What Is an "Expert" in Analytics? Lies, Damned Lies, and Educational Statistics!

• In Event: 25.068 - Roundtable Session 8
  In Roundtable Session: 25.068-11 - Educational Statistical Research: Accuracy in Analytics and in the Context of Quasi-Experimental Designs

Abstract

The analytics necessary for research programs may be completed relatively quickly through hand calculations or the use of a statistical software. However, does correctly completing the computations or submitting the proper software command provide an accurate answer to a research or policy question? The purpose of this paper is to review the challenges educational statisticians must address to ensure internal, external and face validity for the analytical models they develop and report in research. Further, to provide recommendations on how educational statisticians and graduate degree programs in education may improve development and use of analytics to evaluate student performance.

Author

• Sean W. Mulvenon, University of Arkansas