What works best for teachers and students in science education interventions? Statistics can provide some insight—but only if interpreted in context. For instance, the way a study is conducted may impact the outcome, independent of the actual effectiveness of the intervention.
To help researchers understand study results in context, BSCS Science Learning reviewed hundreds of studies in science education while taking into account the various factors contributing to the outcomes. This work resulted in published findings for student outcomes (AERA Open Journal) and teacher outcomes (The Journal of Research on Educational Effectiveness) and online tools for researchers to use when planning or evaluating studies of science education interventions.
The online tools, POWER Calculator for Student Outcomes, and POWER Calculator for Teacher Outcomes use data from the studies BSCS reviewed to estimate the likely effect size for a new study based on its characteristics, such as the nature of the study, the scientific discipline, characteristics of teachers or students, and other key variables.
When planning a study, researchers can use the POWER calculators to determine how many participants will be required to obtain a statistically significant result, giving researchers and funders increased confidence that they will obtain such a result without spending money and time unnecessarily on participants that are not needed. Once a study is completed, the tool enables users to interpret the size of their study’s effect in the context of similar studies.
In addition to the research findings and POWER calculators, BSCS has published the two data sets related to student and teacher outcomes. Researchers who want to do their own meta-analyses of the studies can explore the data from different angles, while efficiently using BSCS’s coding system.
Results from these meta-analyses are published in AERA Open and the Journal of Research on Educational Effectiveness.
This material is based upon work supported by the National Science Foundation under Grant Nos. 1118555 and 1544236. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.