Demographic variable mutability as an organizing distinction for predictor selection¶

Theory · All theories
Evidence · 5 claims (4 for, 1 mixed) · 1 study (1 theoretical), q1 · 0 of 1 report an effect size · 5 claims rest on one study

Description¶

The article organizes the debate around what a demographic variable is and whether schools can change it. Demographic variables "cannot be manipulated (Lee & Scheule, 2010), either due to ethical reasons or feasibility" — a school can improve attendance but generally cannot provide housing. The framework distinguishes immutable variables (race, home language/ESL status), mutable ones schools explicitly work to change (ELL designation via English instruction, cohort-normed age via grade policies), and complex cases needing variable-by-variable analysis. This mutability lens underpins the actionability argument: immutable predictors cannot reflect intervention progress.

Design Implications¶

Context¶

Requirements¶

• Variable-by-variable analysis of whether a demographic label can legitimately change, since incentives to relabel can be salutary or less so

Constraints¶

• In communities where red-shirting is common, cohort-normed age's meaning may be complicated; a learner who legitimately changes label may still share attributes with the original group

Target Learners¶

• Students in K-12 and higher education at-risk prediction contexts

Target Learning Objectives¶

• Selecting actionable, ethically appropriate predictors for student outcome models

Claims¶

• Demographic Predictors Reduce Actionability [+W]
• The argument that excluding demographic predictors amounts to color-blind racism is contested: inclusion can also produce racist decision-making [+W]
• Heterogeneity within demographic categories means group variables as predictors can disadvantage atypical group members and underrepresented groups [+W]
• Using demographic variables as predictors risks reinforcing biases embedded in training labels, including self-fulfilling prophecies [+W]
• Evidence is mixed on whether including demographic variables as predictors improves model performance in educational prediction [~W]

Related Theories¶

Examples¶

-

Key Sources¶

• Baker, R. S., Esbenshade, L., Vitale, J., & Karumbaiah, S. (2023). Using Demographic Data as Predictor Variables: a Questionable Choice. Journal of Educational Data Mining, 15(2). https://jedm.educationaldatamining.org/index.php/JEDM/article/view/619. doi:10.35542/osf.io/y4wvj