level of tolerance or prejudice, and (3) there is no cal significance: the chance you are willing to take third variable that can explain away the observed that a given relationship might have been caused correlation as spurious. by chance in the selection of subjects for study. (This will be discussed further in Chapter 7, on Nomothetic Causal Analysis sampling.) and Hypothesis-Testing Finally, you may specify the tests for spurious- The nomothetic model of causal analysis lends itself ness that any observed relationship must survive. to hypothesis-testing (see Chapter 1), though hy- Not only will you hypothesize, for example, that potheses are not required in nomothetical research. increased education will reduce levels of prejudice, To test a hypothesis, you would carefully specify but you will specify further that the hypothesized the variables you think are causally related, as well relationship will not be the product of, say, political as specifying the manner in which you will mea- orientations. sure them. (These steps will be discussed in detail in the following chapter under the terms conceptual- False Criteria for ization and operationalization.) Nomothetic Causality In addition to hypothesizing that two vari- Because notions of cause and effect are well ables will be correlated with each other, you may entrenched in everyday language and logic, it’s specify the strength of the relationship you expect important to specify some of the things social
Necessary and Sufficient Causes ■ 97 researchers do not mean when they speak of causal Necessary relationships. When they say that one variable and Sufficient Causes causes another, they do not necessarily mean to suggest complete causation, to account for excep- A necessary cause represents a condition that must be tional cases, or to claim that the causation exists in present for the effect to follow. For example, it is a majority of cases. necessary for you to take college courses in order to get a degree. Take away the courses, and the degree Complete Causation never follows. However, simply taking the courses is not a sufficient cause of getting a degree. You need Whereas an idiographic explanation of causation to take the right ones and pass them. Similarly, is relatively complete, a nomothetic explanation being female is a necessary condition of becoming is probabilistic and usually incomplete. As we’ve pregnant, but it is not a sufficient cause. Other- seen, social researchers may say that political ori- wise, all women would get pregnant. Figure 4-2 entations cause attitudes toward legalizing mari- illustrates this relationship between the variables of juana even though not all liberals approve nor all gender and pregnancy as a matrix showing the pos- conservatives disapprove. Thus, we say that political sible outcomes of combining these variables. orientation is one of the causes of the attitude, but not the only one. A sufficient cause, on the other hand, represents a condition that, if it is present, guarantees the effect Exceptional Cases in question. This is not to say that a sufficient cause is the only possible cause of a particular effect. For In nomothetic explanations, exceptions do not dis- example, skipping an exam in this course would prove a causal relationship. For example, it is con- be a sufficient cause for failing it, though students sistently found that women are more religious than could fail it other ways as well. Thus, a cause can be men in the United States. Thus, gender may be a sufficient, but not necessary. Figure 4-3 illustrates cause of religiosity, even if your uncle is a religious the relationship between taking or not taking the zealot or you know a woman who is an avowed exam and either passing or failing it. atheist. Those exceptional cases do not disprove the overall, causal pattern. The discovery of a cause that is both neces- sary and sufficient is, of course, the most satisfying Majority of Cases outcome in research. If juvenile delinquency were the effect under examination, it would be nice to Causal relationships can be true even if they don’t discover a single condition that (1) must be present apply in a majority of cases. For example, we say for delinquency to develop and (2) always results that children who are not supervised after school in delinquency. In such a case, you would surely are more likely to become delinquent than those feel that you knew precisely what caused juvenile who are supervised are; hence, lack of supervision delinquency. is a cause of delinquency. This causal relationship holds true even if only a small percentage of those Unfortunately, we never discover single causes not supervised become delinquent. As long as they that are absolutely necessary and absolutely are more likely than those who are supervised to be sufficient when analyzing the nomothetic re- delinquent, we say there is a causal relationship. lationships among variables. It is not uncommon, however, to find causal factors that are either The social science view of causation may vary 100 percent necessary (you must be female to be- from what you are accustomed to, because people come pregnant) or 100 percent sufficient (skipping commonly use the term cause to mean something an exam will inevitably cause you to fail it). that completely causes another thing. The some- what different standard used by social researchers In the idiographic analysis of single cases, you can be seen more clearly in terms of necessary and may reach a depth of explanation from which sufficient causes. it is reasonable to assume that things could not
98 ■ Chapter 4: Research Design FIGURE 4-2 Necessary Cause. Being female is a necessary cause of pregnancy; that is, you can’t get pregnant unless you are female. experiences would have ended up going to col- lege.) At the same time, there could always be other causal paths to the same result. Thus, the idiographic causes are sufficient but not necessary. FIGURE 4-3 Units of Analysis Sufficient Cause. Not taking the exam is a sufficient cause of failing it, even though there are other ways of failing (such as In social research, there is virtually no limit to what answering randomly). or whom can be studied, or the units of analysis. This topic is relevant to all forms of social research, have turned out differently, suggesting you have although its implications are clearest in the case of determined the sufficient causes for a particular nomothetic, quantitative studies. result. (Anyone with all the same details of your genetic inheritance, upbringing, and subsequent The idea for units of analysis may seem slippery at first, because research—especially nomothetic units of analysis The what or whom being stud- research— often studies large collections of people ied. In social science research, the most typical units or things, or aggregates. It’s important to distinguish of analysis are individual people. between the unit of analysis and the aggregates that we generalize about. For instance, a researcher may study a class of people, such as Democrats, college undergraduates, African American women under 30, or some other collection. But if the researcher is interested in exploring, describing, or explaining how different groups of individuals behave as individuals, the unit of analysis is the indi- vidual, not the group. This is true even though the researcher uses the information about individuals
Units of Analysis ■ 99 to generalize about aggregates of individuals, as in Political Science 110 than students with poor study saying that more Democrats than Republicans favor habits did. You would operationalize the variable legalizing marijuana. Think of it this way: Having study habits and measure this variable, perhaps in an attitude about marijuana is something that can terms of hours of study per week. You might then only be an attribute of an individual, not a group; aggregate students with good study habits and that is, there is no one group “mind” that can have those with poor study habits and see which group an attitude. So even when we generalize about received the best grades in the course. The purpose Democrats, we’re generalizing about an attribute of the study would be to explain why some groups they possess as individuals. of students do better in the course than others do, but the unit of analysis is still individual students. In contrast, we may sometimes want to study groups, considered as individual “actors” or entities Units of analysis in a study are usually also the that have attributes as groups. For instance, we units of observation. Thus, to study success in a po- might want to compare the characteristics of dif- litical science course, we would observe individual ferent types of street gangs. In that case our unit of students. Sometimes, however, we “observe” our analysis would be gangs (not members of gangs), units of analysis indirectly. For example, suppose and we might proceed to make generalizations we want to find out whether disagreements about about different types of gangs. For example, we the death penalty tend to cause divorce. In this might conclude that male gangs are more violent case, we might “observe” individual husbands and than female gangs. Each gang (unit of analysis) wives by asking them about their attitudes about would be described in terms of two variables: capital punishment, in order to distinguish couples (1) What sex are the members? and (2) How who agree and disagree on this issue. In this case, violent are its activities? So we might study 52 our units of observation are individual wives and gangs, reporting that 40 were male and 12 were husbands, but our units of analysis (the things we female, and so forth. The “gang” would be the unit want to study) are couples. of analysis, even though some of the characteristics were drawn from the components (members) of Units of analysis, then, are those things we the gangs. examine in order to create summary descriptions of all such units and to explain differences among Social researchers tend to choose individual them. In most research projects, the unit of analysis people as their units of analysis. You may note will probably be clear to you. When the unit of the characteristics of individual people—gender, analysis is not clear, however, it’s essential to deter- age, region of birth, attitudes, and so forth. You mine what it is; otherwise, you cannot determine can then combine these descriptions to provide what observations are to be made about whom a composite picture of the group the individuals or what. represent, whether a street-corner gang or a whole society. Some studies try to describe or explain more than one unit of analysis. In these cases, the For example, you may note the age and gender researcher must anticipate what conclusions she of each student enrolled in Political Science 110 or he wishes to draw with regard to which units and then characterize the group of students as of analysis. For example, we may want to discover being 53 percent men and 47 percent women and what kinds of college students (individuals) are as having a mean age of 18.6 years. Although the most successful in their careers; we may also want final description would be of the class as a whole, to learn what kinds of colleges (organizations) pro- the description is based on characteristics that duce the most-successful graduates. members of the class have as individuals. Here’s an example that illustrates the complex- The same distinction between units of analy- ity of units of analysis. Murder is a fairly personal sis and aggregates occurs in explanatory studies. matter: One individual kills another individual. Suppose you wished to discover whether students However, when Charis Kubrin and Ronald Weitzer with good study habits received better grades in (2003: 157) ask, “Why do these neighborhoods
100 ■ Chapter 4: Research Design generate high homicide rates?” the unit of analysis aggregating and manipulating the descriptions of in that phrase is neighborhood. You can probably individuals. imagine some kinds of neighborhoods (e.g., poor, urban) that would have high homicide rates and Any type of individual may be the unit of some (e.g., wealthy, suburban) that would have analysis for social research. This point is more low rates. In this particular conversation, the unit important than it may seem at first. The norm of of analysis (neighborhood) would be categorized in generalized understanding in social research should terms of variables such as economic level, locale, and suggest that scientific findings are most valuable homicide rate. when they apply to all kinds of people. In practice, however, social researchers seldom study all kinds In their analysis, however, Kubrin and Weitzer of people. At the very least, their studies are typi- were also interested in different types of homicide: cally limited to the people living in a single country, in particular, those that occurred in retaliation for though some comparative studies stretch across some earlier event, such as an assault or insult. Can national boundaries. Often, though, studies are you identify the unit of analysis common to all of quite circumscribed. the following excerpts? Examples of classes of individuals that might 1. The sample of killings . . . be chosen for study include students, gays and lesbians, auto workers, voters, single parents, and 2. The coding instrument includes over 80 items faculty members. Note that each of these terms related to the homicide. implies some population of individuals. Descriptive studies with individuals as their units of analy- 3. Of the 2,161 homicides that occurred from sis typically aim to describe the population that 1985 [to] 1995 . . . comprises those individuals, whereas explanatory studies aim to discover the social dynamics operat- 4. Of those with an identified motive, 19.5 ing within that population. percent (n ϭ 337) are retaliatory. (Kubrin and Weitzer 2003: 163) As the units of analysis, individuals may be characterized in terms of their membership in In each of these excerpts, the unit of analysis social groupings. Thus, an individual may be is homicide (also called killing or murder). Some- described as belonging to a rich family or to a poor times you can identify the unit of analysis in the one, or a person may be described as having a description of the sampling methods, as in the college-educated mother or not. We might examine first excerpt. A discussion of classification methods in a research project whether people with college- might also identify the unit of analysis, as in the educated mothers are more likely to attend college second excerpt (80 ways to code the homicides). than are those with non-college-educated mothers Often, numerical summaries point the way: 2,161 or whether high school graduates in rich families homicides; 19.5 percent (of the homicides). With a are more likely than those in poor families to at- little practice you’ll be able to identify the units of tend college. In each case, the unit of analysis—the analysis in most social research reports, even when “thing” whose characteristics we are seeking to more than one is used in a given analysis. describe or explain—is the individual. We then ag- gregate these individuals and make generalizations To explore this topic in more depth, let’s con- about the population they belong to. sider several common units of analysis in social research. Individuals Groups As mentioned, individual human beings are Social groups can also be units of analysis in social perhaps the most typical units of analysis for research. That is, we may be interested in charac- social research. Social researchers tend to describe teristics that belong to one group, considered as and explain social groups and interactions by
Units of Analysis ■ 101 a single entity. If you were to study the members study of street gangs, you might discover whether of a criminal gang to learn about criminals, the large gangs are more likely than small ones to en- individual (criminal) would be the unit of analysis; gage in intergang warfare. Thus, you would arrive but if you studied all the gangs in a city to learn the at conclusions about the population of gangs by differences, say, between big gangs and small ones, using individual groups as your unit of analysis. between “uptown” and “downtown” gangs, and so forth, you would be interested in gangs rather than Organizations their individual members. In this case, the unit of analysis would be the gang, a social group. Formal social organizations may also be the units of analysis in social research. For example, a Here’s another example. Suppose you were researcher might study corporations, by which interested in the question of access to computers he or she implies a population of all corporations. in different segments of society. You might describe Individual corporations might be characterized in families in terms of total annual income and ac- terms of their number of employees, net annual cording to whether or not they had computers. profits, gross assets, number of defense contracts, You could then aggregate families and describe the percentage of employees from racial or ethnic mean income of families and the percentage with minority groups, and so forth. We might determine computers. You would then be in a position to whether large corporations hire a larger or smaller determine whether families with higher incomes percentage of minority group employees than small were more likely to have computers than were corporations do. Other examples of formal social those with lower incomes. In this case, the unit of organizations suitable as units of analysis include analysis would be families. church congregations, colleges, army divisions, academic departments, and supermarkets. As with other units of analysis, we can derive the characteristics of social groups from those of Figure 4-4 provides a graphic illustration of their individual members. Thus, we might describe some different units of analysis and the statements a family in terms of the age, race, or education of that might be made about them. its head. In a descriptive study, we might find the percentage of all families that have a college- Social Interactions educated head of family. In an explanatory study, we might determine whether such families have, Sometimes social interactions are the relevant on average, more or fewer children than do fami- units of analysis. Instead of individual humans, lies headed by people who have not graduated you can study what goes on between them: tele- from college. In each of these examples, the family phone calls, kisses, dancing, arguments, fistfights, is the unit of analysis. In contrast, had we asked e-mail exchanges, chat-room discussions, and so whether college-educated individuals have more or forth. As you saw in Chapter 2, social interac- fewer children than do their less-educated counter- tion is the basis for one of the primary theoretical parts, then the individual would have been the unit paradigms in the social sciences, and the number of of analysis. units of analysis that social interactions provide is nearly infinite. Other units of analysis at the group level could be friendship cliques, married couples, census Even though individuals are usually the actors blocks, cities, or geographic regions. As with indi- in social interactions, there is a difference between viduals, each of these terms implies some popula- (1) comparing the kinds of people who subscribe tion. Street gangs implies some population that in- to different Internet service providers (individuals cludes all street gangs, perhaps in a given city. You being the units of analysis) and (2) comparing the might then describe this population by generalizing length of chat-room interactions on those same from your findings about individual gangs. For in- ISPs (interactions being the units of analysis). stance, you might describe the geographic distribu- tion of gangs throughout a city. In an explanatory
FIGURE 4-4 Illustrations of Units of Analysis. Units of analysis in social research can be individuals, groups, or even nonhuman entities.
Units of Analysis ■ 103 Social Artifacts consciousness, taking paintings as the units of anal- ysis and describing each, in part, by the nationality Another unit of analysis is the social artifact, or of its creator. Or you might examine a newspaper’s any product of social beings or their behavior. One editorials regarding a local university, for the pur- class of artifacts includes concrete objects such as pose of describing, or perhaps explaining, changes books, poems, paintings, automobiles, buildings, in the newspaper’s editorial position on the univer- songs, pottery, jokes, student excuses for missing sity over time. In this example, individual editorials exams, and scientific discoveries. would be the units of analysis. See “How to Do It: Identifying the Unit of Analysis” for more. For example, Lenore Weitzman and her associ- ates (1972) were interested in learning how gender Units of Analysis in Review roles are taught. They chose children’s picture books as their unit of analysis. Specifically, they The examples in this section should suggest the examined books that had received the Caldecott nearly infinite variety of possible units of analysis Medal. Their results were as follows: in social research. Although individual human beings are typical objects of study, many research We found that females were underrepre- questions can be answered more appropriately sented in the titles, central roles, pictures, and through the examination of other units of analy- stories of every sample of books we examined. sis. Indeed, social researchers can study just about Most children’s books are about boys, men, anything that bears on social life. male animals, and deal exclusively with male adventures. Most pictures show men singly or Moreover, the types of units of analysis named in groups. Even when women can be found in in this section do not begin to exhaust the possi- the books, they often play insignificant roles, bilities. Morris Rosenberg (1968: 234– 48), for ex- remaining both inconspicuous and nameless. ample, speaks of individual, group, organizational, institutional, spatial, cultural, and societal units of (Weitzman et al. 1972: 1128) analysis. John Lofland and his associates (2006: 122–32) speak of practices, episodes, encounters, In a more recent study, Roger Clark, Rachel roles and social types, social and personal relation- Lennon, and Leana Morris (1993) concluded that ships, groups and cliques, organizations, settle- male and female characters at that time were por- ments and habitats, subcultures, and lifestyles as trayed less stereotypically than before, observing a suitable units of study. The important thing here is clear progress toward portraying men and women to grasp the logic of units of analysis. Once you do, in nontraditional roles. However, they did not find the possibilities for fruitful research are limited only total equality between the sexes. by your imagination. As this example suggests, just as people or Categorizing possible units of analysis might social groups imply populations, each social object make the concept seem more complicated than it implies a set of all objects of the same class: all needs to be. What you call a given unit of anal- books, all novels, all biographies, all introductory ysis—a group, a formal organization, or a social sociology textbooks, all cookbooks, all press confer- artifact—is irrelevant. The key is to be clear about ences. In a study using books as the units of analy- what your unit of analysis is. When you embark on sis, an individual book might be characterized by a research project, you must decide whether you’re its size, weight, length, price, content, number of studying marriages or marriage partners, crimes pictures, number sold, or description of the author. or criminals, corporations or corporate executives. Then the population of all books or of a particular kind of book could be analyzed for the purpose of social artifact Any product of social beings or their description or explanation: What kinds of books sell behavior. Can be a unit of analysis. best and why, for example. Similarly, a social researcher could analyze whether paintings by Russian, Chinese, or U.S. artists showed the greatest degree of working-class
104 ■ Chapter 4: Research Design How to Do It: Identifying the Unit of Analysis The unit of analysis is an important element in research design and nudity is shown,but who or what shows nudity? Movies.Movies are the later in data analysis.However,students sometimes find identifying units of analysis. it elusive.The easiest way to identify the unit of analysis is to examine a statement regarding the variables under study. Finally,how about this statement:“Twenty-four percent of the families have more than one adult earning $30,000 or more.”To be sure, Consider the following:“The average household income was adults are earning the income,but the statement is about whether fami- $40,000.”Income is the variable of interest,but who or what has lies have such adults.To make this statement,we would study several income? Households,in this instance.We would arrive at the given families.For each,we would ask whether they had more than one adult statement by examining the incomes of several households.To calculate earning in excess of $30,000;each family would be scored as“yes”or the mean (average) income,we would add up all the household incomes “no”in that respect.Finally,we would calculate the percentage of families and divide by the number of households.Household is the unit of analy- scored as“yes.”The family,therefore,is the unit of analysis. sis.It is the unit being analyzed in terms of the variable,income. Consider another statement:“Italian movies show more nudity than do American movies.”The variable here is the extent to which Otherwise, you run the risk of drawing invalid something learned about an ecological unit says conclusions because your assertions about one unit something about the individuals making up that of analysis are actually based on the examination unit. Let’s consider a hypothetical illustration of this of another. We’ll see an example of this issue in the fallacy. next section as we look at the ecological fallacy. Suppose we’re interested in learning something Faulty Reasoning about Units about the nature of electoral support received by a of Analysis: The Ecological female political candidate in a recent citywide elec- Fallacy and Reductionism tion. Let’s assume we have the vote tally for each precinct so we can tell which precincts gave her the At this point, it’s appropriate to introduce two types greatest support and which the least. Assume also of faulty reasoning that you should be aware of: the that we have census data describing some charac- ecological fallacy and reductionism. Each repre- teristics of these precincts. Our analysis of such data sents a potential pitfall regarding units of analysis, might show that precincts with relatively young and either can occur in doing research and drawing voters gave the female candidate a greater propor- conclusions from the results. tion of their votes than precincts with older voters did. We might be tempted to conclude from these The Ecological Fallacy findings that younger voters are more likely to vote for female candidates than older voters are—in In this context, “ecological” refers to groups or other words, that age affects support for the sets or systems: something larger than individu- woman. In reaching such a conclusion, we run the als. The ecological fallacy is the assumption that risk of committing the ecological fallacy, because it may have been the older voters in those “young” ecological fallacy Erroneously drawing conclu- precincts who voted for the woman. Our problem sions about individuals solely from the observation is that we have examined precincts as our units of of groups. analysis but wish to draw conclusions about voters. The same problem would arise if we discov- ered that crime rates were higher in cities having large African American populations than in those
Units of Analysis ■ 105 with few African Americans. We would not know terms of limited and/or lower-order concepts. The if the crimes were actually committed by African reductionist explanation is not altogether wrong; Americans. Or if we found suicide rates higher in it is simply too limited. Thus, you might attempt Protestant countries than in Catholic ones, we still to predict this year’s winners and losers in the could not know for sure that more Protestants than National Basketball Association by focusing on the Catholics committed suicide. abilities of the individual players on each team. This is certainly not stupid or irrelevant, but the suc- In spite of these hazards, social researchers cess or failure of teams involves more than just the often have little choice but to address a particular individuals in them; it involves coaching, team- research question through an ecological analysis. work, strategies, finances, facilities, fan loyalty, and Perhaps the most appropriate data are simply not so forth. To understand why some teams do better available. For example, the precinct vote tallies and than others, you would make team the unit of anal- the precinct characteristics mentioned in our initial ysis, and the quality of players would be one variable example may be easy to obtain, but we may not you would probably want to use in describing and have the resources to conduct a postelection survey classifying the teams. of individual voters. In such cases, we may reach a tentative conclusion, recognizing and noting the Further, different academic disciplines ap- risk of an ecological fallacy. proach the same phenomenon quite differently. Sociologists tend to consider sociological variables Although you should be careful not to com- (such as values, norms, and roles), economists ponder mit the ecological fallacy, don’t let these warnings economic variables (such as supply and demand lead you into committing what we might call the and marginal value), and psychologists examine individualistic fallacy. Some people who approach psychological variables (such as personality types and social research for the first time have trouble rec- traumas). Explaining all or most human behavior onciling general patterns of attitudes and actions in terms of economic factors is called economic re- with individual exceptions. But generalizations ductionism, explaining it in terms of psychological and probabilistic statements are not invalidated by factors is called psychological reductionism, and so individual exceptions. Your knowing a rich Demo- forth. Notice how this issue relates to the discussion crat, for example, doesn’t deny the fact that most of theoretical paradigms in Chapter 2. rich people vote Republican—as a general pattern. Similarly, if you know someone who has gotten For many social scientists, the field of socio- rich without any formal education, that doesn’t biology is a prime example of reductionism, sug- deny the general pattern of higher education relat- gesting that all social phenomena can be explained ing to higher income. in terms of biological factors. Thus, for example, Edward O. Wilson (1975) sought to explain altru- The ecological fallacy deals with something else istic behavior in human beings in terms of genetic altogether—confusing units of analysis in such a makeup. In his neo-Darwinian view, Wilson sug- way that we draw conclusions about individuals gests that humans have evolved in such a way that solely from the observation of groups. Although individuals sometimes need to sacrifice themselves the patterns observed between variables at the for the benefit of the whole species. Some people level of groups may be genuine, the danger lies in reasoning from the observed attributes of groups reductionism A fault of some researchers: a strict to the attributes of the individuals who made up limitation (reduction) of the kinds of concepts to those groups, even though we have not actually be considered relevant to the phenomenon under observed individuals. study. Reductionism sociobiology A paradigm based in the view that social behavior can be explained solely in terms of A second type of faulty reasoning related to units of genetic characteristics and behavior. analysis is reductionism. Reductionism involves attempts to explain a particular phenomenon in
106 ■ Chapter 4: Research Design might explain such sacrifice in terms of ideals or Time plays many roles in the design and warm feelings between humans. However, genes execution of research, quite aside from the time are the essential unit in Wilson’s paradigm, produc- it takes to do research. Earlier we noted that the ing his famous dictum that human beings are “only time sequence of events and situations is critical DNA’s way of making more DNA.” to determining causation (a point we’ll return to in Part 4). Time also affects the generalizability of Reductionism of any type tends to suggest that research findings. Do the descriptions and explana- particular units of analysis or variables are more tions resulting from a particular study accurately relevant than others. Suppose we ask what caused represent the situation of ten years ago, ten years the American Revolution. Was it a shared commit- from now, or only the present? Researchers have ment to the value of individual liberty? The eco- two principal options available to deal with the nomic plight of the colonies in relation to Britain? issue of time in the design of their research: cross- The megalomania of the founders? As soon as we sectional studies and longitudinal studies. inquire about the single cause, we run the risk of reductionism. If we were to regard shared values Cross-Sectional Studies as the cause of the American Revolution, our unit of analysis would be the individual colonist. An A cross-sectional study involves observations economist, though, might choose the 13 colonies of a sample, or cross section, of a population or as units of analysis and examine the economic phenomenon that are made at one point in time. organizations and conditions of each. A psycholo- Exploratory and descriptive studies are often cross- gist might choose individual leaders as the units of sectional. A single U.S. Census, for instance, is a analysis for purposes of examining their personali- study aimed at describing the U.S. population at a ties. Of course, there’s nothing wrong in choosing given time. these units of analysis as part of an explanation of the American Revolution, but I think you can see Many explanatory studies are also cross- how each alone would not produce a complete sectional. A researcher conducting a large-scale answer. national survey to examine the sources of racial and religious prejudice would, in all likelihood, be Like the ecological fallacy, reductionism can dealing with a single time frame—taking a snap- occur when we use inappropriate units of analy- shot, so to speak, of the sources of prejudice at a sis. The appropriate unit of analysis for a given particular point in history. research question, however, is not always clear. Social researchers, especially across disciplinary Explanatory cross-sectional studies have an boundaries, often debate this issue. inherent problem. Although their conclusions are based on observations made at only one time, typi- The Time Dimension cally they aim at understanding causal processes that occur over time. This problem is somewhat So far in this chapter, we’ve regarded research akin to that of determining the speed of a moving design as a process for deciding what aspects we’ll object on the basis of a high-speed, still photograph observe, of whom, and for what purpose. Now we that freezes the movement of the object. must consider a set of time-related options that cuts across each of these earlier considerations. We can Yanjie Bian, for example, conducted a survey of choose to make observations more or less at one workers in Tianjin, China, for the purpose of study- time or over a long period. ing stratification in contemporary, urban Chinese society. In undertaking the survey in 1988, how- cross-sectional study A study based on observa- ever, he was conscious of the important changes tions representing a single point in time. brought about by a series of national campaigns, such as the Great Proletarian Cultural Revolution, dating from the Chinese Revolution in 1949 (which
The Time Dimension ■ 107 brought the Chinese Communists into power) and Schachter were specifically interested in learning continuing into the present. what happened to a flying saucer cult when their predictions of an alien encounter failed to come These campaigns altered political atmospheres true. Would the cult members close down the and affected people’s work and nonwork activi- group, or would they become all the more com- ties. Because of these campaigns, it is difficult to mitted to their beliefs? A longitudinal study was draw conclusions from a cross-sectional social required to provide an answer. (They redoubled survey, such as the one presented in this book, their efforts to get new members.) about general patterns of Chinese workplaces and their effects on workers. Such conclusions Longitudinal studies can be more difficult for may be limited to one period of time and are quantitative studies such as large-scale surveys. subject to further tests based on data collected Nonetheless, they are often the best way to study at other times. changes over time. There are three special types of longitudinal studies that you should know about: (1994: 19) trend studies, cohort studies, and panel studies. The problem of generalizations about social Trend Studies life from a “snapshot” is one this book repeatedly addresses. One solution is suggested by Bian’s final A trend study is a type of longitudinal study that comment—about data collected “at other times”: examines changes within a population over time. Social research often involves revisiting phenom- A simple example is a comparison of U.S. Censuses ena and building on the results of earlier research. over a period of decades, showing shifts in the makeup of the national population. A similar use Longitudinal Studies of archival data was made by Michael Carpini and Scott Keeter (1991), who wanted to know whether In contrast to cross-sectional studies, a longitudi- contemporary U.S. citizens were better or more nal study is designed to permit observations of the poorly informed about politics than citizens of an same phenomenon over an extended period. For earlier generation were. To find out, they compared example, a researcher can participate in and ob- the results of several Gallup Polls conducted during serve the activities of a UFO cult from its inception the 1940s and 1950s with a 1989 survey that asked to its demise. Other longitudinal studies use records several of the same questions tapping political or artifacts to study changes over time. In analyses knowledge. of newspaper editorials or Supreme Court decisions over time, for example, the studies are longitudinal Overall, the analysis suggested that contem- whether the researcher’s actual observations and porary citizens were slightly better informed than analyses were made at one time or over the course earlier generations were. In 1989, 74 percent of of the actual events under study. the sample could name the vice president of the United States, compared with 67 percent in 1952. Many field research projects, involving direct Substantially higher percentages of people in 1989 observation and perhaps in-depth interviews, are naturally longitudinal. Thus, for example, when longitudinal study A study design involving the Ramona Asher and Gary Fine (1991) studied the collection of data at different points in time. life experiences of the wives of alcoholic men, they were in a position to examine the evolution of trend study A type of longitudinal study in which troubled marital relationships over time, sometimes a given characteristic of some population is moni- even including the reactions of the subjects to the tored over time. An example would be the series of research itself. Gallup Polls showing the electorate’s preferences for political candidates over the course of a campaign, In the classic study When Prophecy Fails (1956), even though different samples were interviewed at Leon Festinger, Henry Reicker, and Stanley each point.
108 ■ Chapter 4: Research Design 40 Age 30 20 0 1990 2000 1980 Year FIGURE 4-5 A Cohort Study Design. Each of the three groups shown here is a sample representing people who were born in 1960. than in 1947 could explain presidential vetoes and married in 1994, and so forth. An example of a congressional overrides of vetoes. On the other cohort study would be a series of national surveys, hand, more of the 1947 sample could identify their conducted perhaps every 20 years, to study the U.S. representative (38 percent) than the 1989 attitudes of the cohort born during World War II sample (29 percent) could. toward U.S. involvement in global affairs. A sample of people 15–20 years old might be surveyed in An in-depth analysis, however, indicates that 1960, another sample of those 35– 40 years old in the slight increase in political knowledge resulted 1980, and another sample of those 55–60 years old from the fact that the people in the 1989 sample in 2000. Although the specific set of people studied were more highly educated than those from earlier in each survey would differ, each sample would samples were. When educational levels were taken represent the cohort born between 1940 and 1945. into account, the researchers concluded that politi- cal knowledge has actually declined within specific Figure 4-5 offers a graphic illustration of a co- educational groups. hort design. In the example, three studies are being compared: one was conducted in 1980, another Cohort Studies in 1990, and the third in 2000. Those who were 20 years old in the 1980 study are compared with In a cohort study, a researcher examines specific those who were 30 in the 1990 study and those subpopulations, or cohorts, as they change over who were 40 in the 2000 study. Although the sub- time. Typically, a cohort is an age group, such jects being described in each of the three groups are as people born during the 1950s, but it can also different, each set of subjects represents the same be some other time grouping, such as people cohort: those who were born in 1960. born during the Vietnam War, people who got James Davis (1992) turned to a cohort analysis cohort study A study in which some specific sub- in an attempt to understand shifting political orien- population, or cohort, is studied over time, although tations during the 1970s and 1980s in the United data may be collected from different members in States. Overall, he found a liberal trend on issues each set of observations. For example, a study of the such as race, gender, religion, politics, crime, and occupational history of the class of 1970 in which free speech. But did this trend represent people in questionnaires were sent every five years would be general getting a bit more liberal, or did it merely a cohort study. reflect liberal younger generations replacing the conservative older ones?
The Time Dimension ■ 109 TABLE 4-1 Age and Political Liberalism Survey dates 1972 to 1974 1977 to 1980 1982 to 1984 1987 to 1989 25–29 30–34 35–39 Age of cohort 20–24 68% 73% 73% Percent who would let the Communist speak 72% To answer this question, Davis examined educational funding in U.S. society over time in national surveys (from the General Social Sur- general. The researchers add to this the concept vey, of which he is a founder) conducted in four of “generational replacement,” meaning that the time periods, five years apart. In each survey, he older respondents in a survey grew up during a grouped the respondents into age groups, also five time when there was less support for education in years apart. This strategy allowed him to compare general, whereas the younger respondents grew up different age groups at any given point in time as during a time of greater overall support. well as to follow the political development of each age group over time. A cohort analysis allowed the researchers to de- termine what happened to the attitudes of specific One of the questions he examined was cohorts over time. Here, for example, are the whether a person who admitted to being a Com- percentages of Americans born during the 1940s munist should be allowed to speak in the respon- who felt educational spending was too low, when dents’ communities. Consistently, the younger members of that cohort were interviewed over time respondents in each time period were more willing (Plutzer and Berkman 2005: 76): to let the Communist speak than the older ones were. Among those aged 20– 40 in the first set Year Interviewed Percent Who Say Educational of the survey, for example, 72 percent took this 1970s Funding Is Too Low liberal position, contrasted with 27 percent among 1980s respondents 80 and older. What Davis found when 1990s 58 he examined the youngest cohort over time is 2000s 66 shown in Table 4-1. This pattern of a slight, con- 74 servative shift in the 1970s, followed by a liberal 79 rebound in the 1980s, typifies the several cohorts Davis analyzed (J. Davis 1992: 269). As these data indicate, those who were born during the 1940s have steadily increased their sup- In another study, Eric Plutzer and Michael port for educational funding as they have passed Berkman (2005) used a cohort design to com- through and beyond the child-rearing years. pletely reverse a prior conclusion regarding aging and support for education. Logically, as people Panel Studies grow well beyond the child-rearing years, we might expect them to reduce their commitment to Though similar to trend and cohort studies, a educational funding. Moreover, cross-sectional data panel study examines the same set of people each support that expectation. The researchers present time. For example, we could interview the same several data sets showing those over 65 voicing less support for educational funding than those under panel study A type of longitudinal study, in which 65 did. data are collected from the same set of people (the sample or panel) at several points in time. Such simplistic analyses, however, leave out an important variable: increasing support for
110 ■ Chapter 4: Research Design sample of voters every month during an election affiliation. A trend study might look at shifts in campaign, asking for whom they intended to vote. U.S. religious affiliations over time, as the Gallup Though such a study would allow us to analyze Poll does on a regular basis. A cohort study might overall trends in voter preferences for different can- follow shifts in religious affiliations among “the De- didates, it would also show the precise patterns of pression generation,” specifically, say, people who persistence and change in intentions. For example, were 20 to 30 years old in 1932. We could study a a trend study that showed that Candidates A and B sample of people 30– 40 years old in 1942, a new each had exactly half of the voters on September 1 sample of people aged 40–50 in 1952, and so forth and on October 1 as well could indicate that none throughout their lifespan. A panel study could start of the electorate had changed voting plans, that with a sample of the whole population or of some all of the voters had changed their intentions, or special subset and study those specific individuals something in-between. A panel study would elimi- over time. Notice that only the panel study would nate this confusion by showing what kinds of vot- give a full picture of the shifts among the various ers switched from A to B and what kinds switched categories of affiliations, including “none.” Cohort from B to A, as well as other facts. and trend studies would uncover only net changes. Joseph Veroff, Shirley Hatchett, and Elizabeth Longitudinal studies have an obvious advan- Douvan (1992) wanted to learn about marital ad- tage over cross-sectional ones in providing infor- justment among newlyweds, specifically regarding mation describing processes over time. But this differences between white and African American advantage often comes at a heavy cost in both time couples. To get subjects for study, they selected and money, especially in a large-scale survey. Ob- a sample of couples who applied for marriage servations may have to be made at the time events licenses in Wayne County, Michigan, April through are occurring, and the method of observation may June 1986. require many research workers. Concerned about the possible impact their Panel studies, which offer the most compre- research might have on the couples’ marital ad- hensive data on changes over time, face a special justment, the researchers divided their sample in problem: panel attrition. Some of the respondents half at random: an experimental group and a control studied in the first wave of the survey might not group (concepts we’ll explore further in Chapter 8). participate in later waves. (This is comparable to Couples in the former group were intensively inter- the problem of experimental mortality discussed in viewed over a four-year period, whereas the latter Chapter 8.) The danger is that those who drop out group was contacted only briefly each year. of the study may be atypical, thereby distorting the results of the study. Thus, when Carol Aneshensel By studying the same couples over time, the and her colleagues conducted a panel study of ado- researchers could follow the specific problems that lescent girls (comparing Latinas and non-Latinas), arose and the way the couples dealt with them. they looked for and found differences in character- As a by-product of their research, they found istics of survey dropouts among Latinas born in the that those studied the most intensely seemed to United States and those born in Mexico. These dif- achieve a somewhat better marital adjustment. ferences needed to be taken into account to avoid The researchers felt that the interviews could have misleading conclusions about differences between forced couples to discuss matters they might have Latinas and non-Latinas (Aneshensel et al. 1989). otherwise buried. Comparing the Three Types Approximating Longitudinal of Longitudinal Studies Studies To reinforce the distinctions among trend, cohort, Longitudinal studies do not always provide a fea- and panel studies, let’s contrast the three study sible or practical means of studying processes that designs in terms of the same variable: religious take place over time. Fortunately, researchers often
The Time Dimension ■ 111 can draw approximate conclusions about such pressure, and so forth. By reading across the age- processes even when only cross-sectional data are group ratings for each health condition, you would available. Here are some ways to do that. have something approximating the health history of individuals. Thus, you might conclude that the Sometimes cross-sectional data imply processes average person develops vision problems before over time on the basis of simple logic. For example, hearing problems. You would need to be cautious in the study of student drug use conducted at the in this assumption, however, because the differ- University of Hawaii (Chapter 2), students were ences might reflect societywide trends. Perhaps asked to report whether they had ever tried each improved hearing examinations instituted in the of several illegal drugs. The study found that some schools had affected only the young people in your students had tried both marijuana and LSD, some study. had tried only one, and others had tried neither. Because these data were collected at one time, and Asking people to recall their pasts is another because some students presumably would experi- common way of approximating observations over ment with drugs later on, it would appear that such time. Researchers use that method when they a study could not tell whether students were more ask people where they were born or when they likely to try marijuana or LSD first. graduated from high school or whom they voted for in 1988. Qualitative researchers often conduct A closer examination of the data showed, how- in-depth “life history” interviews. For example, ever, that although some students reported having C. Lynn Carr (1998) used this technique in a study tried marijuana but not LSD, there were no stu- of “tomboyism.” Her respondents, aged 25– 40, dents in the study who had tried only LSD. From were asked to reconstruct aspects of their lives from this finding it was inferred—as common sense childhood on, including experiences of identifying suggested—that marijuana use preceded LSD use. themselves as tomboys. If the process of drug experimentation occurred in the opposite time order, then a study at a given The danger in this technique is evident. Some- time should have found some students who had times people have faulty memories; sometimes tried LSD but not marijuana, and it should have they lie. When people are asked in postelection found no students who had tried only marijuana. polls whom they voted for, the results inevitably show more people voting for the winner than Researchers can also make logical inferences actually did so on election day. As part of a series of whenever the time order of variables is clear. If in-depth interviews, such a report can be validated we discovered in a cross-sectional study of college in the context of other reported details; however, students that those educated in private high schools results based on a single question in a survey must received better college grades than those educated be regarded with caution. in public high schools did, we would conclude that the type of high school attended affected college Cohorts can also be used to infer processes over grades, not the other way around. Thus, even time from cross-sectional data. For example, when though we made our observations at only one time, Prem Saxena and his colleagues (2004) wanted to we would feel justified in drawing conclusions examine whether wartime conditions would affect about processes taking place across time. the age at which people married, he used cross- sectional data from a survey of Lebanese women. Very often, age differences discovered in a During the Lebanese Civil War from 1975 to 1990, cross-sectional study form the basis for inferring many young men migrated to other countries. By processes across time. Suppose you’re interested in noting the year in which the survey respondents the pattern of worsening health over the course of first married, he could determine that the average the typical life cycle. You might study the results age-at-first-marriage increased with the onset of of annual checkups in a large hospital. You could the war. group health records according to the ages of those examined and rate each age group in terms of This discussion of the ways that time figures several health conditions—sight, hearing, blood into social research suggests several questions you
112 ■ Chapter 4: Research Design should confront in your own research projects. In nation of participant observation, interviews designing any study, be sure to look at both the ex- with participants, and review of organizational plicit and implicit assumptions you’re making about records. (Kilburn 1998: 89) time. Are you interested in describing some process that occurs over time, or are you simply going to • Using interviews obtained during fieldwork describe what exists now? If you want to describe a process occurring over time, will you be able to in Palestine in 1992, 1993, and 1994, and make observations at different points in the process, employing historical and archival records, I or will you have to approximate such observations argue that Palestinian feminist discourses were by drawing logical inferences from what you can shaped and influenced by the sociopolitical observe now? If you opt for a longitudinal design, context in which Palestinian women acted which method best serves your research purposes? and with which they interacted. (Abdulhadi 1998: 649) Examples of Research Strategies • This article reports on women’s experiences As the preceding discussions have implied, social research follows many paths. The following short of breastfeeding in public as revealed through excerpts further illustrate this point. As you read in-depth interviews with 51 women. (Stearns each excerpt, note both the content of each study 1999: 308) and the method used to study the chosen topic. Does the study seem to be exploring, describing, or • Using interview and observational field data, explaining (or some combination of these)? What are the sources of data in each study? Can you I demonstrate how a system of temporary identify the unit of analysis? Is the dimension of employment in a participative workplace both time relevant? If so, how will it be handled? exploited and shaped entry-level workers’ aspirations and occupational goals. (V. Smith • This case study of unobtrusive mobilizing by 1998: 411) Southern California Rape Crisis Center uses • I collected data [on White Separatist rhetoric] archival, observational, and interview data to explore how a feminist organization worked to from several media of public discourse, includ- change police, schools, prosecutors, and some ing periodicals, books, pamphlets, transcripts state and national organizations from 1974 to from radio and television talk shows, and 1994. (Schmitt and Martin 1999: 364) newspaper and magazine accounts. (Berbrier 1998: 435) • Using life history narratives, the present study • In the analysis that follows, racial and gender investigates processes of agency and conscious- ness among 14 women who identified them- inequality in employment and retirement will selves as tomboys. (Carr 1998: 528) be analyzed, using a national sample of persons who began receiving Social Security Old Age • By drawing on interviews with activists in the benefits in 1980–81. (Hogan and Perrucci 1998: 528) former Estonian Soviet Socialist Republic, we specify the conditions by which accommoda- • Drawing from interviews with female crack tive and oppositional subcultures exist and are successfully transformed into social move- dealers, this paper explores the techniques ments. (Johnston and Snow 1998: 473) they use to avoid arrest. (Jacobs and Miller 1998: 550) • This paper presents the results of an ethno- How to Design graphic study of an AIDS service organization a Research Project located in a small city. It is based on a combi- You’ve now seen some of the options available to social researchers in designing projects. I know there are a lot of components, and the relationships
How to Design a Research Project ■ 113 among them may not be totally clear, so here’s a purpose, the researcher needs to make a variety way of pulling them together. Let’s assume you of decisions, as indicated in the remainder of the were to undertake research. Where would you diagram. start? Then, where would you go? To make this discussion more concrete, let’s Although research design occurs at the begin- take a specific research example. Suppose you’re ning of a research project, it involves all the steps concerned with the issue of abortion and have of the subsequent project. This discussion, then, a special interest in learning why some college provides both guidance on how to start a research students support abortion rights and others oppose project and an overview of the topics that follow in them. Going a step further, let’s say you’ve formed later chapters of this book. the impression that students in the humanities and social sciences seem generally more inclined Figure 4-6 presents a schematic view of the to support the idea of abortion rights than those traditional image of research design. I present this in the natural sciences do. (That kind of thinking view reluctantly, because it may suggest more of a often leads people to design and conduct social step-by-step order to research than actual practice research.) bears out. Nonetheless, this idealized overview of the process provides a context for the specific So, where do you start? You have an idea you details of particular components of social research. want to pursue, one that involves abortion attitudes Essentially, it is another and more detailed picture and choice of college major. In terms of the options of the scientific process presented in Chapter 2. we’ve discussed in this chapter, you probably have both descriptive and explanatory interests, but you At the top of the diagram are interests, ideas, might decide you only want to explore the issue. and theories, the possible beginning points for You might wonder what sorts of attitudes students a line of research. The letters (A, B, X, Y, and so with different majors have about abortion (ex- forth) represent variables or concepts such as ploratory), what percentage of the student body prejudice or alienation. Thus, you might have a supports a woman’s right to an abortion (descrip- general interest in finding out what causes some tive), or what causes some to support it and others people to be more prejudiced than others, or you to oppose it (explanation). The units of analysis in might want to know some of the consequences of this case would be individuals: college students. But alienation. Alternatively, your inquiry might begin we’re jumping the gun. As you can see, even be- with a specific idea about the way things are. For fore we’ve “started,” we’ve started. The reciprocal example, you might have the idea that working on processes described in Figure 4-6 begin even before an assembly line causes alienation. The question you’ve made a commitment to a project. Let’s look marks in the diagram indicate that you aren’t sure more formally at the various steps, then, keeping things are the way you suspect they are—that’s this reciprocal motion in mind. why you’re doing the research. Notice that a theory is represented as a set of complex relationships Getting Started among several variables. At the outset of your project, your aim would prob- The double arrows between “interest,” “idea,” ably be exploratory. At this point, you might choose and “theory” suggest that a movement back and among several possible activities in pursuing your forth across these several possible beginnings often interest in student attitudes about abortion rights. takes place. An initial interest may lead to the for- To begin with, you might want to read something mulation of an idea, which may be fit into a larger about the issue. If you have a hunch that attitudes theory, and the theory may produce new ideas and are somehow related to college major, you might create new interests. find out what other researchers may have written about that. Appendix A of this book will help you Any or all of these three may suggest the make use of your college library. In addition, you need for empirical research. The purpose of such research can be to explore an interest, test a specific idea, or validate a complex theory. Whatever the
FIGURE 4-6 Traditional Image of Research Design.
How to Design a Research Project ■ 115 would probably talk to some people who support process of conceptualization in depth. For now, let’s abortion rights and some who don’t. You might see what it might involve in the case of our hypo- attend meetings of abortion-related groups. All thetical example. these activities could help prepare you to handle the various decisions of research design we’re about If you’re going to study how college students to examine. feel about abortion and why, the first thing you’ll have to specify is what you mean by “the right to Before designing your study, you must define an abortion.” Because support for abortion prob- the purpose of your project. What kind of study ably varies according to the circumstances, you’ll will you undertake—exploratory, descriptive, want to pay attention to the different conditions explanatory? Do you plan to write a research paper under which people might approve or disapprove to satisfy a course or thesis requirement? Is your of abortion: for example, when the woman’s life is purpose to gain information that will support you in danger, in the case of rape or incest, or simply as in arguing for or against abortion rights? Do you a matter of personal choice. want to write an article for the campus newspaper or an academic journal? In reviewing the previous Similarly, you’ll need to specify exact mean- research literature regarding abortion rights, you ings for all the other concepts you plan to study. should note the design decisions other researchers If you want to study the relationship of opinion have made, always asking whether the same deci- about abortion to college major, you’ll have to sions would satisfy your purpose. decide whether you want to consider only officially declared majors or to include students’ intentions Usually, your purpose for undertaking research as well. What will you do with those who have no can be expressed as a report. A good first step in major? designing your project is to outline such a report (see Chapter 17 for help on this). Although your In surveys and experiments, you need to final report may not look much like your ini- specify such concepts in advance. In less tightly tial image of it, this exercise will help you figure structured research, such as open-ended inter- out which research designs are most appropri- views, an important part of the research may in- ate. During this step, clearly describe the kinds of volve the discovery of different dimensions, aspects, statements you want to make when the research or nuances of concepts. In such cases, the research is complete. Here are some examples of such state- itself may uncover and report aspects of social life ments: “Students frequently mentioned abortion that were not evident at the outset of the project. rights in the context of discussing social issues that concerned them personally.” “X percent of State Choice of Research Method U. students favor a woman’s right to choose an abortion.” “Engineers are (more/less) likely than As we’ll discuss in Part 3, each research method sociologists to favor abortion rights.” has its strengths and weaknesses, and certain concepts are more appropriately studied through Conceptualization some methods than through others. In our study of attitudes toward abortion rights, a survey might be Once you have a well-defined purpose and a clear the most appropriate method: either interviewing description of the kinds of outcomes you want to students or asking them to fill out a questionnaire. achieve, you can proceed to the next step in the Surveys are particularly well suited to the study of design of your study—conceptualization. We often public opinion. This is not to say that you couldn’t talk pretty casually about social science concepts make good use of the other methods presented in such as prejudice, alienation, religiosity, and liberal- Part 3. For example, you might use the method of ism, but it’s necessary to clarify what we mean content analysis to examine letters to the editor by these concepts, in order to draw meaningful and analyze the different images of abortion that conclusions about them. Chapter 5 examines this letter writers have. Field research would provide an avenue to understanding how people interact with
116 ■ Chapter 4: Research Design one another regarding the issue of abortion, how In every case, then, we select a sample from among they discuss it, and how they change their minds. the data that might be collected and studied. The Other research methods introduced in Part 3 could sampling of information, of course, occurs in ev- also be used in studying this topic. Usually, the best eryday life and often produces biased observations. study design uses more than one research method, (Recall the discussion of “selective observation” in taking advantage of their different strengths. If you Chapter 1.) Social researchers are more deliberate look back at the brief examples of actual studies at in their sampling of what will be observed. the end of the preceding section, you’ll see several instances where the researchers used many meth- Chapter 7 describes methods for selecting ods in a single study. samples that adequately reflect the whole popula- tion that interests us. Notice in Figure 4-6 that deci- Operationalization sions about population and sampling are related to decisions about the research method to be used. Once you’ve specified the concepts to be studied Whereas probability sampling techniques would be and chosen a research method, the next step is op- relevant to a large-scale survey or a content analy- erationalization, or deciding on your measurement sis, a field researcher might need to select only techniques (discussed further in Chapters 5 and 6). those informants who will yield a balanced picture The meaning of variables in a study is determined of the situation under study, and an experimenter in part by how they are measured. Part of the task might assign subjects to experimental and control here is deciding how the desired data will be col- groups in a manner that creates comparability. lected: direct observation, review of official docu- ments, a questionnaire, or some other technique. In your hypothetical study of abortion at- titudes, the relevant population would be the If you decided to use a survey to study attitudes student population of your college. As you’ll toward abortion rights, part of operationalization discover in Chapter 7, however, selecting a sample is determining the wording of questionnaire items. will require you to get more specific than that. Will For example, you might operationalize your main you include part-time as well as full-time students? variable by asking respondents whether they would Only degree candidates or everyone? International approve of a woman’s right to have an abortion students as well as U.S. citizens? Undergraduates, under each of the conditions you’ve conceptual- graduate students, or both? There are many such ized: in the case of rape or incest, if her life were questions—each of which must be answered in threatened by the pregnancy, and so forth. You’d terms of your research purpose. If your purpose design the questionnaire so that it asked respon- is to predict how students would vote in a local dents to express approval or disapproval for each referendum on abortion, you might want to situation. Similarly, you would specify exactly how limit your population to those eligible and likely respondents would indicate their college major, as to vote. well as what choices to provide those who have not declared a major. Observations Population and Sampling Having decided what to study among whom by what method, you’re now ready to make obser- In addition to refining concepts and measurements, vations—to collect empirical data. The chapters you must decide whom or what to study. The popu- of Part 3, which describe the various research lation for a study is that group (usually of people) methods, give the different observation techniques about whom we want to draw conclusions. We’re appropriate to each. almost never able to study all the members of the population that interests us, however, and we can To conduct a survey on abortion, you might never make every possible observation of them. want to print questionnaires and mail them to a sample selected from the student body. Alterna- tively, you could arrange to have a team of inter-
How to Design a Research Project ■ 117 viewers conduct the survey over the telephone. Moving beyond simple description, you might The relative advantages and disadvantages of these describe the opinions of subsets of the student and other possibilities are discussed in Chapter 9. body, such as different college majors. Provided that your design called for trapping other informa- Data Processing tion about respondents, you could also look at men versus women; freshmen, sophomores, juniors, Depending on the research method chosen, you’ll seniors, and graduate students; or other categories have amassed a volume of observations in a form that you’ve included. The description of subgroups that probably isn’t immediately interpretable. If could then lead you into an explanatory analysis. you’ve spent a month observing a street-corner gang firsthand, you’ll now have enough field notes Application to fill a book. In a historical study of ethnic diversity at your school, you may have amassed volumes The final stage of the research process involves the of official documents, interviews with administra- uses made of the research you’ve conducted and tors and others, and so forth. Chapters 13 and 14 the conclusions you’ve reached. To start, you’ll describe some of the ways social science data are probably want to communicate your findings so processed or transformed for qualitative or quanti- that others will know what you’ve learned. It may tative analysis. be appropriate to prepare—and even publish—a written report. Perhaps you’ll make oral presenta- In the case of a survey, the “raw” observations tions, such as papers delivered to professional and are typically in the form of questionnaires with scientific meetings. Other students would also be boxes checked, answers written in spaces, and the interested in hearing what you’ve learned about like. The data-processing phase of a survey typically them. involves the classification (coding) of written-in answers and the transfer of all information to a You may want to go beyond simply reporting computer. what you’ve learned to discussing the implica- tions of your findings. Do they say anything about Analysis actions that might be taken in support of policy goals? Both the proponents and the opponents of Once the collected data are in a suitable form, abortion rights would be interested. you’re ready to interpret them for the purpose of drawing conclusions that reflect the interests, ideas, Finally, be sure to consider what your research and theories that initiated the inquiry. Chapters 13 suggests in regard to further research on your sub- and 14 describe a few of the many options avail- ject. What mistakes should be corrected in future able to you in analyzing data. In Figure 4-6, notice studies? What avenues— opened up slightly in that the results of your analyses feed back into your study—should be pursued further? your initial interests, ideas, and theories. Often this feedback represents the beginning of another cycle Research Design in Review of inquiry. As this overview shows, research design involves a In the survey of student attitudes about abor- set of decisions regarding what topic is to be stud- tion rights, the analysis phase would pursue both ied among what population with what research descriptive and explanatory aims. You might begin methods for what purpose. Although you’ll want by calculating the percentages of students who to consider many ways of studying a subject—and favored or opposed each of the several different use your imagination as well as your knowledge of versions of abortion rights. Taken together, these a variety of methods—research design is the pro- several percentages would provide a good picture cess of focusing your perspective for the purposes of student opinion on the issue. of a particular study.
118 ■ Chapter 4: Research Design If you’re doing a research project for one of by attending meetings of certain groups? Could your courses, many aspects of research design you glean the data you need from books in the may be specified for you in advance, including the library? method (such as an experiment) or the topic (as in a course on a particular subject, such as prejudice). As you answer these questions, you’ll find The following summary assumes that you’re free yourself well into the process of research design. to choose both your topic and your research Keep in mind your own research abilities and the strategy. resources available to you. There’s little point in designing a perfect study that you can’t actually In designing a research project, you’ll find carry out. You may want to try a research method it useful to begin by assessing three things: your you haven’t used before so you can learn from it, interests, your abilities, and the available resources. but be careful not to put yourself at too great a Each of these considerations will suggest a large disadvantage. number of possible studies. Once you have a general idea of what you Simulate the beginning of a somewhat con- want to study and how, carefully review previous ventional research project: Ask yourself what research in journals and books to see how other you’re interested in understanding. Surely you researchers have addressed the topic and what they have several questions about social behavior and have learned about it. Your review of the litera- attitudes. Why are some people politically liberal ture may lead you to revise your research design: and others politically conservative? Why are some Perhaps you’ll decide to use a previous researcher’s people more religious than others? Why do people method or even replicate an earlier study. A join militia groups? Do colleges and universities still standard procedure in the physical sciences, the discriminate against minority faculty members? independent replication of research projects is just Why would a woman stay in an abusive relation- as important in the social sciences, although social ship? Spend some time thinking about the kinds of researchers tend to overlook that. Or, you might questions that interest and concern you. want to go beyond replication and study some as- pect of the topic that you feel previous researchers Once you have a few questions you’d be in- have overlooked. terested in answering for yourself, think about the kind of information needed to answer them. What Here’s another approach you might take. research units of analysis would provide the most Suppose a topic has been studied previously using relevant information: college students, corpora- field research methods. Can you design an experi- tions, voters, cities, or corporations? This question ment that would test the findings those earlier will probably be inseparable in your thoughts from researchers produced? Or, can you think of existing the question of research topics. Then ask which statistics that could be used to test their conclu- aspects of the units of analysis would provide the sions? Did a mass survey yield results that you’d information you need in order to answer your like to explore in greater detail through on-the-spot research question. observations and in-depth interviews? The use of several different research methods to test the Once you have some ideas about the kind of same finding is sometimes called triangulation, and information relevant to your purpose, ask yourself you should always keep it in mind as a valuable how you might go about getting that information. research strategy. Because each research method Are the relevant data likely to be already available has particular strengths and weaknesses, there is somewhere (say, in a government publication), or always a danger that research findings will reflect, would you have to collect them yourself? If you at least in part, the method of inquiry. In the best of think you would have to collect them, how would all worlds, your own research design should bring you go about doing it? Would you need to survey a more than one research method to bear on the large number of people, or interview a few people topic. in depth? Could you learn what you need to know
The Research Proposal ■ 119 The Research Proposal in your course, you may want to skip ahead to Chapter 17. It will familiarize you with the differ- Quite often, in the design of a research project, ent types of research literature, how to find what you’ll have to lay out the details of your plan for you want, and how to read it. There is a special someone else’s review and/or approval. In the case discussion of how to use online resources and of a course project, for example, your instructor how to avoid being misled by information on the might very well want to see a “proposal” before Internet. you set off to work. Later in your career, if you wanted to undertake a major project, you might In part, your review of the literature will be need to obtain funding from a foundation or gov- shaped by the data-collection method(s) you ernment agency, who would most definitely want intend to use in your study. Reviewing the designs a detailed proposal that describes how you would of previous studies using that same technique can spend their money. You might respond to a Request give you a head start in planning your own study. for Proposals (RFP), which both public and private At the same time, you should focus your search agencies often circulate in search of someone to do on your research topic, regardless of the methods research for them. other researchers have used. So, if you’re planning field research on, say, interracial marriages, you This chapter continues with a brief discussion might gain some useful insights from the findings of how you might prepare a research proposal. of surveys on the topic; further, past field research This will give you one more overview of the whole on interracial marriages could be invaluable in your research process that the rest of this book details. designing a survey on the topic. Elements of a Research Proposal Because the literature review will appear early in your research proposal, you should write it Although some funding agencies (or your instruc- with an eye to introducing the reader to the topic tor, for that matter) may have specific requirements you will address, laying out in a logical manner for the elements or structure of a research proposal, what has already been learned on the topic by here are some basic elements you should include. past researchers, then leading up to the holes or loose ends in our knowledge of the topic, which Problem or Objective you propose to remedy. Or, a little differently, your review of the literature may point to inconsisten- What exactly do you want to study? Why is it cies or disagreements to be found among the exist- worth studying? Does the proposed study have ing research findings. In that case, your proposed practical significance? Does it contribute to the research will aim to resolve the ambiguities that construction of social theories? plague us. I don’t know about you, but I’m already excited about the research you’re proposing to Literature Review undertake. What have others said about this topic? What theo- Subjects for Study ries address it and what do they say? What previ- ous research exists? Are there consistent findings, Whom or what will you study in order to collect or do past studies disagree? Are there flaws in the data? Identify the subjects in general, theoretical body of existing research that you think you can terms; then, in specific, more concrete terms, iden- remedy? tify who is available for study and how you’ll reach them. Will it be appropriate to select a sample? If so, Chapter 17 has a lengthier discussion of this how will you do that? If there is any possibility that topic. You’ll find that reading social science re- your research will affect those you study, how will search reports requires special skills. If you need to you insure that the research does not harm them? undertake a review of the literature at this point
120 ■ Chapter 4: Research Design Beyond these general questions, the specific Budget research method you’ll use will further specify the matter. If you’re planning to undertake an experi- When you ask someone to cover the costs of ment, a survey, or field research, for example, the your research, you need to provide a budget techniques for subject selection will vary quite a that specifies where the money will go. Large, bit. Happily, Chapter 7 of this book discusses sam- expensive projects include budgetary categories pling techniques for both qualitative and quantita- such as personnel, equipment, supplies, tele- tive studies. phones, and postage. Even for a project you’ll pay for yourself, it’s a good idea to spend some time Measurement anticipating expenses: office supplies, photocopy- ing, CD-ROMs, telephone calls, transportation, What are the key variables in your study? How will and so on. you define and measure them? Do your definitions and measurement methods duplicate or differ from As you can see, if you’re interested in conduct- those of previous research on this topic? If you ing a social research project, it’s a good idea to have already developed your measurement device prepare a research proposal for your own pur- (a questionnaire, for example) or will be using poses, even if you aren’t required to do so by your something previously developed by others, it might instructor or a funding agency. If you’re going to be appropriate to include a copy in an appendix to invest your time and energy in such a project, you your proposal. should do what you can to insure a return on that investment. Data-Collection Methods Now that you’ve had a broad overview of social How will you actually collect the data for your research, you can move on to the remaining chap- study? Will you conduct an experiment or a ters in this book and learn exactly how to design survey? Will you undertake field research or will and execute each specific step. If you’ve found a you focus on the reanalysis of statistics already cre- research topic that really interests you, you’ll want ated by others? Perhaps you’ll use more than one to keep it in mind as you see how you might go method. about studying it. As always, however, you should keep the ethical dimension of research design in Analysis mind as you explore your options. Indicate the kind of analysis you plan to conduct. The Ethics of Research Design Spell out the purpose and logic of your analysis. Are you interested in precise description? Do you Designing a research project must include serious intend to explain why things are the way they consideration of the ethical dimension of social are? Do you plan to account for variations in some science research. To begin, if your study requires quality: for example, why some students are more the participation of human subjects, you must liberal than others? What possible explanatory vari- determine whether the likely benefits of the re- ables will your analysis consider, and how will you search will justify the time and effort you’ll ask of know if you’ve explained variations adequately? them. Schedule Then you’ll want to design the study in con- currence with the ethical concerns discussed in It’s often appropriate to provide a schedule for the Chapter 3. For example, you’ll want to insure that various stages of research. Even if you don’t do this the subjects’ privacy and well-being is protected. As for the proposal, do it for yourself. Unless you have I’ve indicated earlier, it may be appropriate for your a timeline for accomplishing the several stages of research design to be reviewed by an Institutional research and keeping track of how you’re doing, Review Board. you may end up in trouble.
Main Points ■ 121 MAIN POINTS research is the individual person, but it may also be a social group, a formal organization, a social Introduction interaction, a social artifact, or some other phe- nomenon such as a lifestyle. • Any research design requires researchers to • The ecological fallacy involves taking conclusions specify as clearly as possible what they want to find out and then determine the best way to drawn solely from the analysis of groups (e.g., do it. corporations) and applying them to individuals (e.g., the employees of corporations). Three Purposes of Research • Reductionism is the attempt to understand a • The principal purposes of social research in- complex phenomenon in terms of a narrow set clude exploration, description, and explanation. of concepts, such as attempting to explain the Research studies often combine more than one American Revolution solely in terms of economics purpose. (or political idealism or psychology). • Exploration is the attempt to develop an initial, The Time Dimension rough understanding of some phenomenon. • Research into processes that occur over time • Description is the precise measurement and presents social challenges that can be addressed through cross-sectional studies or longitudinal reporting of the characteristics of some population studies. or phenomenon under study. • Cross-sectional studies are based on observations • Explanation is the discovery and reporting of made at one time. Although this characteristic relationships among different aspects of the phe- limits such studies, researchers can sometimes use nomenon under study. Whereas descriptive stud- them to make inferences about processes that oc- ies answer the question “What’s so?” explanatory cur over time. ones tend to answer the question “Why?” • In longitudinal studies, observations are made The Logic of Nomothetic Explanation at many times. Such observations may be made • Both idiographic and nomothetic models of of samples drawn from general populations (trend studies), samples drawn from more explanation rest on the idea of causation. The specific subpopulations (cohort studies), or the idiographic model aims at a complete under- same sample of people each time (panel standing of a particular phenomenon, using all studies). relevant causal factors. The nomothetic model aims at a general understanding—not necessarily How to Design a Research Project complete— of a class of phenomena, using a small number of relevant causal factors. • Research design starts with an initial interest, idea, • There are three basic criteria for establishing cau- or theoretical expectation and proceeds through a series of interrelated steps to narrow the focus sation in nomothetic analyses: (1) The variables of the study so that concepts, methods, and must be empirically associated, or correlated, procedures are well defined. A good research plan (2) the causal variable must occur earlier in time accounts for all these steps in advance. than the variable it is said to affect, and (3) the observed effect cannot be explained as the effect • At the outset, a researcher specifies the meaning of a different variable. of the concepts or variables to be studied (con- Necessary and Sufficient Causes ceptualization), chooses a research method or methods (e.g., experiments versus surveys), and • Mere association, or correlation, does not in itself specifies the population to be studied and, if ap- plicable, how it will be sampled. establish causation. A spurious causal relationship is an association that in reality is caused by one or • To operationalize the concepts to be studied, the more other variables. researcher states precisely how variables in the Units of Analysis study will be measured. Research then proceeds through observation, data processing, analysis, • Units of analysis are the people or things whose and application, such as reporting the results and assessing their implications. characteristics social researchers observe, describe, and explain. Typically, the unit of analysis in social
122 ■ Chapter 4: Research Design The Research Proposal the time order is just the opposite of what was assumed? • A research proposal provides a preview of why 2. Here are some examples of real research topics. a study will be undertaken and how it will be For each one, can you name the unit of analysis? conducted. A research proposal is often required (The answers are at the end of this chapter.) to get permission or necessary resources. Even when not required, a proposal is a useful device a. Women watch TV more than men because for planning. they are likely to work fewer hours outside the home than men. . . . Black people watch The Ethics of Research Design an average of approximately three-quarters of an hour more television per day than white • Your research design should indicate how your people. (Hughes 1980: 290) study will abide by the ethical requirements of b. Of the 130 incorporated U.S. cities with more social research. than 100,000 inhabitants in 1960, 126 had at least two short-term nonproprietary general • It may be appropriate for your research proposal hospitals accredited by the American Hospital Association. (Turk 1980: 317) to be reviewed by an Institutional Review Board. c. The early TM [transcendental meditation] KEY TERMS organizations were small and informal. The Los Angeles group, begun in June 1959, met The following terms are defined in context in the at a member’s house where, incidentally, chapter and at the bottom of the page where the Maharishi was living. (Johnston 1980: 337) term is introduced, as well as in the comprehensive glossary at the back of the book. d. However, it appears that the nursing staffs exercise strong influence over . . . a decision cohort study reductionism to change the nursing care system. . . . Con- correlation social artifact versely, among those decisions dominated by cross-sectional study sociobiology the administration and the medical staffs . . . ecological fallacy spurious relationship (Comstock 1980: 77) longitudinal study trend study panel study units of analysis e. Though 667,000 out of 2 million farmers in the United States are women, women histori- PROPOSING SOCIAL RESEARCH: DESIGN cally have not been viewed as farmers, but rather, as the farmer’s wife. (Votaw 1979: 8) This chapter has laid out many different ways social research can be structured. In designing your research f. The analysis of community opposition project, you will need to specify which among these to group homes for the mentally handicapped you will use. Is your purpose that of exploring a topic, . . . indicates that deteriorating neighborhoods providing a detailed description, or explaining the are most likely to organize in opposition, but social differences and processes you may observe? If that upper-middle class neighborhoods are you are planning a causal analysis, you should say most likely to enjoy private access to local something about how you will organize and pursue officials. (Graham and Hogan 1990: 513) that goal. g. Some analysts during the 1960s predicted that Further, will your project collect data at one point the rise of economic ambition and political in time or compare data across time? What data col- militancy among blacks would foster discon- lection technique(s) will you employ? You will revisit tent with the “otherworldly” black mainline these and similar questions as you delve into your churches. (Ellison and Sherkat 1990: 551) project. h. This analysis explores whether propositions REVIEW QUESTIONS AND EXERCISES and empirical findings of contemporary theories of organizations directly apply to 1. One example in this chapter suggested that politi- both private product producing organizations cal orientations cause attitudes toward legalizing (PPOs) and public human service organiza- marijuana. Can you make an argument that tions (PSOs). (Schiflett and Zey 1990: 569) i. This paper examines variations in job title structures across work roles. Analyzing 3,173
Online Study Resources ■ 123 job titles in the California civil service system 2. As you review, take advantage of the CengageNOW in 1985, we investigate how and why lines of personalized study plan, based on your quiz work vary in the proliferation of job categories results. Use this study plan with its interactive ex- that differentiate ranks, functions, or particu- ercises and other resources to master the material. lar organizational locations. (Strang and Baron 1990: 479) 3. When you’re finished with your review, take the posttest to confirm that you’re ready to move on 3. Review the logic of spuriousness. Can you think to the next chapter. up an example where an observed relationship between two variables could actually be explained WEBSITE FOR THE PRACTICE away by a third variable? OF SOCIAL RESEARCH 12TH EDITION 4. Using InfoTrac College Edition or printed journals Go to your book’s website at www.cengage.com/ in the library, locate a research project involving sociology/babbie for tools to aid you in studying for a panel study. Describe the nature of the study your exams. You’ll find Tutorial Quizzes with feedback, design and its primary findings. Internet Exercises, Flash Cards, Glossaries, and Essay Quiz- zes, as well as InfoTrac College Edition search terms, sug- SPSS EXERCISES gestions for additional reading, Web Links, and primers for using data-analysis software such as SPSS. See the booklet that accompanies your text for ex- ercises using SPSS (Statistical Package for the Social ANSWERS TO UNITS OF ANALYSIS QUIZ, EXERCISE 2 Sciences). There are exercises offered for each chapter, and you’ll also find a detailed primer on using SPSS. a. Men and women, black and white people (individuals) Online Study Resources b. Incorporated U.S. cities (groups) If your book came with an access code card, visit c. Transcendental meditation organizations (groups) www.cengage.com/login to register. To purchase d. Nursing staffs (groups) access, please visit www.ichapters.com. e. Farmers (individuals) 1. Before you do your final review of the chapter, f. Neighborhoods (groups) g. Blacks (individuals) take the CengageNOW pretest to help identify the h. Service and production organizations (formal areas on which you should concentrate. You’ll find information on this online tool, as well as organizations) instructions on how to access all of its great re- i. Job titles (artifacts) sources, in the front of the book.
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 534
- 535
- 536
- 537
- 538
- 539
- 540
- 541
- 542
- 543
- 544
- 545
- 546
- 547
- 548
- 549
- 550
- 551
- 552
- 553
- 554
- 555
- 556
- 557
- 558
- 559
- 560
- 561
- 562
- 563
- 564
- 565
- 566
- 567
- 568
- 569
- 570
- 571
- 572
- 573
- 574
- 575
- 576
- 577
- 578
- 579
- 580
- 581
- 582
- 583
- 584
- 585
- 586
- 587
- 588
- 589
- 590
- 591
- 592
- 593
- 594
- 595
- 596
- 597
- 598
- 599
- 600
- 601
- 602
- 603
- 604
- 605
- 606
- 607
- 608
- 609
- 610
- 611
- 612
- 613
- 614
- 615
- 616
- 617
- 618
- 619
- 620
- 621
- 622
- 623
- 624
- 625
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 625
Pages:
