Correcting Selection Bias in Standardized Test Comparisons
This paper tackles the critical issue of sample selection bias in cross-country comparisons using international assessments such as PISA (Program for International Student Assessment). While PISA is widely used to rank educational performance, its reliance on samples of students still in school at age 15 introduces survival bias, potentially distorting comparisons. To address this, I developed an econometric framework based on a quantile selection model. Under a stochastic dominance assumption, the selection-corrected quantile function is partially identified, with point identification achieved under parametric restrictions on the joint distribution of test scores and selection. Applying this method to PISA 2018 data, I demonstrate that correcting for selection bias leads to significant changes in country rankings based on mean performance. These findings underscore the importance of accounting for sample selection bias to ensure accurate and meaningful international educational comparisons.
Code (0)
등록된 구현이 없습니다.
Tasks
counterfactualSelection biasMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Non-Clicks Mean Irrelevant? Propensity Ratio Scoring As a Correction
Recent advances in unbiased learning to rank (LTR) count on Inverse Propensity Scoring (IPS) to eliminate bias in implicit feedback. Though theoretically sound in correcting the bias introduced by treating clicked docume…
Learning-To-RankCorrecting for Selection Bias in Learning-to-rank Systems
Click data collected by modern recommendation systems are an important source of observational data that can be utilized to train learning-to-rank (LTR) systems. However, these data suffer from a number of biases that ca…
counterfactualLearning-To-RankPositionRecommendation Systems+1Improving Evaluation of Debiasing in Image Classification
Image classifiers often rely overly on peripheral attributes that have a strong correlation with the target class (i.e., dataset bias) when making predictions. Due to the dataset bias, the model correctly classifies data…
Classificationimage-classificationImage ClassificationFastAMI -- a Monte Carlo Approach to the Adjustment for Chance in Clustering Comparison Metrics
Clustering is at the very core of machine learning, and its applications proliferate with the increasing availability of data. However, as datasets grow, comparing clusterings with an adjustment for chance becomes comput…
ClusteringAn experimental approach: The graph of graphs
One of the essential issues in decision problems and preference modeling is the number of comparisons and their pattern to ask from the decision maker. We focus on the optimal patterns of pairwise comparisons and the seq…