RECENT SEX SURVEYS: SURVEYS OF SPECIAL POPULATIONS
In recent years, there have been several major surveys of the sexual attitudes and behaviors of special, more restricted populations. Two of these surveys (Kantner and Zelnik, 1972 and 1973; Sorenson, 1973) investigated teen-age Americans. Using probability sampling (a technique for maximizing the representativeness of a sample and thus the generalizability of results), Kantner and Zelnik surveyed females between the ages of 15 and 19. Since this sample is fairly representative of the female teen-age population at large, it has provided much useful data with respect to the relationships between various demographic factors and sexuality. Of special import here were the analyses of black-white differences, since about one-third of the approximately 4,600 respondents were black.
Sorenson (1973) surveyed adolescents of both sexes between the ages of 13 and 19. Using accepted sampling techniques to identify over 2,000 households containing 839 adolescents, Sorenson then sought informed consent to participate from both the adolescents and their parents. This yielded a total sample of 393 respondents. Thus, over 50% of the participants identified initially did not take part in the study, certainly a limiting factor because we cannot know precisely the effects of this selection bias. It should be noted that this high nonresponse rate was due mostly to the parents' failure to give their consent since almost 80% of the adolescents agreed to participate.
As part of a broader study of American women of reproductive age, Westoff (1974) gathered some data on female sexuality. Some 10,000 women were questioned, approximately one-half in 1965 and the others in 1970. Like the Kantner and Zelnik study, Westoff used probability sampling techniques and managed to obtain a sample that was fairly representative of American women in this age group as a whole. Two additional aspects of this study are relevant. First, since most of the questions were of a nonsexual nature, such as attitudes about family size, the questionnaire as a whole would seem to have been much less threatening to the respondents than a questionnaire dealing exclusively with sexuality. Therefore, the women chosen were probably more likely to participate and also to respond in an honest fashion, thus increasing the generalizability of the results. This speculation seems to be confirmed by the fact that the refusal rate for responding to the sexual questions was consistently low, typically less than 10%. Secondly, since the two samples were surveyed five years apart, this study has been useful in assessing possible changes in sexual attitudes and behaviors over that period of time.
Men's Health-Erectile Dysfunction