TY - JOUR
AU - Bobko, Philip
AB - Moderated and Appropriate Regression Research A Demonstration Inappropriate Strategy: of Information Loss Due to Scale Coarseness J. Russell Craig University Rutgers K. Pinto Jeffrey of Maine University Bobko Philip Rutgers University Paunonen and Jackson demonstrated research. Since that the in the (1988) time, infrequency that moderated a test provides stepwise regression literature of applied psychological findings sup- of interaction effects that the nominal protects of moderator effects has fueled concerns portive I error rate. the However, stepwise Type procedure over the of the power procedure (Zedeck, 1971). has also been characterized as to detect failing More Bobko and Venkatraman recently, (1986) interaction effects in studies. This issue empirical has led to the method’s have concerns questions regarding (1989) reported specific regarding statistical in power (Bobko, 1986; Zedeck, 1971) the or &dquo;fit&dquo; between an interac- correspondence research. It is demonstrated because applied that, tive framework and statistical conceptual pro- of a research used in strategy frequently empirical cedures used to evaluate the framework. the of II in error investigations, probability Type The in interactions has a true interaction effect is unknown. difficulty detecting detecting the number of scale used in often been attributed to Specifically, steps multicollinearity among the variable 
TI - Appropriate Moderated Regression and Inappropriate Research Strategy: A Demonstration of Information Loss Due to Scale Coarseness
JF - Applied Psychological Measurement
DO - 10.1177/014662169101500305
DA - 1991-09-01
UR - https://www.deepdyve.com/lp/sage/appropriate-moderated-regression-and-inappropriate-research-strategy-a-1Q0i7w5pIB
SP - 257
EP - 266
VL - 15
IS - 3
DP - DeepDyve
ER -