The Cameron group’s obituary study

Source: Psychology Dept, Univ of California, Davis
Undated, but appears to be from the mid 1990s

Cameron, Playfair, and Wellum (1994) counted obituaries in various gay community publications and claimed to be able to use them to calculate the average life expectancy for homosexuals.

Their conclusion – that homosexual men and women have a shorter life span than heterosexual men and women – provides a textbook example of the perils of using data from a convenience sample to generalize to an entire population.

Death Notices and Obituaries

Most city newspapers include a section containing death notices for community residents. These notices – which can carry a small fee for printing – typically list the name, age, address, and survivors of the deceased, along with information about funeral or memorial services. Funeral directors often assist the loved ones of the deceased in submitting such notices.

Gay community newspapers do not have sections of death notices. When the AIDS epidemic began to claim the lives of so many gay and bisexual men in the 1980s, however, many gay newspapers began to print obituaries. Except in the case of prominent community figures, these obituaries are typically written by (or based on information from) the loved ones of the deceased.

Assuming that the deceased person wasn’t famous, an obituary appears in a gay community newspaper only if (1) a loved one or friend notifies the newspaper about the death (and, in many cases, writes the obituary) and (2) the editor decides to print the obituary.

Consequently, many gay men and lesbians who die never have an obituary in a gay community publication. Here are just a few examples of who is left out of gay newspapers’ obituaries.

* gay men and lesbians who were not involved in the gay community

* gay men and lesbians who were in the closet about their sexual orientation

* gay men and lesbians whose loved ones or family didn’t want their homosexuality to be known

* gay men and lesbians whose loved ones or family simply didn’t think of sending an obituary to a gay community newspaper

* gay men and lesbians whose loved ones did not write an obituary for some other reason (e.g., they were too grief stricken)

* gay men and lesbians who died without leaving anyone to write an obituary for a gay publication (e.g., those whose loved ones and relatives died before them).

An accurate estimate of the life span of gay men and lesbians would have to count such people. By restricting their analysis to obituaries in gay newspapers, however, the Cameron group systematically excluded them from the sample.

Internal Inconsistencies

The inadequacy of the Cameron group’s approach is evident from internal inconsistencies within their own data. Compare the data about lesbians reported in their obituary study, for example, to data from their so-called national survey.

In their obituary study, the Cameron group claimed that the average lesbian life-span is similar to that of gay men who do not have AIDS (“under 50 years” versus “mid-40s,” respectively). But if this is true, and if obituaries are indeed a valid source for this type of data, the ratio of gay male obituaries to lesbian obituaries should be about the same as the ratio of gay men to lesbians in the population.

From their survey data, the Cameron group has claimed to know the number of gay men and lesbians in the population. If we believed their numbers, we would set the ratio of gay men-to-lesbians at about 1.6-to-1 (or approximately 2.6-to-1 if bisexuals are omitted).

But the ratio of gay male-to-lesbian obituaries in the Cameron group’s study is quite different – approximately 6-to-1 if AIDS and violent deaths are excluded, 32-to-1 if they are included.

Thus, at least one data set has to be wrong. Either the obituaries data do not include a representative sample of lesbians, or the Cameron group’s population estimates based on their survey data are invalid.

An observer with training in research methodology would most likely conclude that both sets of data are fatally flawed.

This example is provided as simply one illustration of the flaws in the Cameron group’s methods.

Conclusion

Obituaries in gay community newspapers do not provide a representative sampling of the community. This is evident in the fact that only only 2% of the Cameron group’s obituaries were for lesbians. Moreover, community newspapers tend overwhelmingly to report deaths due to AIDS (only 11% of Cameron’s gay male obituaries were not related to AIDS). In addition, community newspapers tend not to print obituaries for people who are not actively involved in the local gay community, those who are in the closet, and those whose loved ones simply don’t submit an obituary to a local gay newspaper.

The Cameron group’s gay obituary study reports many numbers and statistics. However, they are absolutely worthless for estimating the life expectancy of gay men and lesbians.

Postscript

In a 1997 column in the Weekly Standard, former Secretary of Education William Bennett referred to the findings of Cameron et al.’s obituary study, although he did not cite Cameron by name. He again referred to Cameron’s conclusion about the truncated life span of homosexuals in an appearance on ABC’s “This Week” program.

In 1998, after Andrew Sullivan wrote an article challenging the statistic, Bennett wrote in a letter to the New Republic (1998, February 23, page 4): “Given what I now know, I believe there are flaws with Paul Cameron’s study. One cannot extrapolate from his methodology and say that the average male homosexual life span is 43 years.”

Editor’s note: See also this post

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




Archives


%d bloggers like this: