Employing Latent Class Regression Analysis to Examine Logistics Theory: An Application of Truck Driver Retention

Abstract

Multiple regression analysis assumes that one model or theory is relevant for the entire population, yet research has shown that this assumption is often false and may severely limit valid theory development and testing. Latent class regression analysis overcomes this limitation and allows the researcher to identify and develop regression models that are relevant for different segments within the same population. Latent class regression analysis is introduced and is used to analyze truck drivers' intentions to stay with the same firm. This article demonstrates the advantages of testing logistics theory with latent class regression analysis and provides numerous applications for practitioners.

Take This Job and Shove It: Examining the Influence of Role Stressors and Emotional Exhaustion on Organizational Commitment and Identification in Professional Truck Drivers

“…Some of this churning can be attributed to increased driver demand. However, other reasons cited have centered on compensation packages, unsatisfactory equipment, strained relationships with management, and inadequate driver support programs (Richard et al 1995;Garver et al 2008;Watson 2011 (Berry and Parasuraman 1992;Berry et al 1976). Although truck drivers, unlike traditional employees may be "physically removed" from the day-to-day operations of an organization, maintaining effective communication and relations with drivers regarding the company's goals and values is paramount.…”

Carrier Selection:

“…Thus, latent class cluster analysis (LCCA) was used to determine whether unique need-based customer segments exist in terms of the importance attached to the carrier-selection variables. LCCA also produces both fi t statistics that guide the selection of the appropriate number of segments, and probabilities of segment membership, which are helpful in determining how well the technique has segmented the market (Garver, Williams, and Taylor 2008 ). LCCA also produces both fi t statistics that guide the selection of the appropriate number of segments, and probabilities of segment membership, which are helpful in determining how well the technique has segmented the market (Garver, Williams, and Taylor 2008 ).…”

“…Because it is diffi cult for practitioners to work with more than fi ve segments, previous segmentation studies suggest no more than fi ve segments (e.g., Garver, Williams, and Taylor 2008 ). Because it is diffi cult for practitioners to work with more than fi ve segments, previous segmentation studies suggest no more than fi ve segments (e.g., Garver, Williams, and Taylor 2008 ).…”

Needs assessment of university leadership programs

“…Examining the overall results can be misleading because the student segments may place differing importance values on the attributes (Garver 2009;Garver, Divine, & Spralls, 2009;Garver, Williams, & Taylor, 2008). Differences among segments will often remain hidden when examining overall results.…”