BRUNO D. ZUMBO
OPENING REMARKS TO THE SPECIAL ISSUE ON VALIDITY THEORY AND THE METHODS USED IN VALIDATION: PERSPECTIVES FROM THE SOCIAL AND BEHAVIORAL SCIENCES
To my knowledge there has been no discussion of contemporary views of validity as they apply to quality-of-life and social indicators research in recent years. A special issue is an opportunity for experts in a field to engage in a dialogue on a topic of special interest or critical need. The concept, method, and process of validation is central to quality-of-life and social indicators research, for without validation, any inferences made from a measure are meaningless. Throughout this volume, the terms measure, observation, score, test, index, indicator, and scale will be used interchangeably and in their broadest senses to mean any coding or summarization of observed phenomenon. Furthermore, lest we fall into traditional camps, validity applies equally to both so-called subjective and objective indicators. The issue at hand is that one needs to make an inference from a score about the state or status of an observational unit whether it is something that is at first glance objective (e.g., annual earnings), or subjective (e.g., self-reported well-being). My point is that whether one is measuring annual earnings or subjective well-being one has just one (or more) of an extensible set of indicators of a construct of interest. Also, the score on the indicator is not equated with the construct it attempts to tap, and often nor is it considered to define the construct as in strict conventional operationism. Some advocates of the exclusive use of objective indicators appear to forget (or This special issue arose out of discussions with Alex Michalos about current developments in validity theory. I would like to thank Alex for his support and encouragement throughout this project. I also extend my thanks to Ed Diener who was very supportive and helpful at the crucial early stages of this project. Social Indicators Research 45: 1–3, 1998.
2
BRUNO D. ZUMBO
downplay) that they too rely on constructs and inferences in their social indicators. What is at issue is whether the constructs are based on an elaborated theory and hence have the status of being explicitly theoretical (i.e., objective indicators), or whether the constructs are merely embedded in a network of expected or hypothesized empirical relationships (i.e., subjective indicators). Therefore, for subjective indicators the validation results and the substantive theory development are inextricably intertwined, whereas this is less of a case for the former. The papers selected for this special issue come from scholars in a wide range of fields in the social and behavioral sciences (e.g., sociology, economics, psychology, statistics, psychometrics, educational measurement, health research, management and business research) and reflect the international and interdisciplinary flavor not only of quality-of-life research but also of this journal. In an attempt to organize the diverse themes and yet reflect current thinking in validity I have separated theory and methodology. This is not to imply that theory and method are disjoint but rather it is meant to highlight one of the central concepts in contemporary validity theory – validity is not simply a technique or method. This special issue is divided into three sections. The first section contains and portrays a conceptual framework of validity that is undergoing consolidation, debate, and re-conceptualization. The authors raise new questions and re-awaken long-standing debates that lie at the heart of empirical science and speak to our collective desire to formalize and better articulate the concepts and measures we employ. As we are reminded in the philosophies of science, linking concepts to observations (often in a nomological network) is a fundamental strategy to clarify the meaning of a measure. The second section is loosely organized around the process of inquiry, i.e., methodology, in validation research. These papers reflect some of the central methodological issues at hand: translation involving multi-cultural and multi-language studies, various measures of association/correlation and their effects on validity and reliability, powerful and flexible estimation methods in regression, classical test theory, and variable ordering in statistical models. Because the problems they treat are ubiquitous, these papers are an essential read for any researcher.
OPENING REMARKS TO SPECIAL ISSUE
3
The third section of this special issue is devoted to ongoing studies of measures, i.e., the work of validation. These papers reflect the diversity in measures used in quality-of-life research. One sees more traditional self-report measures, scale formats, and interviews, as well as relatively new approaches from goal-oriented activity psychology and personality research involving personal goals methodology and personal projects methodology. Finally, it is with gratitude that I acknowledge the hard work of an exceptional group of reviewers. Crocker L., University of Florida Duxbury L., Carleton University Edwards J., University of Michigan Emmons R., University of California, Davis Fekken C., Queen’s University Fischer D., University of Saskatchewan Frasier C., Monash University, Australia Goffin R., University of Western Ontario Gorsuch R., Fuller Theological Seminary, California Hancock G., University of Maryland Holden R., Queen’s University Hubley A., University of Northern British Columbia (multiple reviews) Kaplan D., University of Delaware Kornbrot D., University of Hertfordshire, UK Land K., Duke University MacIntyre P., University College of Cape Breton Maraun M., Simon Fraser University McDowell I., University of Ottawa
McQuitty S., University of Northern British Columbia Michalos A., University of Northern British Columbia Moss P., University of Michigan Neizert M., Laurentian University Pedhazur Schmelkin L., Hofstra University Phipps S., Dalhousie University Popham J., IOX Assessment Associates, Los Angeles, CA Pratschke J., Universita di Salerno, Italy Pychyl T. Carleton University Raykov T., University of Melbourne, Parkville, VIC, Australia Reckase M., ACT, Iowa City, Iowa Schutz R., University of British Columbia Shepard L., University of Colorado at Boulder Steyer R., Friedrich-Schiller-Universität Jena, Institut für Psychologie Tousignant M., Université du Québec á Montréal Wheeler R., St. Louis, MO Williams R., University of Miami Zedeck S., University of California, Berkeley Zimmerman D., Surrey, British Columbia
Department of Psychology University of Northern British Columbia Prince George, B.C. Canada V2N 4Z9