An equivalence test showed FLEX and conventional students' final exam results were statistically equivalent despite halved classroom time¶

Claim · All claims
Evidence · 1 study · causal r2 · q3
Currency · newest study 2018 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? The two-sample equivalence test rejected H0: the exam results of the FLEX and CONV groups can be regarded as statistically equivalent (test statistic T = 1.64 below the noncentral Fisher quantile of 3.83). → Claude Müller 2018

Evidence¶

Claude Müller 2018¶

Claude Müller, Michael Stahl, Mark Alder, Maximilian Müller. (2018). Learning Effectiveness and Students' Perceptions in a Flexible Learning Course. European Journal of Open, Distance and e-Learning, 21(2). https://eric.ed.gov/?id=EJ1208426

q3 · i? · causal · r2

A two-sample test for equivalence (following Meyners, 2012; Wellek, 2010) was applied to the same 2014 IBA final exam data because the project's goal was equivalence, not superiority. The test statistic T = 1.64 fell below the quantile 3.83, supporting statistical equivalence of the groups' results.

"The additional equivalence test shows a value of the test statistics T (1.64) lower than the corresponding quantile of the noncentral Fisher distribution (3.83). Thus, the H 0 hypothesis can be rejected, and the exam results for the FLEX and the CONV groups can be regarded as statistically equivalent."

Discussion¶

Related Claims¶

• The deviation of BIT students' exam results from BA means did not change significantly between conventional (2012/13) and FLEX (2014) formats — related
• FLEX students with 57% less face-to-face time showed no significant difference in final exam results compared with conventionally taught students — possibly the same claim (merge candidate)