Do You Understand The Data?

 So, I wanted to make this blog a little different from the first one. I decided that I wanted to interview my roommate Kevin to get his insight on this subject today. A lot, if not half, of the interview that I had made was lost, so I'm going to paraphrase what he said.

I started by presenting him with the same option that Justice Kennedy did back in 2013 concerning the decision of redefining marriage. In the Supreme Court there needs to be a unanimous vote in order for a decision to be made. Four of the justices were four redefining marriage, while four others were against redefining marriage. I asked Kevin what he would do if he had the Swing Vote, and he was presented with the assertion from the American Psychology Association (APA) that, "not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect relative to children of heterosexual parents." He thought for a moment before he said that he would decide to redefine marriage, according to that statement.

After that, I told him that that's what Justice Kennedy also decided. However, statements by some Supreme Court Justices made it clear that they did not review and understand the data that was presented to them. So, that's where we go into the data. Of the 59 studies that were made, 55 of them were found to be of no statistical power, or they were not applicable. What it means to have no statistical power is that you may not be able to detect a statistically significant result even when it has practical significance. It is also called a type II error.

This is the part where I asked Kevin, "when you are conducting research, what are the two groups that you have?" He responded with," first, you have the group that you're conducting the research on, and then there's the group that you give the placebo to." I said, "that's right; there's the treatment group, and then there's the control group." Most of the data was inaccurate because when it said that it was not applicable, it means that there is no control group to compare against.

I presented to him another concern about the data as well. And that was that, eight of these studies had something in common that could be a possible error: in the Bozette, 1980 study, all the fathers in the sample were Caucasian. In the Flaks, 1995 study, 60 parents, all of whom were white, comprised the sample. In the Hoeffer, 1981 study, all 40 mothers were white. In the Huggins, 1989 study, all the children, mothers, and fathers in the sample were Caucasian. So was the case with the other four studies. 

The next issue I brought up to him was that, of the data that was reported to be statistically significant, two of those studies included single mothers.

With each of these issues that I brought up to Kevin, his response was always, "oh, that's not a good way to look at the data." What Kevin understood was the complexity of gathering data about the family. I found Kevin to be a really considerate critic. What I mean by that, is when we talked about the issue of the data containing all white people, I found him making arguments for and against being more privileged or having more opportunity. When we talked about the issue of single mothers being studied, he recognized that there needed to be two parents. But again, I found him arguing with himself whether those should be the biological parents or whether they could be homosexual parents.

Nearing the end of the interview, I asked him the question again, "if you were the Supreme Court Justice that had the swing vote, and you were told that "not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect relative to children of heterosexual parents", what would you decide now?" His response was, "I would say I don't understand 100% of the data. I would need to look into it more and if the data is incomplete, or if there's a direct bias, then we would need more information before making a decision. I guess from the information you've shown me today, it would be to keep the definition the same until further research can be conducted. I guess this was in 2005 when the APA brief was presented, so I think if I decided to keep it as it was, then I would spend a lot more resources, time, and energy in making sure that the data is correct to get more information on it."

I appreciated Kevin weighing in on this topic today, and I would love to hear everyone's thoughts and opinions. Truthfully, gathering family data is extremely hard. The best way, in my opinion, to gather family data is through lots of time. We would need to study the effects of many children of homosexual parents from the time they are children, to teenagers, to adults, to married adults, to married adults with children, to their children. 

Thanks for reading my blog today!