Generalizability from second language research samples |
James Dean Brown University of Hawai'i at Manoa |
[ p. 21 ]
[ p. 22 ]
However, if the opposite is true, that is the sample is relatively homogeneous and large with equal sized groups, and the strata are not important to the analysis, a random sample may prove better because it removes the need to define the strata in the population and sample proportionately from each stratum.[ p. 23 ]
Given the fact that many of the sets of students used in second language studies are samples of convenience and the fact that samples of convenience themselves are typically the populations beyond which it is irresponsible to generalize, perhaps we should be less concerned with generalizability and more concerned with the transferability of the results of a study (as suggested by the discussion in Lazaraton, 1995, p. 465). Transferability is the demonstration of the "applicability of the results of a study in one setting to another context, or other contexts" (Brown, 2001, p. 226). In other words, given that we very often cannot generalize our results beyond the population of convenience, perhaps we should abandon the notion of generalizability and, instead, describe the groups of students in these populations of convenience thickly (i.e., in considerable detail) so other researchers and the readers of our studies can decide for themselves if the results are transferable to the settings that they are dealing with.Where to Submit Questions: |
Please submit questions for this column to the following address: |
JD Brown Department of Second Language Studies University of Hawai'i at Manoa 1890 East-West Road Honolulu, HI 96822 USA |
[ p. 24 ]