The particular need for replication in the quantitative study of SLA: A case study of the mnemonic effect of assonance in collocations

What is it about?

Experimental and quasi-experimental studies in the field of instructed second language acquisition are almost always small in scale in the sense that results are derived from just a couple dozen learners and, usually, an even smaller number of targeted linguistic items. P values and estimates of effect size derived from such studies are in general far less stable across across replications than most researchers apparently recognize. We explain why serviceable estimates of effect size require multiple replications that are as exact as practically possible. We argue too for increased use of small-scale interim meta-analysis.

Featured Image