Appendix C

The Shaffer et al.(1995) study: Sexual Orientation in Adolescents who Commit Suicide

The Shaffer et al.(1995) study data: The fact that the Shaffer et al.(1995) researchers did not obtain any "homosexuality" information from the 145 control teenagers in the study suggests that they were not skilled in obtaining such taboo information from teenagers. Studies reveal that at least 5% of male teenagers would be in the "homosexuality" categories investigated(88,105). As noted in Appendix A, however, these studies suffer from problems of methodology, such as concealment occurring at rates which produce significant underestimates of the degree to which "homosexuality" exists in the male population.

With respect to the assumed 100 control males studied, statistically some of them should have reported having had same-sex experiences. Bagley(1994) reported that 6 percent of males in the Bagley et al.(1994) sample had experienced repeated but unwanted sexual experiences with a man before the age of 15 (which is average for the male population), with another 8 to 10 percent reporting such experiences as a one-time event. Bagley(1994) also reported that 3% of males had been consensually and repeatedly sexually involved with at least one man before the age of 15.27 Overlapping with these respondents were another 5% of males having sexual experiences with at least one other male approximately their own age before the age of 15. Therefore, from the Shaffer et al.(1995) random sampling for about 100 males, a minimum of 12% (about 12 males) should have reported same-sex experiences as either wanted (6%) and unwanted (6%). None of the control males, however, reported having had any same-gender sexual experiences.

With respect to the 95 adolescent males who committed suicide, only 3 were deemed to meet the criteria set by the researchers, either revealing their gay identity to others (one case), or having had same-gender sex. The methods used to determine such facts about a teenage boy, however, are faulty. I was homosexually active from the age of 5 to 19 with neither parents or teachers knowing anything about this. If I had committed suicide, my friends would never have never revealed what was happening between us sexually. To do this would have required them to admit their own involvement, and only highly skilled researcher could have obtained the required information from them; likewise if the situation was reversed and information was being culled from me if a friend I was having sex with had committed suicide. The Shaffer et. al.(1995) researchers do not appear to have such skills.

Uribe & Harbeck(1992) reported that most of the males in their sample of 37 high school gay/bisexual males, "did what they could to conceal their homosexuality because to affirm it was too painful. They devised elaborate concealment strategies, and the result of these strategies was to cripple them emotionally and socially... Thirty-five out of 37 were homosexually active when interviewed, and had been since the average age of 14. For the majority, their first sexual experiences were "with unknown males"(37:19-22). I have also found such behaviour to be common enough in Calgary with gay boys who were terrified of having their homosexual identity discovered by their classmates, friends, siblings, teachers, and especially their parents.

What, therefore, could a researcher have done to discover that one such teenage males was homosexually active if he had committed suicide? Ask his parents? His teachers? And what if the teacher or counsellor knows the truth because he was having sex with the boys? Would a friend designated by his parents know? I doubt it. And if so, will the friend talk? Obviously, before doing their study, the Shaffer et al.(1995) researchers did not do their homework.

To date, researchers still have not asked gay/bisexual male teenagers who made suicide attempts rated in the "moderate to high lethality" category questions like:

  1. If you had died, would your parents or teachers have known that you were gay and homosexually active? Would a friend whom your parents had told a researcher to talk to have known the truth about you? If so, do you think this person would have revealed your secret to anyone?
  2. If you had died, is there a way that a good investigator could have obtained the information needed to prove that you were gay or bisexual, and maybe homosexually active?
The results of such questions would certainly have given the Shaffer researchers an estimate of the margin of error they could expect by using their method of determining the "homosexual" designation for suicide victims. This margin of error could have also permitted them to estimate the actual percentage of homosexual male suicide victims. Instead, the researchers concluded that, on the basis of the 3 suicide victims proven to be gay and/or homosexually active, the "study finds no evidence that suicide is a common characteristic of gay youth."(104:64) Without doubt, Shaffer et. al., failed in one of the "overarching principles" listed in the same volume of SLTB in which their study was published. "Researchers must know the community they are studying."(106:83)

Far more troubling, however, is the conclusion: "...that when suicide does occur among gay teenagers, that it is not a direct consequence of stigmatization or lack of support."(104:64) This conclusion is based only on three suicides of known homosexual male teenagers and is equivalent to concluding - from the Bagley(1994) result indicating 3 homosexually active gay/bisexual males who had attempted suicide, of which 2 had been sexually abused - that gay and bisexual males primarily attempt suicide because of being sexually abused as children. Or, alternatively, that they do not attempt suicide for a particular reason because the factor was not present. The only conclusion which can be made from the Bagley(1994) data is related to the percentage of young adult males who were homosexually active as young adults and had attempted suicide during their youth.

Anything significant about the possible factors related to the suicide attempts for gay or bisexual males - such as a particular problem or the cumulative effects of many interrelated problems - must be studied by obtaining a large enough representative sample of homosexually active gay and bisexual males who have attempted suicide. Only with such samples can the statistical significance or insignificance of postulated factors be determined.

Finally, the Shaffer et al.(1995) study produced four victims in the effeminate category. Remafedi et al.(1991) reported that effeminate gay males were three times more likely to attempt suicide than other gay males(31), thus confirming what gay males have always known. Effeminate males, since early childhood, receive the brunt of society's hatred for gay males, and are therefore the most stigmatized of all male youth. They are also strongly suspected to be homosexually oriented, even though Shaffer et. al opposed this conclusion.

The Bell & Weinberg(1981) study did not have one heterosexual male in their sample rate themselves in the 0 to 2 category on a 0-6 feminine to masculine scale, while 28% of predominantly gay males gave themselves this "label" for the period defined to be "while you were growing up."(113:75) This study also determined that gender nonconformity was the most distinguihing feature between gay and heterosexual males(22:79-81). In the 1987 Sissy Boy Syndrome study, the 44 "feminine" boys studied over many years became young adult men with a 75% probability of being gay or bisexual, as rated by the Kinsey 0-6 fantasy/behaviour scale. A control group of 35 "conventionally 'masculine' boys" only produced one young adult male in the bisexual category(115:99-101).

Therefore, noticeably effeminate males are most likely gay, or they may be (mostly heterosexual) transvestites or transsexuals who form a very small percentage of the male population. That 4 out of the 95 male suicide victims were effeminate suggests that these male are at much higher risk for suicide than other males. Shaffer et al.(1995) were nonetheless silent about this, even if it was noted that not one of the males in the control group was rated to be "abnormally effeminate" using the same scale used to define victims to have been effeminate(104:70).

[Main Page]
25 Nov 1995