ERIC Number: ED447194
Record Type: RIE
Publication Date: 2000-Apr-25
Examining Categories of Rival Hypotheses for Educational Research.
McMillan, James H.
Implicit in the work of S. Huck and H. Sandler (1979) is the idea that the concept "rival hypotheses" refers to some kind of alternative explanation. Rather than being threats to internal validity, rival hypotheses, in their view, are interpretations that differ from those of the researcher. This paper broadens the idea to include any interpretations or explanations that are important in understanding and using results, whether these are inconsistent with those of the researcher or simply noted as plausible. Huck and Sandler present 20 categories of rival hypotheses, going beyond the classic list of 7 threats to internal validity (Campbell and Stanley, 1963). This paper organizes categories of threats, dividing them into threats to external validity (or generalizability) and threats to internal validity. Three categories within internal credibility correspond to statistical conclusion, relationship conclusion, and causal conclusion explanations. These three categories are further subdivided, as shown in a table of potential threats to validity for experimental and non-experimental designs. (SLD)
Publication Type: Opinion Papers; Reports - Descriptive; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the American Educational Research Association (New Orleans, LA, April 24-28, 2000).