Partial credit accounts for only a small proportion of IF-AT exam scores, though weaker students depend on it more¶

Claim · All claims
Evidence · 1 study · design r2 · q2 · i1 small
Currency · newest study 2013 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? On the final exam (mean score 67%), removing partial credit lowered the average by only 5 points to 62%, yet as many as 10 of 49 students would have failed without partial credit, each gaining less than 8 percentage points from it. → Slepkov 2013

Evidence¶

Slepkov 2013¶

Slepkov, A. D. (2013). Integrated Testlets and the Immediate Feedback Assessment Technique. https://arxiv.org/abs/1308.4365 doi:10.48550/arxiv.1308.4365

q2 · i1 · design · r2

Score decomposition of the IF-AT final exam: the mean was 67% but "without partial credit—i.e. a [1,0,0,0] marking scheme—the average was only 5 points lower at 62%". The author notes an inverse correlation between partial credit granted and exam score, mostly due to opportunity.

"as many as 10 of 49 students would have failed the final exam (earned less than 50%) without partial credit, but partial credit accounted for less than 8 percentage points for each of these students."

Discussion¶

Related Claims¶

• IF-AT exam scores correlate with constructed-response quiz scores (r = 0.67), supporting validity — related
• Partial credit under the IF-AT is granted in a discriminating manner: top final-exam scorers earned 65% ± 23% of available partial credit versus 39% ± 10% for bottom scorers — related
• Polychotomously scored IF-AT physics exam items show excellent discrimination, with a mean item-total correlation of r′ = 0.41 ± 0.13 across 45 items — related
• Intra-testlet items were slightly more difficult than stand-alone questions (p′ = 0.69 ± 0.15 vs 0.75 ± 0.15) — related