How will you vote in the Marriage Equality referendum? Mod Note Post 1

1.1.1. Population sample

The first problem is the sample of individuals chosen to undertake the questionnaire. The study interviewed individuals only between the ages of 18-39 and questions them on their childhood. This means that the author is only looking at a snapshot of time between 1972 and 2011. There are two problems with this. Firstly, it means that the study has no depth. There is no longitudinality which is something that usually makes for the most robust conclusions. The author even makes an acknowledgement of this fact:

"There are several things the NFSS is not. The NFSS is not a longitudinal study, and therefore cannot attempt to broach questions of causation."Secondly, this snapshot covers an area of massive flux with regard to same-sex parenting. It moves from a time when homosexual parenting was entirely unusual and mostly prohibited to a time when it is becoming an accepted family form and yet seeks to make broad conclusions about the snapshot as a whole.

Also with regard to sample selection, Regnerus opens his study with an attack on previous studies such as the NLLFS (National Longitudinal Lesbian Family Study) for their poor sample selection. However, Regnerus used Knowledge Networks to gather his sample data despite having prior knowledge that a) KN had a limited ability to connect him with people who fell under the criteria of the study and b) there were other companies available who could have supplied more rigorous data.


1.1.2. Classification problems

There are several problems in this section in the design of the experiment.

The first problem is the classification of family form. The survey uses the following categories for analysis:

1.IBF: Lived in intact biological family (with mother and father) from 0 to 18, and parents are still married at present (N = 919).
2.LM: R reported R’s mother had a same-sex romantic (lesbian) relationship with a woman, regardless of any other household transitions (N = 163).
3.GF: R reported R’s father had a same-sex romantic (gay) relationship with a man, regardless of any other household transitions (N= 73).
4.Adopted: R was adopted by one or two strangers at birth or before age 2 (N = 101).
5.Divorced later or had joint custody: R reported living with biological mother and father from birth to age 18, but parents are not married at present (N = 116).
6.Stepfamily: Biological parents were either never married or else divorced, and R’s primary custodial parent was married to someone else before R turned 18 (N = 394).
7.Single parent: Biological parents were either never married or else divorced, and R’s primary custodial parent did not marry (or remarry) before R turned 18 (N = 816).
8.All others: Includes all other family structure/event combinations, such as respondents with a deceased parent (N = 406).

The problem here is that while broken homes are included in both types of same-sex families, they are definitionally excluded from heterosexual couples so that the comparisons made later on are fundamentally flawed.

The second problem with the classification above is that any instance of infidelity with a same-sex partner moves the respondent into either LM or GF regardless of the nature of the infidelity (i.e. one night stand vs. long term relationship).

The final problem with the classification above is that a number of respondents were found to fit into more than one category. To resolve this overlap Regnerus makes arbitrary decisions in order to achieve the results he wants:

"That is, a small minority of respondents might fit more than one group. I have, however, forced their mutual exclusivity here for analytic purposes. For example, a respondent whose mother had a same-sex relationship might also qualify in Group 5 or Group 7, but in this case my analytical interest is in maximizing the sample size of Groups 2 and 3 so the respondent would be placed in Group 2 (LMs). Since Group 3 (GFs) is the smallest and most difficult to locate randomly in the population, its composition trumped that of others, even LMs. (There were 12 cases of respondents who reported both a mother and a father having a same-sex relationship; all are analyzed here as GFs, after ancillary analyses revealed comparable exposure to both their mother and father)."That, btw, is as good an example of bad science as you're ever likely to see.

In classifying the data this way, Regnerus creates a homogeneous IBF category and compares it against LM and GF categories which are made up of a number of different relationship types both stable and unstable. That dog won't hunt Monsignor.


1.1.3. Analytical problems

The first problem in this category is Regnerus' idea of what constitutes a bad outcome for children. In particular Regnerus classifies the "current" receipt of public assistance (i.e. social welfare) as a bad outcome.
The problem with this analysis is that America is currently in the depths of one of the deepest economic crises it has faced, yet Regnerus makes no attempt to analyses the other socio-economic factors that play into this in analysing the data.

The second and bigger problem is that the classification errors made by Regnerus in his experiment design leads him to making the wrong comparison. Children in a family where one or both parents were in a same-sex relationship were found to have similar outcomes to those in divorced, cohabiting and step families and markedly different from intact heterosexual families. This is because he designed the study this way. What he actually ended up doing was comparing stable families against non-stable ones and the conclusions don't offer any new insight into the field which is what research is supposed to do.

To put it simply he analysed bad data with a badly designed experiment to get bad results and ended up with bad conclusions.