| Abstract: | This article replies to J. R. Hollenbeck, D. S. DeRue, and M. Mannor's comment (see record 2006-00819-001) critiquing R. S. Peterson, D. B. Smith, P. V. Martorana, and P. D. Owens's use (see record 2003-08045-002) of a large number of statistical tests in research with a small sample. Although Hollenbeck et al's point of view is valid, it paints a one-sided picture of the trade-offs inherent in empirical research when data are scarce and the questions important. This reply specifically discusses the dilemmas Peterson et al faced in conducting empirical research in a nascent area and suggests that theory development in such a situation can be well served by studies that use alternative or new methods with small samples. Theory development scholarship using small-sample research methods (e.g., case studies and Q sorting from archival sources) can be useful for stimulating ideas, theory, and research programs that can be tested with large-sample quantitative research. (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
