Evidence Of Hereditary IQ In Norway

We study a decade of achievement gaps for fifth-, eighth-, and 10th-grade students in Norway using administrative population data. Norway is a wealthy and egalitarian country with a homogeneous educational system, yet achievement gaps between students at the 90th and 10th percentiles of parental income and between students whose parents have at least a master and at most a high school degree are found to be large (0.55–0.93 and 0.70–0.99 SD), equivalent to about 2 to 2.5 years of schooling, and increasing by grade level. Achievement gaps by parental income, but not by parental education, increased over the time period, underscoring the different ways these two socioeconomic status components relate to achievement and the potential for policy to alter gaps.

From here.

This is not a surprising result. 

The evidence that IQ is a well defined real quantity that is heavily hereditary is well established. And, while IQ isn't necessary a great measure of everything that is important, it is absolutely an excellent indicator of a person's capacity for academic achievement in ordinary schoolwork.

This is, however, a particularly "clean" result, because Norway's wealth and egalitarianism remove many of the environmental confounds to definitively establishing the validity of hereditary IQ as real in other countries.

Also, the data set is huge:

We use data from national registries covering the entire Norwegian population of fifth, eighth, and 10th graders from 2007 to 2018 (cohort birth years 1991–2008) and their parents/guardians. The data allow us to examine (a) changes in gaps over time (i.e., between-cohorts comparisons for, e.g., fifth grade) and (b) within-cohorts achievement patterns across grades (i.e., how a cohort performs in fifth, eighth, and 10th grades). The total sample used in the analyses includes 1,103,081 children/adolescents. Data are obtained from Statistics Norway and include the education, income, and population registries, linked by anonymized personal identification numbers.

Honestly, the most surprising thing about the effect measured is how small it is. 

There are about 2.5 standard deviations of in IQ between the 10th percentile and the 90th percentile, so even with 50% regression to the mean, you'd expect a 1.25 standard deviation gap in achievement (although both income and education are imperfect indirect indicators of IQ).

The discussion section of the paper also adds useful analysis and comparisons:

There are three main findings in our study. 

First, even in a fairly egalitarian country with a homogeneous educational system, achievement gaps by both parental income and parental education are large, equivalent to about 2 to 2.5 years of schooling. 

Second, the gaps are increasing with student grade level, increasing by about 10 percentage points from fifth grade to eighth grade, with a similar increase to 10th grade. This increase is consistent with OECD (2018), which found large and growing gaps for Norway from age 10 to ages 25 to 29. 

Third, for income gaps, but not for education gaps, there has been an increase in gaps over time, equivalent to about 3 to 4 months of schooling. When we decompose the change for income, we observe that achievement remains high and stable for the 90th percentile and that achievement is decreasing over time for the 10th percentile. 

How do the gaps we observe for Norway compare to the United States and other countries? 

Such comparisons, as noted previously, are not straightforward but give us some idea of how the magnitude of our estimates compare to the magnitude in estimates in studies that are somewhat comparable. Our estimates are consistent with, but in general smaller than, those reported from the United States. For example, Reardon (2011b) reported income gaps that increase from about 0.75 SD in 1940 to about 1.25 SD in 2000, and Reardon and Portilla (2016) reported income achievement gaps for kindergarteners (ages 5–6) above 0.99, substantially above gaps for our lowest grade level. 

In an international comparison, Chmielewski and Reardon (2016) found the income achievement gap in Norway to be 0.75 of a standard deviation in the PIRLS 2001 data, compared with 1.25 of a standard deviation in the United States, using same-age data from the ECLS-K. 

Our estimates for achievement gaps by parental education are also slightly smaller than those reported from the United States. Using the Reardon (2011b) 90/10 estimation method, more appropriate for comparisons, we found gaps ranging from 0.86 to 1.15 SD (Appendix A, Figure A2, available on the journal website). Reardon (2011a) found achievement gaps by parental education in the United States to be stable over time largely above 1 SD, with some signs of a slight increase in the end of the period. 

While not directly comparable, the socioeconomic (SES) achievement gaps reported by Chmielewski (2019) for the United States are estimated to be between 1.1 and 1.2 for the period 1950 to 2000, whereas trends for other countries are largely above 1.0. Hanushek et al. (2020), measuring the difference between average performance above the 75th and below the 25th percentiles, found that gaps in the United States increased from .84 to .91 from 1961 to 2001. Similar estimates for our data, using parental education as our gap measure, yield gaps in the range of .51 and .87, again, largely lower than the U.S. estimates and in line with Broer et al. (2019) and OECD (2018), which showed that Norway has large gaps but that they are relatively smaller than those for the United States.

Despite smaller absolute achievement gaps in Norway compared to those found in the United States, it is notable that the gaps we observe in Norway are disturbingly large, given the sociopolitical context in which they have arisen. If income disparities (being about twice as large in the United States compared to Norway; OECD, 2021) relate to income achievement gaps, the relative difference in achievement gaps between the two countries is smaller than what one might expect. This is particularly evident if we compare the education, social security, and health-care system in Norway with those of the United States, expected to reduce inequality. From a developmental perspective, however, even in a more egalitarian society, there may be strong associations between student achievement and the resources that families have available to their children, whether measured by parent income or education.

The difference in the achievement gaps between Norway and the U.S. provide a rough means of estimating the extent to which environmental factors in the U.S. led to achievement gaps not arising from hereditary IQ. This excess achievement gap is on the order of half of a standard deviation, or about 7-8 IQ points.