Rossiter (2007) Predictive Validity of Multiple- vs. Single-Item

Rossiter (2007) Predictive Validity of Multiple- vs. Single-Item

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Journal of Marketing Research Vol. XLIV (May 2007), 175–184 175 © 2007, American Marketing Association ISSN: 0022-2437 (print), 1547-7193 (electronic) *Lars Bergkvist is Assistant Professor of Marketing, Division of Busi- ness and Humanities, UNSW Asia, Singapore (e-mail: l.bergkvist@ unswasia.edu.sg). When this study was conducted, he was Senior Lecturer in Marketing, School of Management and Marketing, University of Wol- longong, Australia. John R. Rossiter is Research Professor of Marketing, Marketing Research Innovation Centre, University of Wollongong, Aus- tralia (e-mail: [email protected]). He is also Visiting Professor of Marketing, Rotterdam School of Management, Erasmus University, the Netherlands. The authors gratefully acknowledge research support from the Association of Swedish Advertisers, the Swedish Newspaper Publish- ers’ Association, and the Advertising Association of Sweden. To read and contribute to reader and author dialogue on JMR , visit http://www.marketingpower.com/jmrblog. LARS BERGKVIST and JOHN R. ROSSITER* This study compares the predictive validity of single-item and multiple- item measures of attitude toward the ad (A Ad ) and attitude toward the brand (A Brand ), which are two of the most widely measured constructs in marketing. The authors assess the ability of A Ad to predict A Brand in copy tests of four print advertisements for diverse new products. There is no difference in the predictive validity of the multiple-item and single-item measures. The authors conclude that for the many constructs in marketing that consist of a concrete singular object and a concrete attribute, such as A Ad or A Brand , single-item measures should be used. The Predictive Validity of Multiple-Item Versus Single-Item Measures of the Same Constructs In his extremely influential Journal of Marketing Research ( JMR ) article, Churchill (1979, p. 66) states the following: “In sum, marketers are much better served with multi-item than single-item measures of their constructs, and they should take the time to develop them.” In making that recommendation, Churchill followed the tradition of psychometrics for the measurement of abilities and traits (e.g., Guilford 1954; and especially Nunnally 1978). In the 28 years since Churchill’s article, academics have increas- ingly used multiple items to measure every marketing con- struct. To be more technically precise, they have used multiple items to measure the attribute of the construct (e.g., attitude, quality, liking), as distinguished from the object of the construct (e.g., a company, a brand, an ad). In his C-OAR-SE procedure for scale development, Rossiter (2002) proposes that if the object can be conceptualized as concrete and singular, it does not require multiple items to represent it in the measure, and if the attribute can be con- ceptualized as concrete, it does not require multiple items either. However, Churchill’s article, as well as Peter’s (1979) JMR article on multiple-item reliability, has influ- enced the measurement of marketing constructs to such an
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 03/06/2012 for the course ECONOMIC 203 taught by Professor Veiga during the Spring '12 term at Universidad Complutense de Madrid.

Page1 / 10

Rossiter (2007) Predictive Validity of Multiple- vs. Single-Item

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online