Regardless of the survey strategy used, sex researchers find that it is difficult to secure a representative sample. This occurs because many people do not want to participate in sex studies. For instance, assuming that you used proper sampling procedures to choose your sample of older married couples in the example discussed earlier, what proportion of your representative sample would actually be willing to answer your questions? Nonresponse, the refusal to participate in a research study, is a common problem that consistently plagues sex survey research (Turner, 1999; Wiederman, 2001).
No one has ever conducted a major sex survey in which 100% of the selected subjects voluntarily participated. In fact, some studies include results obtained from samples in which only a small minority responded. This situation raises an important question: Are people who agree to take part in sex surveys any different from those who refuse?
Perhaps volunteer subjects in sex research are a representative cross section of the population, but we have no theoretical or statistical basis for that conclusion. As a matter of fact, the opposite might well be true. People who volunteer to participate may be the ones who are the most eager to share their experiences, who have explored a wide range of activities, or who feel most comfortable with their sexuality. (Or it may be that the most experienced people are those who are least willing to respond because they feel that their behaviors are atypical or extreme.) A preponderance of experienced, inexperienced, liberal, or conservative individuals can bias any sample.
Research suggests that self-selection, or volunteer bias, is an important concern for sex researchers (Plaud et al., 1999; Wiederman, 2001). Studies strongly suggest that volunteers for sex research are more sexually experienced and hold more positive attitudes toward sexuality and sex research than do nonvolunteers (Boynton, 2003; Plaud et al., 1999). In addition, research indicates that women are less likely than men to volunteer for sex research (Boynton, 2003; Plaud et al., 1999), a finding that suggests female sex research samples may be more influenced by volunteer bias than male samples.
Sex Research: Methods and problems
|
demographic bias A kind of sampling bias in which certain segments of society (such as White, middle-class, white-collar workers) are disproportionately represented in a study population. |
Another problem inherent in sex survey research has to do with the accuracy of subjects’ responses. Most data about human sexual behavior are obtained from respondents’ own reports of their experiences, and people’s actual behavior can be quite different from what they report (Catania, 1999; Ochs & Binik, 1999). How many people accurately remember when they first masturbated, and with what frequency, or at what age they first experienced orgasm? Some people may also distort or falsify their self-disclosures to maintain or even enhance their social image (Catania, 1999). This tendency to provide socially desirable responses can involve people who consciously or unconsciously conceal certain facts about their sexual histories because they view them as abnormal, foolish, or painful to remember. People can also feel pressure to deny or minimize their experiences regarding behaviors for which strong taboos exist, such as incest, homosexuality, and masturbation. In other cases, people can purposely inflate their experience, perhaps out of a desire to appear more liberal, experienced, or proficient. A third type of problem that affects sex surveys is demographic bias. Most of the data available from sex research in the United States have come from samples weighted heavily toward White, middle-class volunteers. Typically, college students and educated white-collar workers are overrepresented. Ethnic and racial minorities and less — educated individuals are underrepresented. How much of an effect do nonresponse and demographic bias have on sex research findings? We cannot say for sure. But as long as elements of society, including the less educated and ethnic and racial minorities, are underrepresented, we must be cautious in generalizing findings to the population at large. The informational deficit pertaining to sexual behaviors of American racial and ethnic minorities is lessening with the emergence of several studies that included members of various ethnic minority groups in the United States. These studies demonstrate that ethnicity often exerts considerable influence on sexual attitudes and behaviors (Okazaki, 2002). Throughout this textbook, we describe results from several studies as we discuss ethnicity in relation to a variety of topics.
Психология взаимоотношений © 2022
Frontier Theme
|