Limits...
Constrained Dual Scaling for Detecting Response Styles in Categorical Data.

Schoonees PC, van de Velden M, Groenen PJ - Psychometrika (2015)

Bottom Line: Response styles occur when respondents use rating scales differently for reasons not related to the questions, often biasing results.A spline-based constrained version of DS is devised which can detect the presence of four prominent types of response styles, and is extended to allow for multiple response styles.An alternating nonnegative least squares algorithm is devised for estimating the parameters.

View Article: PubMed Central - PubMed

Affiliation: Econometric Institute, Erasmus University Rotterdam, Rotterdam, The Netherlands. schoonees@gmail.com.

ABSTRACT
Dual scaling (DS) is a multivariate exploratory method equivalent to correspondence analysis when analysing contingency tables. However, for the analysis of rating data, different proposals appear in the DS and correspondence analysis literature. It is shown here that a peculiarity of the DS method can be exploited to detect differences in response styles. Response styles occur when respondents use rating scales differently for reasons not related to the questions, often biasing results. A spline-based constrained version of DS is devised which can detect the presence of four prominent types of response styles, and is extended to allow for multiple response styles. An alternating nonnegative least squares algorithm is devised for estimating the parameters. The new method is appraised both by simulation studies and an empirical application.

Show MeSH
Response styles used in the simulation study. Each curve represents a different style.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4644217&req=5

Fig4: Response styles used in the simulation study. Each curve represents a different style.

Mentions: The 108 experimental conditions consisted of the following. The number of objects was varied over 10, 20 and 30 items. The rating scales employed were either or -point scales. Sample sizes of , and , respectively, were used. The number of groups were either 3 or 5. For each of these , it was assumed that one of the groups has a linear response mapping (that is, a group with no response style). The additional groups exhibited response styles through nonlinear mappings. For , these additional groups were acquiescence and extreme responding, since Baumgartner and Steenkamp (2001) found that these are most prevalent in survey data. For , groups for disacquiescence and midpoint responding were also added. The corresponding spline functions used to simulate from are shown in FigureĀ 4. The sample of respondents was assigned to the groups by allocating either 20, 50 or 80 % of respondents equally among the response style groups. These percentages represent the amount of contamination in the simulated data. The remaining percentage of respondents was assigned to the group exhibiting no response style. The latent standard deviation was fixed at 0.1 for all experiments.Fig. 4


Constrained Dual Scaling for Detecting Response Styles in Categorical Data.

Schoonees PC, van de Velden M, Groenen PJ - Psychometrika (2015)

Response styles used in the simulation study. Each curve represents a different style.
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4644217&req=5

Fig4: Response styles used in the simulation study. Each curve represents a different style.
Mentions: The 108 experimental conditions consisted of the following. The number of objects was varied over 10, 20 and 30 items. The rating scales employed were either or -point scales. Sample sizes of , and , respectively, were used. The number of groups were either 3 or 5. For each of these , it was assumed that one of the groups has a linear response mapping (that is, a group with no response style). The additional groups exhibited response styles through nonlinear mappings. For , these additional groups were acquiescence and extreme responding, since Baumgartner and Steenkamp (2001) found that these are most prevalent in survey data. For , groups for disacquiescence and midpoint responding were also added. The corresponding spline functions used to simulate from are shown in FigureĀ 4. The sample of respondents was assigned to the groups by allocating either 20, 50 or 80 % of respondents equally among the response style groups. These percentages represent the amount of contamination in the simulated data. The remaining percentage of respondents was assigned to the group exhibiting no response style. The latent standard deviation was fixed at 0.1 for all experiments.Fig. 4

Bottom Line: Response styles occur when respondents use rating scales differently for reasons not related to the questions, often biasing results.A spline-based constrained version of DS is devised which can detect the presence of four prominent types of response styles, and is extended to allow for multiple response styles.An alternating nonnegative least squares algorithm is devised for estimating the parameters.

View Article: PubMed Central - PubMed

Affiliation: Econometric Institute, Erasmus University Rotterdam, Rotterdam, The Netherlands. schoonees@gmail.com.

ABSTRACT
Dual scaling (DS) is a multivariate exploratory method equivalent to correspondence analysis when analysing contingency tables. However, for the analysis of rating data, different proposals appear in the DS and correspondence analysis literature. It is shown here that a peculiarity of the DS method can be exploited to detect differences in response styles. Response styles occur when respondents use rating scales differently for reasons not related to the questions, often biasing results. A spline-based constrained version of DS is devised which can detect the presence of four prominent types of response styles, and is extended to allow for multiple response styles. An alternating nonnegative least squares algorithm is devised for estimating the parameters. The new method is appraised both by simulation studies and an empirical application.

Show MeSH