A Methodology for Appropriate Testing When Data is Heterogeneous Using EXCEL

Main Article Content

Nguyen Khanh
Jimin Lee
Susan Reiser
Donna Parsons
Sara Russell
Robert DeWitt Yearout

Abstract

A Methodology for Appropriate Testing When Data is Heterogeneous was originally published and copy written in the mid-1990s in Turbo Pascal and a 16-bit operating system.  While working on an ergonomic dissertation (Yearout, 1987), the author determined that the perceptual lighting preference data was heterogeneous and not normal.  Drs. Milliken and Johnson, the authors of Analysis of Messy Data Volume I: Designed Experiments (1989), advised that Satterthwaite’s Approximation with Bonferroni’s Adjustment to correct for pairwise error be used to analyze the heterogeneous data. This technique of applying linear combinations with adjusted degrees of freedom allowed the use of t-Table criteria to make group comparisons without using standard nonparametric techniques.  Thus data with unequal variances and unequal sample sizes could be analyzed without losing valuable information.  Variances to the 4th power were so large that they could not be reentered into basic calculators.  The solution was to develop an original software package which was written in Turbo Pascal on a 7 ¼ inch disk 16-bit operating system.  Current operating systems of 32 and 64 bits and more efficient programming languages have made the software obsolete and unusable. Using the old system could result either in many returns being incorrect or the system terminating.  The purpose of this research was to develop a spreadsheet algorithm with multiple interactive EXCEL worksheets that will efficiently apply Satterthwaite’s Approximation with Bonferroni’s Adjustment to solve the messy data problem.  To ensure that the pedagogy is accurate, the resulting package was successfully tested in the classroom with academically diverse students.  A comparison between this technique and EXCEL’s Add-Ins Analysis ToolPak for a t-test Two-Sample Assuming Unequal Variances was conducted using several different data sets.  The results of this comparison were that the EXCEL Add-Ins returned incorrect significant differences.  Engineers, ergonomists, psychologists, and social scientists will find the developed program very useful. A major benefit is that spreadsheets will continue to be current regardless of evolving operating systems’ status.

Article Details

How to Cite
Khanh, N., Lee, J., Reiser, S., Parsons, D., Russell, S., & Yearout, R. D. (2016). A Methodology for Appropriate Testing When Data is Heterogeneous Using EXCEL. Industrial and Systems Engineering Review, 4(1), 54-66. https://doi.org/10.37266/ISER.2016v4i1.pp54-66
Section
Articles
Author Biography

Robert DeWitt Yearout, University of North Carolina at Asheville (UNCA)

Professor Industrial Engineering Management

LTC Speciial Forces US Army (Retired)

References

R. Barger, R. Yearout, , and G. Yates, Statistical Package for Analyzing Messy Ergonomic Data, Advances in Industrial Ergonomics and Safety VII, Proceedings of the Annual International Foundation for Industrial Ergonomics and Safety Research Conference, Taylor and Francis, London 1995, pp. 203-210.

Microsoft (2013). Microsoft Excel. Redmond, Washington: Microsoft, 2013. Computer Software.

Milliken, G. and Johnson, D. 1984, Analysis of Messy Data Volume I: Designed Experiments, Belmont: Lifetime Learning Publications. pp. 19-25

Netter, J., Wasserman, W., and Kutner, M., 1990, Applied Linear Statistical Models, 3ed, Irwin, Boston, pp. 160-161 and 734-735.

Satterthwaite, F.E., 1946. Biometrics Bulletin, No. 2 pp. 11-114

Sprent, P., 1989, Applied Non-Parametric Statistical Methods, Chapman & Hall, New York, pp 1-4.

Yearout, R. 1987, Task Lighting for Visual Display Unit Work Stations, Kansas State University, Manhatten, Annex 1 Appendix G.

Yearout, R. Barger, R. Yates, G. and Lisnerski D. A Methodology for Appropriate Testing When Data are Heterogeneous, International Journal of Industrial Ergonomics , 1999, Elsevier Science NL, Amsterdam The Netherlands. Vol. 24, No. 1, pp. 129-134.

Most read articles by the same author(s)