THE IMPACT OF WORLD-CLASS MANUFACTURING PRACTICES ON SMALL MANUFACTURERS - by
David K. Johnson

Chapter IV - Survey and Analysis

Introduction

In order to test the viability of world-class manufacturing practices in a small manufacturing environment a survey instrument was developed and distributed to small manufacturers in the state of Utah. This chapter discusses the methodology used to develop, conduct, and analyze the survey as well as significant findings, which were revealed through the analysis of the data.

Survey Development and Administration

To assure validity of the survey instrument similar studies were researched and used as a foundation to develop a survey instrument suitable for this thesis. Studies, which were used as a basis for the development of the survey instrument, used for this thesis included research conducted by Boyer K., Davy J.A., Fawcett S.E., Flynn B.B, Inman R.A., Al-zu'bi R.S., Saraph J.V., and their respective coauthors. By using proven survey instruments from previous studies as a basis for the development of the survey instrument used for this thesis, many of the issues concerning the validity of the survey instrument are mitigated. Because of this decision many of the questions used in the survey are a direct result of research conducted by the aforementioned authors. Questions from the previously used survey instruments were evaluated to determine their suitability for small manufacturers and for testing the hypotheses of this thesis. In some cases the selected questions were slightly modified in order to improve their applicability to the desired survey instrument. In order to further enhance the validity of the survey instrument, experts from the field of operations management, as well as academic researchers were consulted regarding the selection and development of appropriate questions for measuring world-class manufacturing practices in a small manufacturing environment. After substantial effort and many revisions a suitable survey instrument was developed. (See Appendix A)

The structure of the resulting survey instrument corresponds to the proposed model of this thesis. The survey instrument included sections focused on supply-chain management, lean manufacturing, quality, and resulting performance. The questions in each of these sections attempt to measure the degree to which the respondents apply world-class manufacturing practices to their operations.

In order to determine which companies should be surveyed, a definition of what constitutes a small manufacturer had to be established. Although no correct specifications exist for labeling a company as being small or large, most sources use a cutoff point of 500 employees or less to define a small company. This cutoff point was considered, but later abandoned because the environment of a 20-employee company and a 500-employee company are often considerably different. To determine a reasonable cutoff point for this thesis, a Pareto analysis was performed on all of the companies listed in the Utah Manufacturers Association database. The Pareto analysis revealed that 80% of the manufacturers listed in the database had fewer than 100 employees. In accordance with the 80/20 rule, a determination was made to only include manufacturers having less than 100 employees in the study. For this study, it was also important to assure that participating companies would be large enough that they would be likely to be concerned with world-class manufacturing practices. Because of this, a decision was made to only target companies with more than 20 employees, but less than 100. Applying these constraints to the database search yielded 321 suitable companies to participate in the study.

Representatives from 206 of these companies were contacted by phone and 191 of them agreed to fill out the survey. Of the 191 surveys that where emailed, faxed, or mailed to the company representatives, depending on the their preference, 82 completed surveys were returned yielding a response rate of 43%. The quality of the received data was further validated through a response bias analysis, which compared the data collected from early respondents to data collected from late respondents, and found no evidence of significant response bias.

Back | Next

Copyright © 2005. David K. Johnson. All Rights Reserved.