Collins uses data from the U.K. National Survey on Sexual Attitudes and Lifestyles from 1990-91. The sample covered adults aged 16-59, with a reasonable sample size of over 4,500 responses. The main results are based on a regression analysis where the dependent variable is whether a respondent gave 'being in love' as the reason for their loss of virginity.
It turns out that there are very few explanatory variables that are associated with losing virginity in love - education, how the respondent learned about sex, belief in God, and religion were mostly insignificant (the exception being that Muslims had 73 percent lower odds of reporting losing their virginity in love than the control religious category (those who were not Catholic, Hindu/Sikh, Muslim, or Jewish).
The interesting results are the differences between men and women, in terms of the effect of age. In the paper, Collins misinterprets his results as showing differences between women when younger and when older. However, because all respondents were surveyed at the same time (1990-91), then age is really a proxy for cohort effects (we can't separately control for age at loss of virginity, and the year that virginity was lost, but assuming that most people lose their virginity about the same age, i.e. not decades apart, then this probably isn't too far from correct). These cohort effects tell us whether love is becoming more, or less, important over time in the virginity loss decision (it doesn't tell us differences between younger and older people at the time they lost their virginity). Because respondents were aged 16-59 at the time of the survey (average age around 35), then the time period (for loss of virginity) covered is approximately the late 1940s to the late 1980s.
In terms of results, it turns out that love has become no more or less important for men in the U.K. over this time period - each additional year of age is associated with a 0.1 percent decrease in odds of reporting that their virginity was lost in love (and the effect is very statistically insignificant). For women though, each additional year of age is associated with a 1.6 percent decrease in odds of reporting that their virginity was lost in love (and statistically significant). So, for women love is becoming less important for virginity loss over time (extrapolating over the 40-year period 1940s-1980s, a 40-year difference in age equates to a 48 percent lower odds of reporting losing virginity in love).
This result has the effect of reducing the difference in responses between men and women (women had 3.4 times higher odds overall of reporting losing their virginity in love than men). Sadly (for the romantics among us), this tells us that love is probably becoming less important over time, for women.
- Is sexual freedom a normal good?
- More sex might increase happiness, but not if an economist tells you to
- Are pornography and marriage substitutes for young men?
- Economics majors have the most sex!
- Is sex like driving? HIV prevention and risk compensation
- An un-recognised additional cost of higher education