School-level differences in grades are small and largely explained by attendance and course characteristics¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r3 · q3
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2021: baselines and conditions change

Subclaims¶

q3 i? School variance (0.10) is much smaller than student (0.94) or teacher (0.35) variance, and adding attendance leaves very little school-level variance (0.05 in Model 5c). → Elaine M. Allensworth and Stuart Luppescu 2016 q3 i? Comparing students with similar skills, backgrounds, attendance and courses, a 2.0-GPA student at schools one SD above or below the mean would earn GPAs about 0.2 points higher or lower. → Elaine M. Allensworth and Stuart Luppescu 2016 (2)

Evidence¶

Elaine M. Allensworth and Stuart Luppescu 2016¶

Elaine M. Allensworth and Stuart Luppescu. (2016). Why do students get good grades, or bad ones? The influence of the teacher, class, school, and student. University of Chicago Consortium on School Research working paper. https://consortium.uchicago.edu

q3 · i? · associational · r3

Cross-classified models on the CPS grade dataset show school variance of 0.10 in Model 2b, falling to 0.05 in Model 5c; the article states GPA differences across schools are "largely explained by differences in student attendance and differences in course characteristics."

"There is very little remaining school-level variance after adding student attendance to the models (0.05 in Model 5c). Differences in GPAs across schools are largely explained by differences in student attendance and differences in course characteristics"

Elaine M. Allensworth and Stuart Luppescu 2016 (2)¶

Elaine M. Allensworth and Stuart Luppescu. (2016). Why do students get good grades, or bad ones? The influence of the teacher, class, school, and student. University of Chicago Consortium on School Research working paper. https://consortium.uchicago.edu

q3 · i? · associational · r3

Figure 2 translates school variance components from Models 3a and 5c into GPA units: for comparable students with similar attendance and courses, school effects are modest, with GPAs 0.2 points higher or lower one SD from the mean; at the 98th/2nd percentile extremes the 2.0 student would land between 1.6 and 2.4.

"Students at schools that are one standard deviation above or below the mean have GPAs that are 0.2 points higher or lower than at a typical school, respectively."

Discussion¶

Related Claims¶

• Community-area hardship index explained 65% of between-community-area variation; most variation in participation was at the student level — related
• Teacher effects account for about one-fifth of within-student variation in course grades but average out of GPAs — related
• Differences in grading practices across teachers and schools are small relative to students' actual academic preparation, and GPAs average out those differences — a broader claim this one bears on
• Classroom-level ICCs for elementary test score gains range from 0.02 to 0.15 and school-level ICCs from 0.05 to 0.20, so classroom effects explain about 15 percent of gain-score variance — related