Assume That Professor X Has Decided to Conduct a Quasi-Experiment

Assume that Professor X has decided to conduct a quasi-experiment where students are assigned to an experimental type of class instruction based on their performance on this final exam.(To qualify for participation,students must either score above 90 percent or below 60 percent) .Which problem would be most likely to interfere with the ultimate conclusion that student ability influenced the effectiveness of the instruction?

A) regression to the mean
B) measured differences between the groups of students
C) the scale attenuation problem
D) interactions between matched variables