Reading focus
Syllabus: What Is Intelligence?
Compare cohort trends across families with those between brothers, accounting for birth order and selection. Keep explanations ruled against separate from mechanisms actually identified.
Why you’re reading it. End by distinguishing causes of changing population averages from causes of differences among people.
A question to consider. Younger cohorts of the programme score lower even within families. Why would this weaken a family-composition explanation without proving that education caused the decline or that individual differences have no genetic component?
Why this reading
Average cognitive test performance has risen across cohorts in many settings and later declined in some. This study uses Norwegian conscription records and comparisons between brothers to investigate whether changing family composition can explain the pattern. It illustrates how a causal claim can become more specific without identifying the responsible environmental mechanism.
Before you read
A cohort groups people by a shared period, such as birth year. The Flynn effect concerns changes in average scores across cohorts, not simply an individual's development with age. Family fixed effects compare brothers while accounting for characteristics shared within their family. They do not remove every influence that changes between births or testing occasions.
Read for
- Follow the restrictions of the Norwegian male conscription sample and the distinction between across-family and within-family comparisons.
- Examine how birth order, missing scores, and changing selection into testing bear on the estimated trends.
- Identify which explanations based on changing parental composition are challenged and which environmental mechanisms remain unmeasured.
Consider afterwards
- A rise and subsequent fall in scores appears both across families and between brothers born in different years. Which explanations does this weaken, and why does it not tell us whether schooling, nutrition, media use, or another changing exposure produced the pattern?
Editorial context
The causal title should be interpreted within the study's design and population: Norwegian men from specified birth cohorts with two native-born parents. Within-family trends argue against explanations based solely on changing composition between families. They do not identify a particular exposure or show that genetic differences play no role in differences between individuals. Causes of cohort-mean change and causes of individual variation are distinct questions; findings here should not be generalized automatically to all countries or groups.