Threats to the Validity of Logistics and Supply Chain Management Research

• Date: 01 March 2019
• DOI: http://doi.org/10.1111/jbl.12203
• Published date: 01 March 2019

Threats to the Validity of Logistics and Supply Chain Management

Research

Michael S. Garver

Central Michigan University

To increase the relevance of logistics and supply chain academic research, this paper recommends the development and testing of middle-

range theory and practice-level theory. Yet, there are a number of research issues that arise when academic researchers test middle-range

and practice-level theory, both in measuring constructs and in testing theoretical relationships between constructs. Concerning the measurement

of constructs, this paper recommends that academic researchers pay more attention to content validity and undertake rigorous processes to

ensure content validity. In addition, academic researchers need to more explicitly define constructs as either reflective or formative. If the con-

struct is defined as formative, then the traditional statistical approaches to validate these measurement scales are not recommended. The appro-

priate use of employing single-item measures for concrete constructs is discussed. In regard to conducting hypothesis tests, research issues

associated with multicollinearity and omitted variable bias are discussed. Relative weight analysis is ideal for testing theoretical models and

research hypotheses when survey data are obtained, multicollinearity is present, and there are a large number of independent variables predicting

a dependent variable. Thus, relative weight analysis is ideal for testing research hypotheses in logistics and supply chain management.

Keywords: relevance; middle-range theory; practice-level theory; construct validity; content validity; C-OAR-SE; reflective; formative

single-item measures; multicollinearity; omitted variable bias; relative weight analysis; hypothesis testing; research methods

INTRODUCTION

Being a doctoral student at the University of Tennessee

(1995–1999) was an incredible learning experience, as the halls

were filled with motivated and bright professors and students. I

was blessed to have Dr. Tom Mentzer as a professor, mentor,

and friend, one who encouraged all of the doctoral students to

have an impact on the discipline.

Tom had a vision for the logistics discipline, which was to

increase the scientific rigor of logistics academic research by

conducting theory and hypothesis testing research. A key part of

this vision was to achieve scientific rigor and to maintain the rel-

evance of logistics academic research to practitioners. Tom’s

vision helped to drive the evolution of logistics academic

research at the Journal of Business Logistics, which was moving

away from research that focused on descriptive statistics and

moving toward research that focused upon theory development

followed by rigorous theory and hypothesis testing research.

The Garver and Mentzer (1999) article played an important

role in this evolution, by specifically addressing how to test for

construct validity with structural equation modeling (SEM). The

goal was to take a very technical and complicated topic, and to

make it easy to read, understand, and implement. Without ques-

tion, there were a few key articles in the Journal of Business

Logistics that strongly influenced our work, and they included

the following: Dunn, Seaker, and Waller (1994); Mentzer and

Kahn (1995); and Mentzer and Flint (1997).

Tom’s vision became reality as the evolution of logistics aca-

demic research occurred. In 2005, academic researchers assessed

the state of logistics academic research and found that survey

research dominated the logistics research landscape and that the-

ory and hypothesis testing research increased dramatically over

the prior ten years (Frankel et al. 2005; Sachan and Datta 2005;

Spens and Kov

acs 2005). Commensurate with this increase,

SEM and multiple regression were the most common statistical

tools employed for testing theoretical models and hypothesized

theoretical relationships (Sachan and Datta 2005).

The purpose of this article is to continue the evolution of

logistics research and to chart a new course for future logistics

and supply chain management (SCM/L) research. Whereas the

goal of the Garver and Mentzer (1999) article was to help

researchers do “good”scientific research, the goal of this article

is to help researchers do “better”scientific research. To accom-

plish this purpose, a literature review will first address the rele-

vance of theoretical development at different levels, which

include general theory, middle-range theory, and practice-level

theory. Then, research issues associated with measuring con-

structs will be put forth (ensuring content validity, defining con-

structs as reflective or formative, and employing single-item

measures) along with a new process for developing valid mea-

sures. Finally, research issues that threaten the validity of con-

ducting hypothesis testing research will be put forth

(multicollinearity and omitted variable bias), as well as a

relatively new statistical technique (relative weight analysis) that

overcomes these problems. The evolution of logistics research

continues to forge ahead.

LITERATURE REVIEW

The relevance of theory development and testing

Developing and testing theoretical models is an important goal

of SCM/L academic research (Mentzer and Flint 1997). The first

step of this process is to create and develop a theoretical model,

which contains hypothesized theoretical relationships between

Corresponding author:

Michael S. Garver, Department of Marketing and HSA, Central

Michigan University, 100 Smith Hall, Mount Pleasant, 48804, MI,

USA; E-mail: garve1ms@cmich.edu

Journal of Business Logistics, 2019, 40(1): 30–43 doi: 10.1111/jbl.12203

© 2019 Council of Supply Chain Management Professionals