The Listening Styles Profile-Revised (LSP-R): A Scale Revision and Evidence for Validity

Abstract

The Listening Styles Profile (LSP-16) is the most widely used self-report listening instrument in the communication discipline. Unfortunately, researchers have utilized the instrument despite its uncharacteristically low reliability estimates and unvalidated factor structure. The following manuscript presents results from two studies designed to address these limitations. Study 1 proposes a revised measure (Listening Styles Profile-Revised; LSP-R) based on four factors: relational, analytical, task-oriented, and critical listening. Study 2 was designed to further refine and provide validity evidence for the revised scale. Internal consistency estimates and latent variable test–retest correlations showed the LSP-R to be consistent over repeated administrations; the factors were related to number of listening, information processing, empathy, communication trait, and personality variables. Beyond identifying orientations toward attending to others, the instrument developed here possesses heuristic potential for investigating the role and positive potential of listening within a variety of specific research agendas and theoretical perspectives.

Keywords:

• Construct Validity
• Information Processing
• Listening Styles
• Message Reception
• Scale Development

Notes

Note. Internal consistency estimates (Cronbach's alpha) are presented along the diagonal.

*p < .05, **p < .01.

Note. LCI = Listening Concepts Inventory.

*p < .05, **p < .01.

Note. All regression models were statistically significant, p < .001.

All 57 items are available upon request. Since we utilized the original items from the LSP-16, we were able to analyze model fit of the original scale to examine whether the Bodie and Worthington (2010) results could be replicated. In line with their findings, the LSP-16 model generated poor fit statistics (χ ² (98) = 257.05, p < .001, CFI = .78, SRMR = .10), though approximation error was within acceptable standards (RMSEA = .06 (.05, .07)). In addition, 22 standardized residual covariance values were above 2 in absolute value, 14 of the 16 standardized regression weights were below .60, and reliability estimates were well below acceptable standards. Inspection of scale distributions also indicated that each LSP-16 subscale deviated significantly from normality with the people-oriented scale negatively and the other three scales positively skewed and each scale leptokurtic.

Please contact the first author at [email protected] for a copy of the instrument.

Analyses available upon request.

Analyses available upon request.

Additional information

Notes on contributors

Graham D. Bodie

Graham D. Bodie (Ph.D., Purdue University, 2008) is an Assistant Professor in Communication Studies at Louisiana State University.

Debra L. Worthington

Debra L. Worthington (Ph.D., University of Kansas, 1994) is an Associate Professor in the Department of Communication & Journalism at Auburn University.

Christopher C. Gearhart

Christopher C. Gearhart (Ph.D., Louisiana State University, 2012) is an Instructor in Communication Studies at Louisiana State University.