Simulating the statewide scaleup of a promising teacher education initiative to preempt its unintended consequences for racial inequity

Abstract

Most evaluation research fails to adequately anticipate unintended consequences, thereby implicitly permitting the possibility of negative repercussions so long as they fall out of the purview of the policy under study. Simulating how otherwise promising initiatives might scale into educational policy, however, offers both researchers and policymakers a way to not only investigate but also preemptively address any such possible, pernicious side effects. We provide an illustrative example of this kind of forward-thinking evaluation research by generating hypothetical scenarios of the statewide implementation of an algorithmic teacher education initiative shown in prior research to have positive intended effects on improving the quality of preservice teachers’ clinical placements. By comparing these plausible implementation scenarios against the historical record of clinical placements that programs actually made, we are able to not only uncover the unintended but anticipated inequities that this initiative would likely introduce if adopted at scale but also proactively make algorithmic adjustments that prevent their occurrence without diminishing any intended positive impacts, all before causing real-world harm.