Why it's a bad idea to track students based on cognitive ability tests given early in life

There's a lot of controversy1 around the idea of tracking in schools (i.e., testing students for ability or performance and then grouping them for differential curricula and/or instruction). Two of the many questions asked about this practice are whether the tests used to track students are reliable and whether students should be able to "switch" tracks over time. Well, a new meta-analysis by Breit et al. (2024) suggests that tracking decisions based on cognitive ability (i.e., "intelligence", comprehension knowledge, fluid reasoning, learning efficiency, processing speed, visual processing, and working memory capacity) tests (e.g., the Wechsler Intelligence Scale for Children, Woodcock–Johnson III Tests of Cognitive Abilities) are questionable, particularly for younger children. There's much more to this meta-analysis than what I'm presenting here, but regarding school tracking decisions, this paper suggests there's too much variability to use cognitive ability tests administered prior to age 6. Students will shift around too much from year-to-year from the early years to 6. Really, it's not until age 12 that there's sufficient stability in test scores (technically, the rank-orders implied by test scores) to make a tracking decision, and those decisions should be reassessed after four years (or earlier, if possible). And, sadly, like much of the research in psychology, these findings, and the studies used to derive them, were almost all conducted with WEIRD samples. So, whether these tests work at all for people from other cultures and contexts is unknown. Thus, all these findings (and limitations) make me skeptical of using cognitive ability tests for tracking, particularly those given prior to age 12.

1 Note, both linked articles are endorsing views against or alternatives to tracking.