Working Paper: NBER ID: w25460
Authors: Eric A. Hanushek
Abstract: The evaluation of educational programs has accelerated dramatically in the past quarter century. While such evaluations were once almost exclusively conducted in the U.S., they have broadened dramatically across many countries of the world. At the same time, the methodology has improved, strengthening considerably the internal validity of various studies. We must now consider what conclusions can be drawn from the growing wealth of international results. In particular, available cross-national studies on a variety of topics suggest using caution when generalizing results, because the results vary systematically with a number of institutional characteristics of the different countries that are not explicitly considered in within-country analyses.
Keywords: educational impact evaluation; cross-national studies; policy implications
JEL Codes: H4; I2; J00
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Accountability systems (H83) | Educational achievement (I24) |
| Initial levels of student performance (A21) | Effectiveness of testing and accountability systems (I21) |
| Local autonomy in decision-making (H77) | Student achievement (high-performing countries) (O57) |
| Local autonomy in decision-making (H77) | Student achievement (low-performing systems) (I24) |
| Cognitive skill levels (G53) | Economic outcomes (F69) |
| Institutional factors (D02) | Effects of educational policies (I28) |