RESULTS

Psychological Adjustment 

Table I 
Self-Esteem and Age of Attainment of Positive Sexual Identity as a Function of Having Experienced ADSR

The ADSR Experiences 
Characteristics 
Reactions 
Consent 
Correlations 

Table II 
Correlations Among ADSR Characteristics

Psychological Adjustment 

If ADSRs have adverse effects on the adjustment of gay /bisexual males, it would be expected that, in comparison with controls, ADSR subjects should have lower self-esteem and greater difficulty in attaining a positive sexual identify (i.e., the age of achieving this milestone should be delayed).

In Sample 1, self-esteem scores were not lower for ADSR subjects (M = 23.82) than for control subjects (M = 23.46), t(37) = -.17, p > .10 (all tests reported here are two-tailed), with effect size r = -.03 (positive effect sizes indicate better adjustment for controls; negative effect sizes indicate better adjustment for ADSR subjects). 

Attainment of a positive sexual identity was not delayed for ADSR subjects (M = 18.00) compared to controls (M = 18.35), t(23) = -.43, p > .10, r = -.09. Assessment of this attribute commenced a third of the way into the interviews; data are missing for nearly equal proportions of ADSR (31% ) and control (33% ) subjects. Of those who were asked about achieving a positive sexual identity, the proportion of ADSR subjects (89% ) and control (85% ) subjects who had achieved this milestone did not differ, z = -.28, p > .10, r = -.05. Table I provides adjustment statistics for both samples.

Table I  
Self-Esteem and Age of Attainment of Positive Sexual Identity as a Function of Having Experienced ADSR

In Sample 2, self-esteem scores were not lower for ADSR-identified subjects (M = 21.00) than for control subjects (M = 21.96), t(84) = .62, p > .10, r = .07. Attainment of a positive sexual identity occurred earlier rather than later for ADSR-identified subjects (M = 16.80) relative to controls (M = 19.10), t(70) = -2.89, p < .01, r = -.33. The proportion of ADSR-identified subjects (77%) and control subjects (85% ) achieving a positive sexual identity did not differ, z = .72, p > .10, r = .08.

Sample 1 results provide no evidence for adverse effects of ADSR. Sample 2 results, although in need of qualification because the control sample was likely to be impure, are consistent with those of Sample 1, reinforcing the conclusion of no evidence for adverse effects. 

Combining results from the two samples meta-analytically (Rosenthal, 1984) yielded a very small and statistically non-significant effect size for self-esteem (r = .04, N = 125,95% confidence interval = -.14 to +.21), a medium and statistically significant effect size for age of positive sexual identity (r = -.27, N = 97,95% confidence interval = -.45 to -.07), and a small and statistically non-significant effect size for proportion having achieved a positive sexual identity (r = .05, N = 115,95% confidence interval = -.14 to +.23). All of these results are inconsistent with the traumagenic view (e.g., incest model) of ADSR, particularly the negative effect size for age of achieving a positive sexual identity, which is consistent instead with a beneficial effect. 

The ADSR Experiences

The Appendix contains narratives from all 26 ADSR subjects, which are ordered according to subjects' and partners' ages. The narratives generally provide information on the context in which the ADSRs occurred, the level of familiaity between the partners, the frequency and duration of the sexual relationships, the types of sex involved, subjects' reactions, and their level of consent. Savin-Williams took more notes on first sexual or romantic experiences; as such, these narratives contain more details.

Characteristics

The mean age at which subjects had their first ADSR experience was 15.31 (SD = 1.67), with a range from 12 to 17. The mean age of their older partner was 28.62 (SD = 7.66), with a range from 20 to 46. Thus, on average, there was a 13.31 (SD = 7.83) year age difference between the boys and their older partners, with a range from 5 to 30 years. Forty two percent of the cases involved contacts with strangers; 35% involved contacts with acquaintances; and 23% involved friends (one of these involved an older brother). About two thirds (68%) of the cases involved multiple sexual contacts.  Nearly half of the sexual relationships, (42%) lasted less than a month, whereas a quarter (25%) lasted more than a year. 

Of the 19 cases for which type of sex could be identified, 21% involved masturbation as the most intense form, 42% involved oral sex, and 37% involved anal intercourse. 

Almost all subjects (96% ) were aware of their sexual attraction to males before their ADSR experience-mean age of awareness was 7.92 (SD = 4.10), with a range from 3 to 17. Ninety-six percent had reached puberty at an earlier I age (one reached puberty in the same year as his ADSR experience) -- mean age of puberty was 11.46 (SD = 1.21), with a range from 10 to 14. Three quarters (76%) had already labeled their interests "homosexual" or "gay" before the ADSR occurred (16% labeled their interests in the same year as the ADSR) -- mean age of labeling was 12.52 (SD = 3.02), with a range from 7 to 18. These results regarding age of awareness of attractions and labeling call into question the role of ADSRs in causing same-sex interests.

Reactions

Subjects' mean reaction was positive (M = 3.94, SD = 1.25), although individual reactions ranged from very negative to very positive. Overall, reactions were as follows: 38.5% very positive, 38.5% positive, 7.7% neutral/mixed, 3.8% negative, and 11.5% very negative. Combining categories and rounding, 77% were positive, 8% were neutral, and 15% were negative.

Consent

Overall, subjects were mutually consenting (M = 4.15, SD = .51); consent ranged from acquiescing to encouraging. Thus, forced or coerced contact was not a factor in this sample. To the contrary, nearly a quarter (23.1 %) encouraged the contacts and about two-thirds (69.2%) mutually consented; 7.7% acquiesced. Thus, 92% evidenced positive desire for the sexual involvement.

Correlations

Table II presents correlations among the various ADSR characteristics; statistical significance is based on two-tailed tests. Doll et al. (1992) reported that greater age difference was associated with more negative reactions at time of interview in their sample of gay /bisexual men. In the current sample, this association was not found, r(24) = .12, p > .10. Moreover, younger boys did not react more negatively (or less positively) than older ones, r(24) = -.23, p > .10, and they were just as consenting as older boys, r(24) = -.01, p > .10. Further, contrary to age difference posing a problem for the boys, their willingness and interest in participating in sexual relations increased as the difference in ages between them and the men increased, r(24) = .39, p < .05, and as the ages of their partners increased, r(24) = .40, p < .05. 

Positivity of reactions increased with a greater degree of familiarity with the men, r(24) = .56, p < .01, multiple as opposed to single sexual encounters, r(23) = .60, p < .01, longer lasting sexual relationships, r(22) = .52, p < .01, and greater willingness and interest in participation, r(24) = .43, p < .05. Greater familiarity was associated with more frequent sexual encounters, r(23) = .47, p < .05, and longer lasting sexual relationships, r(22) = .69, p < .01. Younger boys tended to be more familiar or friendly with their partners, r(24) = -.43, p < .05. Finally, two one-way ANOVAs were performed to examine whether reactions and consent were related to type of sex experienced. Results were non-sigificant in both cases: F(2, 16) = .25, p > .10, and F(2, 16) = .36, p > .10,  respectively. 

Table II. Correlations Among ADSR Characteristics

Note. 
Boys' age = boys' ages when ADSR began; 
men's age = men's ages when ADSR began; 
age difference = difference between partners' ages; 
relationship = familiarity between the partners; 
frequency = single vs. multiple sexual contacts; 
duration = length of sexual relations;
reaction ranges from very negative to very positive; 
consent ranges from forced to encouraging (see text for exact definitions). 
Correlations based on N = 26, except for those involving frequency or duration, which involve N = 25 and N = 24, respectively.
* p < .05 two-tailed; 
** p < .01 two-tailed.