Guidelines for choosing between multi-item and single-item scales for construct measurement: a predictive validity perspective (original) (raw)

Marketing research using single-item indicators in structural equation models

2013

This article analyzes the use of single-item indicators in marketing research and their utilization in structural equation modeling (SEM). The study provides a literature review regarding the debate of the use of single-item measures in social sciences research and methodologically in SEM. The analysis of recent studies that use single-item indicators from top marketing journals provides information regarding the types of constructs fit for single-item measurement and their use in SEM. The article presents clarifications to the debate regarding the use of single-item indicators in marketing research, gives examples of types of constructs measurable through single-item indicators and provides recommendations that add knowledge to the empirical analysis and methodology domains of marketing research.

A Census of Multi-Item Scales Used in Marketing Research

1993

A CENSUS OF MULTI-ITEM SCALES USED IN MARKETING RESEARCH The multi-item psychometric scale is a commonly used measurement tool in contemporary marketing research in both academia and industry. Yet, past studies have typically observed that individual measures have been reported with minimal evidence of psychometric quality (Heeler & Ray, 1972,

One, few or many? An integrated framework for identifying the items in measurement scales

Churchill (1979) proposed a detailed procedure for the development of better multi-item measures that has become popular . Recently, however, many scholars have challenged this dominant paradigm . They argue that, in many marketing contexts where the target construct has a precise and concrete definition, long multi-item measures can be substituted by shorter measures with fewer items, or even single-item measures . This has resulted in the controversy around the relative superiority of single-versus multi-item scales . We review the extant literature to summarise various arguments in favour of (or against) multi-item and singleitem measures, respectively . Moreover, we propose an integrated framework for developing a new scale, reducing long multi-item scales to shorter multi-item measures or to single-item measures, or to expand an existing short (single-item) scale . The significant contributions of this paper to the literature are identified .

Marketing Scales Handbook: A Compilation of Multi-Item Measures

1992

MARKETING SCALES HANDBOOK, VOLUME IV: CONSUMER BEHAVIOR covers all new or new uses of previously developed scales in consumer behavior and advertising that appeared between 1998 and 2001. 654 scales are presented along with a description, the origin of the scale, reliability, validity, and other useful information for the academic and professional researcher. This volume includes two indices: by author and by publication.

Multi-item scale usage in marketing journals: 1980 to 1989

Journal of the Academy of Marketing Science, 1993

The use of multi-item scales reported in six marketing journals during the 1980s is analyzed. The analysis replicates some aspects of the Churchill and Peter (1984) study and extends the examination to issues not reviewed previously in marketing. The database for the study is unique in that it attempts to incorporate every instance of scale usage from the defined domain.

On building better construct measures: Implications of a general hierarchical model

Psychology and Marketing, 2008

Four problems occur in the scale development process: (a) defining the construct, (b) drawing items from multiple domains, (c) identifying dimensions, and (d) showing nomological validity. In order to minimize these problems, the authors propose a general hierarchical model (GHM) that provides an organizational structure for placing many of the individual difference constructs used in marketing and consumer behavior. Three principles, which were derived from the GHM, add to the current scale development paradigm: (a) Define and test the construct within a hierarchical network that includes antecedents and consequences, (b) define and test the construct's dimensionality, and (c) match the construct's items to its level in the hierarchical system. By using these steps in scale development, researchers can build more precise measures possessing higher levels of validity and reliability.

The Impact of Item Characteristics on Item and Scale Validity

Multivariate Behavioral Research, 2004

This study describes the relation between personality items' validities, defined as the items' correlations with acquaintance ratings on the Big 5 personality factors, and other itemmetric properties including ambiguity, syntactic complexity, social desirability, content, and trait indicativity. Five external validity coefficients for each item on the California Psychological Inventory were correlated with a number of itemmetric variables often assumed to affect item validity. Item validity correlated positively with social desirability and trait indicativity and negatively with ambiguity across the five factors. Other characteristics had a more limited influence on item validity. Multiple regression analyses revealed trait indicativity-how obviously an item response indicates a trait-to be the most important determinant of item validity. Scales built from itemmetrically sound versus poor items showed differential validity in two additional samples. Implications for the psychological processes underlying responses to personality items are discussed. Goldberg (1993) and Wiggins (1979) are probably best known for their influential work on the Big Five and circumplex taxonomies of personality traits. Unbeknownst to many, however, almost 40 years ago Wiggins and Goldberg (1965) were defining a new field called itemmetrics-the measurement of the properties of individual personality questionnaire items. Much of the early itemmetric research was limited to measuring and reporting Much of this article was written while the author was a visiting research fellow at the University of Bielefeld, supported by a fellowship from the Alexander von Humboldt-Stiftung. I wish to express my gratitude to Alois Angleitner and Robert Wicklund for their help in arranging my research stay, and to the psychology department at Bielefeld and the von Humboldt-Stiftung for their support during that time. I would also like to thank Michael Gerg and Kelly Horner for serving as judges on the item rating tasks in this study. Thanks go to Robert R. McCrae for providing the rational ratings of CPI items according to the five-factor model and to thanks go to the editor of the German CPI, Ansfried Weiner, for generously providing the authorized 462-item revised version of this inventory for the research project before the inventory actually went to press. Finally, I would like to thank