Blog

Van Ijzendoorn, M. H., Juffer, F., & Poelhuis, C. W. K. (2005). Adoption and cognitive development: a meta-analytic comparison of adopted and nonadopted children’s IQ and school performance. Psychological bulletin, 131(2), 301.

It turns out that someone already did a meta-analysis of adoption studies and cognitive ability. It does not solely include cross-country transracial, but it does include some. They report both country of origin and country of adoption, so it is fairly easy to find the studies that one wants to take a closer look at. It is fairly inclusive in what counts as cognitive development, e.g. school results and language tests count, as well as regular IQ tests. They report standardized differences (d), so results are easy to understand.

They do not present aggregated results by country of origin however, so one would have to do that oneself. I haven’t done it (yet?), but the method to do so is this:

1. Obtain the country IQs for all countries in the study. These are readily available from Lynn & Vanhanen (2012) or in the international megadataset.
2. Score all the outcomes by using the adoptive country’s IQ. E.g. if the US has a score of 97, and Koreans adopted to that country have get a d score of .16 in “School results” as they do in the first study listed, then this corresponds to a school IQ performance of 97 – 2.4 = 94.6. Note that this assumes that the comparison sample is unselected (not smarter than average). This is likely false because adoptive parents tend to be higher social class and presumably smarter, so they would send their (adoptive) children to above average schools. Also be careful about norm comparisons because they often use older norms and the Flynn effect thus results in higher IQ scores for the adoptees.
3. Copy relevant study characteristics from the table, e.g. comparison group, sample sizes, age of assessment and type of outcome (school, language, IQ, etc.).
4. Repeat step (2-3) for all studies.
5. BONUS: Look for additional studies. Do this by, a) contacting the authors of recent papers and the meta-analysis, b) search for more using Google Scholar/other academic search engine, c) look thru the studies that cite the included studies for more relevant studies.
6. BONUS: Get someone else to independently repeat steps (2-3) for the same studies. This checks interrater consistency.
7. Aggregate results (weighted mean of various kinds).
8. Correlate aggregated results with origin countries’ IQs to check for spatial transferability, a prediction of genetic models.
9. Do regression analyses to see of study characteristics predict outcomes.
10. Write it up and submit to Open Differential Psychology (good guy) or Intelligence (career-focused bad guy). Write up to Winnower or Human Varieties if lazy or too busy.

The main results table

More likely, you are too lazy to do the above, but you want to sneak peak at the results. Here’s the main table from the paper.