Performing competitive analysis in QFD studies using state multipole moments and bootstrap sampling

ABSTRACT

QFD often makes use of Likert scales to convert opinions into data for analysis. Likert scales are prone to bias in relation to their labeling, interpretation, and distributional assumptions. Analysis is often based on means, without regard to variation or distribution shape, making inferential statistical procedures unreliable. We present a method for extracting increased information from Likert data using State Multipole Moments, and use bootstrap sampling to enable the moments to be statistically compared. We demonstrate application using a published QFD competitor analysis, showing how this method can enable practitioners to perform statistical comparisons resulting in improved decision making.

KEYWORDS:

• quality function deployment (qfd)
• bootstrap sampling
• competitive analysis
• state multipole moments
• likert scale data

About the authors

Zafar Iqbal is an Assistant Professor of statistics at The Islamia University of Bahawalpur, Pakistan, and a doctoral research student based in the School of Engineering and Advanced Technology at Massey University, New Zealand.

Nigel P. Grigg is an Associate Professor in Quality Systems at Massey University. He leads the taught Master of Quality Systems qualification and coordinates postgraduate research in Quality Management, Business Excellence, and related areas. He is a Chartered Quality Professional and Certified Six Sigma Black Belt, and a member of the Chartered Quality Institute, the American Society for Quality, and the New Zealand Organisation for Quality

K. Govindaraju is a Senior Lecturer in statistics in the Institute of Fundamental Sciences at Massey University, New Zealand.

Acknowledgments

The authors would like to express their gratitude to James and Lorinda Camparo, for their help in providing articles, appendices, and an Excel spreadsheet that were used in the development of this article. Our gratitude also goes to the three reviewers and associate editor of Quality Engineering for their helpful comments that greatly improved the original article.