Parents wishing to assist their children with school often turn to grades and test scores to inform their efforts. Which do they weigh more heavily? How do they respond when the two diverge? This question is not merely academic: following the COVID-19 pandemic, the nation has seen learning losses of 0.2 standard deviations masked by grade distributions that failed to reflect these declines.
In this paper, the authors study how parents trade off grades against standardized test scores in their investment decisions. They recruited 2,079 U.S. parents through the survey platform Prolific, screening for parents between ages 25 and 50 with at least one child born between 2000 and 2010, and stratified recruitment across four income brackets to ensure representation across the income spectrum.
Each parent was presented with a series of scenarios involving hypothetical fifth-grade children whose grades (A through F) and test scores (10th through 90th percentile) were independently varied — a key methodological contribution, as prior studies manipulated only a single signal at a time, leaving open how parents reconcile signals that point in different directions.
For each scenario, parents were asked to advise how much time per day and money per week a child’s family should invest in academic support. Both investment types (time and money) were standardized and combined into a single measure to maximize statistical power. Because each parent evaluated multiple scenarios, the authors identify responses from within-parent variation rather than cross-parent differences, controlling for stable individual characteristics. The final dataset comprises 23,321 investment decisions, and reveals the following:
- Parents respond to both grades and test scores, with lower performance on either signal increasing investment. On average, a one-unit decrease in grades (e.g., from B to C) increases investment by 0.14 standard deviations, while a one-unit decrease in test score percentile increases investment by 0.12 standard deviations.
- Parents put a higher weight on grades than tests, on average. This does not appear to be driven by parental income or education. Hispanic parents, however, show a particularly strong preference for grades over test scores.
- Parents exhibit asymmetric responses to conflicting signals. When test scores are high but grades are low, parents invest. When grades are high but test scores are low, however, parents do not invest: their responses are statistically indistinguishable from those for a child performing averagely on both measures. High grades crowd out the investment response that low test scores would otherwise trigger, meaning parents receiving inflated grades will fail to make remedial investments that their children’s actual achievement levels warrant.
Grade inflation’s costs extend beyond signal distortion. If high grades crowd out responses to low test scores, inflation insulates struggling students from remedial investment. This crowd-out mechanism implies that combating grade inflation may be more consequential for parental behavior than expanding test score access. At the same time, the finding that test scores do trigger investment when grades are concordantly low suggests that maintaining standardized testing preserves a valuable information channel — one whose influence depends critically on the quality of grade information parents already possess.