• No, children of same-sex parents do not have lower graduation rates.

    A new Canadian study reportedly found that children of same-sex parents have a lower graduation rate than opposite sex married families. It is currently doing the rounds on your usual anti-gay sites as proof that homosexuals are unable to raise children, are unnatural, the spawn of Satan, harbingers of death etc.

    The study states that “children living with gay and lesbian families in 2006 were about 65% as likely to graduate compared to children living in opposite sex marriage families”. But a quick look at the data provided shows that this simply isn’t the case. The author of the study, Douglas W. Allen, makes a rather strange and glaringly obvious flaw: he includes children who are still in high school awaiting graduation. Allen analyses the 2006 Canadian census and focuses on 17-22 year olds. However, many 17-18 year olds would still be in High School so Allen will be misreporting them as not graduating instead of awaiting graduation. Such a flaw isn’t too problematic unless the average age of one group of parents is lower than the others, because if this is the case then the group with the lower average age will have a lower graduation rate because more of their children will still be awaiting graduation. So let’s see what affect this flaw has upon the results:

    POPULATION AGE AVERAGE                                 GRADUATION RATES
    Opposite sex married:  19.26                                 Opposite sex married:  .72
    Gay parents:  18.96                                                 Gay parents: .60
    Opposite sex common law:  18.91                          Opposite sex common law: .59
    Lesbian parents:  18.79                                           Lesbian parents: .52

    Yep, that’s right, it aligns perfectly. The lower the average age of the sample groups then the lower the graduation rate because many are still in school. So what Allen has cleverly discovered, I hope you are sitting down for this, is that the more children are still attending High School, the less likely they are to have graduated. Truly ground breaking stuff.

    Category: Equal Marriage

    Article by: Humanisticus

    One Pingback/Trackback

    • Peter

      In what journal was this study published? There are lots of right-wing “think tanks” here so I’d be interested where it was published and by whom.

      • It was in Review of Econmics and the Household.

        • Peter


      • Dan

        The author has a history of anti-gay activitism and misrepresenting data, intended to stigmaize. He is an economist, not a social scientist. Other flaws of the study are explained on other sources.

    • Do you know the effect this has on the graduation rate among girls raised by gay male parents? That had the most dramatic number—a mere 11% graduation rate as reported by the study.

    • Kaye

      I appreciate your great description of this methodological problem with the article. I wholeheartedly agree with your critique. I wish you had also commented on two additional problems:

      (1) Dr. Allen reports and draws conclusions from results that are not statistically significant. Specifically, he breaks the sample down by gender and reports that daughters of gay couples are worse off than the other groups. He then states “However, none of these estimates are statistically significant.” In other words, the differences are not meaningful and probably arose by chance.

      (2) In interpreting his results he capitalizes on the general public’s lack of understanding of Odds Ratios. Many people seem to be mistaking the Odds ratio for the graduation rate. The Odds Ratio he reports is 0.65 for gay/lesbian parents relative to heterosexual married parents. This DOES NOT mean that only 65% of the children of gay/lesbian parents are high school graduates. Rather, if you assume an Odds Ratio of 0.65 and assume that 89% of heterosexual couples’ kids graduate (the reported high school graduation rate for Canada in 2006 was 89%), then the graduation rate for gay/lesbian couples’ children would be 84%.

      Here is a nice page on odds and probabilities if anyone wants to check my math: http://pages.uoregon.edu/aarong/teaching/G4075_Outline/node15.html

      Of course, #2 is not relevant, given the methodological problem you mention.

    • literate3

      “Yep, that’s right, it aligns perfectly”
      No it doesn’t.
      First age difference 0.3 grad diff 0.12 =40%
      Second age difference .05, grad diff 0.01=20%
      Third age difference .12 grad diff 0.07=58%
      Married:unmarried opposite sex age diff 0.35 grad diff .13=37%
      Gay:Lesbian age diff .17 grad diff .08 =47%
      So unmarried parents of different sex do better on an age-adjusted scale than same-sex parents.
      If you want to criticise someone for misusing numbers, please get your own right.

    • disqus_pmal

      how about crime involvement, how many kids are involved in crime raised by homo/hetero/divorced/single mom/single dad/orphan, etc. Especially orphan versus homo family?

    • Pingback: Why the bad science of the no campaign shouldn’t sway Ireland’s voters | News Leader()