Hello all,

C, thanks for posting that link. The article on that site is extremely short. I read it after reading the actual study, and found it to be quite superficial, relying too much on the interpretation of the researchers.

Hartmann, P., Reuter, M., Nyborg, H. 2006."The relationship between date of birth and individual differences in personality and general intelligence: A large-scale study" Personality and Individual Differences, 40, 1349 - 1362.

First off, their participants. They used two sources.

One, from the Vietnam Experience Study (VES), was conducted with vets from 1965-71 and again from 1985-86. It is important to note that vets scoring below the 10th % on the cognitive aptitude test that was part of that study were excluded from the sample, as mandated by US Congress. hmmm. The researchers do not provide a number for that bottom 10%. The researchers say that for the 'present purpose' they included a total of 4321 vets from the database, which could be mean that they picked and chose certain people, but they do not go into detail. The study was composed of a general intelligence test, and a personality test, the Eysenkian Personality Questionnaire (ESQ), which has acceptable reliability (reliability means that it tests what we thing it tests).

The problem with the VES sample right off the bat is that it is only men. That kinda shit doesn't really fly these days, except if you are purposefully only studying one gender. Also, the researchers never specify which time they are taking the results from: the first eval from '65 to '71, or the one done in the 80's. And while not all Vietnam vets were in combat, this is a sample with an incredibly high rate of post traumatic stress disorder, which has many effects on personality and general functioning. We also don't know if these guys were injured, on meds, married, not married, etc. They do say that the sample brings in men from all levels of socioeconomic status, saying it is "fairly" representative, but we all know that more poor people probably went to Vietnam (on average) than rich people, further skewing the sample.

Second, from the National Longitudinal Study of Youth 1979 (NLSY1979). How it is a longitudinal study that was only done during one year, I don't know. Unfortunately, the researchers refuse to include details on this data and the vietnam vet sample, like inclusion, exclusion criteria (ANYTHING really), saying it has been documented elsewhere. But usually one at least goes over the basics of a sample for the reader, in order to make it easier to evaluate the soundness of the study. They included 11,448 young adult subjects (ages 15-24, mean age 19, almost evenly split male and female). The youth study only did an intelligence test, not a personality one, so really, the intelligence test numbers for this study are over 15,000, but the personality numbers are only from the VES study.

I don't have enough info on the NLSY study, but it is a large sample. I don't know how they recruited these young people to be in the study. All it was was mass 'intelligence testing,' and this testing was vastly different than the tests they did on the vets, making it harder to group the intelligence results from both samples together, as they could be representing different things (I mean, they tested the kids on 'automobile and shop information,' among others). Basically, imagine SATs for the young kids vs. card sorting tasks and the copying of pictures for the vets.

They did some standard stats. For the Vets, they only found one significant relationship, which was that people who were born during July - December had a slightly (though statistically significant) higher score on the intelligence scale than people born from January through June (see Christoph, I always knew we were smarter - hehe). Otherwise they either found no relationships, or just chose not to report data analyses.

For the youth study, they found that the relationship b/w season of birth and intelligence to be statistically significant (good p score = .025), and they found the reverse relationship for the January - June and July - December people (please disregard my jubilation from before) regarding intelligence. For males the month of birth did correlate with intelligence , but they do not tell us what kind of a relationship that was. Direct quote: "apparently, the overal relationship between season of birth and Jan - June vs. July - Dec is mediated by gender." And remember, they did not have data for personality, as that kind of test wasn't administered to the kids.

Overall, I found the study to be unsatisfying. In no way do I think that month of birth has anything to do with intelligence and personality, although some of those books that break people down by their birth date can be pretty uncanny. That aside, they had a ton of data to work with, but barely showed any of it. They had two completely different samples, one questionable due to life circumstances. The "general intelligence" tests so boldy heralded in the title were completely different. We don't know if data from the vets was taken directly after returning from a war zone, or 15-20 years later. Also, working with data retrospectively can be tricky because it's much easier to pick and choose suitable people, vs. having a prospective study with all of its trappings.

Here's my question: are people really going to stop reading horoscopes as a result of these studies? probably not.

H
- posted by H @ 4/28/2006 12:12:00 PM