Adaptive Designs for Likert-Type Data: An Approach for Implementing Marketing Surveys

The authors discuss a method for implementing marketing surveys in which the questionnaire is “adapted” to the individual respondent. This approach (referred to as “adaptive design”) enables the researcher to pose to the respondent only those scale items (usually a subset of the entire scale) that provide useful “information” about his or her attitude. Such adaptive designs are based on a measurement theory, the Latent Trait Theory (LTT), which has been developed largely in the areas of education and psychology. However, much previous research has focused on adaptive designs that are based on binary LTT models. The authors take the first step in examining adaptive designs for Likert-type data within a marketing context. Initially they discuss and evaluate the graded-response LTT model that utilizes all of the information in Likert-type items. Then, using a simulation study, they apply the graded LTT model to implement an adaptive design for the measurement of consumer discontent. The results of the simulation suggest that an adaptive design based on the graded-response LTT performs reasonably well on several criteria. Furthermore, the findings support theoretical predictions that adaptive designs for Likert-type data can achieve measurement efficiency and precision with a substantially smaller item pool. Implications of adaptive designs for marketing researchers and practitioners, and of LTT in general, are discussed.

[1]  Barbara G. Dodd,et al.  Operational Characteristics of Adaptive Testing Procedures Using the Graded Response Model , 1989 .

[2]  Isaac I. Bejar,et al.  A PROCEDURE FOR INVESTIGATING THE UNIDIMENSIONALITY OF ACHIEVEMENT TESTS BASED ON ITEM PARAMETER ESTIMATES , 1980 .

[3]  Melvin R. Novick,et al.  Some latent train models and their use in inferring an examinee's ability , 1966 .

[4]  David J. Weiss,et al.  APPLICATION OF COMPUTERIZED ADAPTIVE TESTING TO EDUCATIONAL PROBLEMS , 1984 .

[5]  R. D. Bock,et al.  Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm , 1981 .

[6]  F. Lord.  Applications of Item Response Theory To Practical Testing Problems , 1980 .

[7]  Vern W. Urry,et al.  TAILORED TESTING: A SUCCESSFUL APPLICATION OF LATENT TRAIT THEORY* , 1977 .

[8]  D. Weiss.  Adaptive testing by computer. , 1985, Journal of consulting and clinical psychology.

[9]  O. C. Ferrell,et al.  Response Rates and Perceived Questionnaire Length in Mail Surveys , 1979 .

[10]  R. Hambleton,et al.  Item Response Theory , 1984, The History of Educational Measurement.

[11]  David Andrich,et al.  An extension of the rasch model for ratings providing both location and dispersion parameters , 1982 .

[12]  David Thissen,et al.  Marginal maximum likelihood estimation for the one-parameter logistic model , 1982 .

[13]  John B. Carroll,et al.  Phillip Justin Rulon (1900–1968) , 1969 .

[14]  L. Humphreys,et al.  Predictions of Academic Performance in Graduate and Professional School , 1977 .

[15]  David J. Weiss,et al.  Bias and Information of Bayesian Adaptive Testing , 1984 .

[16]  Wagner A. Kamakura,et al.  Measuring Consumer Attitudes toward the Marketplace with Tailored Interviews , 1989 .

[17]  Frank B. Baker,et al.  Methodology Review: Item Parameter Estimation Under the One-, Two-, and Three-Parameter Logistic Models , 1987 .

[18]  Gordon G. Bechtel,et al.  Generalizing the Rasch Model for Consumer Rating Scales , 1985 .

[19]  M. R. Novick,et al.  Statistical Theories of Mental Test Scores. , 1971 .

[20]  Jacob Cohen.  The Cost of Dichotomization , 1983 .

[21]  David J. Weiss,et al.  Improving Measurement Quality and Efficiency with Adaptive Testing , 1982 .

[22]  W. J. Lundstrom,et al.  The Development of a Scale to Measure Consumer Discontent , 1976 .

[23]  David Thissen,et al.  A taxonomy of item response models , 1986 .

[24]  Roderick P. McDonald,et al.  Linear Versus Models in Item Response Theory , 1982 .