A THEORETICAL INVESTIGATION OF PUBLIC ATTITUDES TOWARD SEX EDUCATION

Abstract

Rates of teen pregnancy in the United States are significantly higher than those in most other industrialized nations. This is troubling in light of the consequences to teen mothers, their children, and society. Some suggest that teen pregnancies can be decreased by providing sex education in the public school system, while others believe that sex education is inappropriate for public school. Little research exists, however, on predictors of attitudes toward sex education, and most existing research is dated and lacks a theoretical framework. Guided by the seminal work of Ira L. Reiss, we examine the impact of labor shortages, religiosity, militarism, gender egalitarianism, regulation of sexuality, and a naturalistic view of sexuality on attitudes toward sex education. Results suggest that Hispanics, those with high levels of religiosity, and those who supported regulation of sexuality were significantly less likely to support sex education in public schools. Implications of the findings and directions for future research are discussed.

Notes

¹Reiss (2006) points to a number of changes that may affect views toward sexuality, including the onset of AIDS, the proliferation of “right wing religious groups” (Reiss 2006, p. 210), and the Bush Administration's opposition to funding for sex research. Similarly, the fact that the Supreme Court recently held that a ban on partial birth abortion does not violate a women's right to choose is seen by some as a move toward more conservative policies toward sexuality and women's rights.

²According to Reiss (2006), his theory is more applicable to attitudes rather than actual behaviors.

³They have recently passed a law that criminalizes the customer of the prostitute, but not the prostitute (Weinberg et al. 2000).

⁴If a respondent answered that it would be easy to find another job, it implies that there is a labor shortage; if he or she answered that it would not be easy to find another job, it indicates that there is not a labor shortage.

⁵We recognize that this is a weak measure of labor shortages because it does not directly measure job availability or the state of the economy. Rather, it measures an individual's personal perception of the ease/difficulty of finding another job, which may be related to type of occupation, educational attainment, or another variable not captured here.

Coefficients are odds ratios.

∗p < .10.; ∗∗p < .05; ∗∗∗p < .01.

⁶While it is common to compare R ² values across models for linear regression, such comparisons are less meaningful for logistic regression. For this reason, we generated a deviance statistic (or likelihood ratio test) in order to compare subsequent, more complex models to the first model. This statistic is calculated by subtracting the −2 Log likelihood of the more complex model from that of the simpler model. The difference follows the chi-square distribution, where degree of freedom is equal to the additional parameters in the more complex model (Agresti and Finlay 1997).

⁷It should be noted that the final model loses some statistical power due to decreased sample size (missing data) and this could explain the lack of statistical significance of some of the variables.