Identification systems typically involve conjunctive (“And”), disjunctive (“Or”), and compensatory (“Mean”) rules for combining multiple measures. As correlations among assessments decrease, conjunctive and compensatory systems identify fewer students (unless the cut-off for the mean score is adjusted for shrinkage), while disjunctive rules identify more students for programming. However, both researchers and practitioners in gifted education often assume that correlations among multiple identification measures are the same for students from different backgrounds. If correlations among measures are lower for one group than another, the group with lower correlations would be disadvantaged by conjunctive (AND) and compensatory (MEAN) rules (unless the compensatory rule computes shrinkage factors separately for each subgroup). Conversely, they would be advantaged by disjunctive (OR) rules. The key takeaways from this work are:

• Different combination rules can be implemented to identify similar overall percentages of students.
• Correlations among identification measures do appear to vary somewhat across demographic groups, and this has implications for how combination rules can be expected to perform.
• No combination rule can create parity when mean score differences across subgroups are substantial.