Careless and Random Responding on Clinical and Research Measures in the Addictions: A Concerning Problem and Investigati
- PDF / 191,425 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 62 Downloads / 146 Views
Careless and Random Responding on Clinical and Research Measures in the Addictions: A Concerning Problem and Investigation of their Detection Joseph F. Meyer & Kyle A. Faust & David Faust & Aaron M. Baker & Nathan E. Cook
Published online: 7 December 2012 # Springer Science+Business Media New York 2012
Abstract Even when relatively infrequent, careless and random responding (C/RR) can have robust effects on individual and group data and thereby distort clinical evaluations and research outcomes. Given such potential adverse impacts and the broad use of self-report measures when appraising addictions and addictive behavior, the detection of C/RR can reduce error substantially. Based on earlier research using a video game questionnaire as an exemplar, we cross-validated promising items for detecting C/RR and developed an expanded set of items, in this case using an Internet questionnaire to examine efficacy and generalization. Research participants were instructed to complete the questionnaire in standard fashion (i.e., cooperatively) or to adopt either a careless or random response set. Careless and random responders often obtained elevated mean questionnaire scores. Most items for detecting careless or random responding demonstrated significant differences across groups, and combinations of items showed high levels of accuracy, particularly in detecting random responders. Results suggest that a relatively small number of items, which might only add a minute or two to questionnaire completion time, can detect the great majority of random responders and most careless responders. Guidelines are provided for the development and application of items. Keywords Video game questionnaires . Internet questionnaires . Behavioral addictions . Deviant response sets . Careless responding . Random responding Self-report measures, questionnaires, and surveys have proven invaluable in research and clinical work on the addictions, and their use is almost omnipresent. When we examined publications across 2 years of the present journal (2009, 2011) and a second well-known journal on the J. F. Meyer (*) : K. A. Faust : D. Faust : A. M. Baker : N. E. Cook Department of Psychology, University of Rhode Island, Kingston, RI 02881, USA e-mail: [email protected] K. A. Faust : D. Faust Department of Psychiatry and Human Behavior, Brown University, Providence, RI, USA
Int J Ment Health Addiction (2013) 11:292–306
293
addictions (Psychology of Addictive Behaviors, 2008 & 2010), we found that 171 (74 %) of the 230 research studies used at least one type of questionnaire, self-report, or survey measure. Despite this track record of use and usefulness, almost all such measures depend on respondents’ cooperation. Unfortunately, as is well known, some individuals in both clinical and research settings respond in an overly defensive manner, or they may answer items carelessly or even randomly, thereby compromising the quality of data. Conditions of anonymity, as are common in research, tend to increase rates of careless and random responding
Data Loading...
