The Importance of Samples and Populations

The final warning we wish to provide [in Chapter 5] is really a repetition of a major concern expressed earlier in this chapter. Simply stated, an empirical investigation that incorporates inferential statistics is worthless unless there is a detailed description of the population or the sample. No matter how carefully the researcher describes the measuring instruments and procedures of the study, and regardless of the levels of appropriateness and sophistication of the statistical techniques used to analyze the data, the results will be meaningless unless we are given a clear indication of the population from which the sample was drawn (in the case of probability samples) or the sample itself (in the case of nonprobability samples). Unfortunately, too many researchers get carried away with their ability to use complex inferential techniques when analyzing their data. We can almost guarantee that you will encounter technical write-ups in which the researchers emphasize their analytical skills to the near exclusion of a clear explanation of where their data came from or to whom the results apply. When you come across such studies, give the authors high marks for being able to flex their "data analysis muscles"--but low marks for neglecting the basic inferential nature of their investigations.

(From Chapter 5, p. 135)

Copyright © 2003 Schuyler W. Huck
All rights reserved.