U.S. PIRLS Reading Scores by Student Group and Gender 2021
PIRLS 2021 reports U.S. reading achievement for girls and boys from the delayed fall 2021 administration. The reported U.S. achievement results come from the paper-bridge sample of fifth-grade students.
2021 U.S. Gender Results
U.S. PIRLS Reading Score by Gender
Average PIRLS reading scale score for girls and boys in the 2021 U.S. reported sample.
The girls–boys difference was 7 scale points with a standard error of 4.4 points; the official comparison did not meet the p < .05 threshold.
Source: IEA, PIRLS 2021 International Results in Reading.
Full U.S. Gender Data
| Girls | 50% | 1.3 | 551 | 7.2 |
| Boys | 50% | 1.3 | 544 | 7.1 |
The 2021 Gender Gap Was Within Sampling Uncertainty
- Girls had a numerically higher average reading score than boys, but the 7-point difference did not meet IEA’s p < .05 criterion. The result therefore supports describing a numeric difference rather than a confirmed population-level gender difference.
- The standard errors for the two gender estimates were about seven scale points, close to the size of the observed gap. The uncertainty around the estimates is especially relevant because the reported U.S. achievement results came from the smaller paper-bridge sample rather than the full digital assessment sample.
- The United States did not satisfy IEA sampling participation guidelines in 2021. Weighted overall participation after replacement schools were included was 64 percent, so the participation notation attached to the U.S. results also applies when interpreting subgroup estimates.
Data Sources, Measurement and Comparability
Official U.S. source
NCES PIRLS 2021 U.S. Results reports the U.S. 2021 reading result and documents the delayed U.S. administration.
Gender estimates
IEA PIRLS 2021 International Results in Reading, Exhibit 1.5, reports the U.S. percentages, average scale scores, standard errors, and girls–boys difference used on this page.
Population
The United States delayed testing from spring to fall 2021 because of pandemic-related disruption. The assessed cohort was therefore at the beginning of fifth grade rather than at the end of fourth grade.
Reported assessment mode
The United States administered digitalPIRLS but reported achievement from its paper-bridge study because of psychometric concerns about linking the U.S. paper and digital results. NCES documents this choice in the PIRLS 2021 methodology and technical notes.
Score scale
PIRLS achievement is reported on a 0–1,000 scale. The scale centerpoint was set at 500 and the standard deviation at 100 when the scale was established in 2001. Scores are population estimates rather than individual student scores.
Gender comparison
IEA reported a 7-point girls–boys difference with a standard error of 4.4 points. The comparison did not meet the p < .05 threshold. Standard errors for the girls’ and boys’ score estimates were 7.2 and 7.1 points, respectively.
Sample and participation
The U.S. paper-bridge sample assessed 1,657 students. Weighted school participation was 54 percent before replacement and 67 percent after replacement; class participation was 100 percent, student participation was 95 percent, and overall participation after replacement was 64 percent. IEA marked the United States as not satisfying its sampling participation guidelines.
Comparison with earlier cycles
The 2021 U.S. population had additional schooling because assessment occurred about six months later than originally planned. NCES and IEA advise caution when comparing the delayed fifth-grade cohort with fourth-grade results from earlier PIRLS cycles.
U.S.-specific race/ethnicity and school poverty estimates: NCES explains that race/ethnicity and free or reduced-price lunch variables are contained in U.S. national data files rather than the standard international file. Its current PIRLS data-files page lists national public-use files with these U.S.-specific variables through 2016, not 2021. For that reason, 2021 race/ethnicity and FRPL reading scores are not presented as if they were published U.S. subgroup results.