Working Paper: NBER ID: w22558
Authors: Anna Aizer; Janet Currie; Peter Simon; Patrick Vivier
Abstract: We construct a unique individual-level longitudinal dataset linking preschool blood lead levels with third grade test scores for eight birth cohorts of Rhode Island children born between 1997 and 2005. Using these data, we show that reductions of lead from even historically low levels have significant positive effects on children's reading test scores in third grade. Our preferred estimates use the introduction of a lead remediation program as an instrument in order to control for the possibility of confounding and for considerable error in measured lead exposures. The estimates suggest that a one unit decrease in average blood lead levels reduces the probability of being substantially below proficient in reading by 3.1 percentage points (on a baseline of 12 percent). Moreover, as we show, poor and minority children are more likely to be exposed to lead, suggesting that lead poisoning may be one of the causes of continuing gaps in test scores between disadvantaged and other children.
Keywords: blood lead; children; test scores; education; public health
JEL Codes: I24; J15; Q53; Q58
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Blood Lead Levels (Y10) | Educational Outcomes (I21) |
| Decrease in Blood Lead Levels (J89) | Improvement in Reading Test Scores (I21) |
| Increase in Blood Lead Levels (I19) | Decrease in Reading Scores (I21) |
| Blood Lead Levels (Y10) | Math Scores (Y10) |
| Lead Exposure (L72) | Cognitive Outcomes (D91) |
| Lead Remediation Programs (R28) | Mitigation of Effects of Lead Exposure (I19) |