Frontiers of Entrepreneurship Research
1997 Edition

Return to 1997 Topical Index

Order hard copy editions of Frontiers of  Entrepreneurship Research by mail


A METHODOLOGICAL INVESTIGATION OF THE VALIDITY AND USEFULNESS OF PARAMETRIC AND NONPARAMETRIC STATISTICAL DATA ANALYSIS TECHNIQUES FOR NEW VENTURE RESEARCH*

Kenneth C. Robinson, Kennesaw State University
Charles W. Hofer, University of Georgia

*This research was funded in part by the Center for Entrepreneurial Leadership Inc. and the Ewing Marion Kauffman Foundation,


INTRODUCTION
STATISTICAL DATA ANALYSIS ISSUES IN ENTREPRENEURSHIP RESEARCH

Distributional Assumptions Underlying Statistical Data Analysis Techniques
Prior Research on Distributional Characteristics of Dependent Variables

NEW VENTURE RESEARCH: METHODOLOGICAL APPROACHES AND RESULTS

Prior Findings vis--vis Scope
Prior Findings via-a-vis Strategic Weapons

RESEARCH METHODS

Research Design
Sample/Site Selection
Variable Specification & Classification
Data Gathering Methods
Data Analysis Techniques for Assessing Distributional Characteristics
Data Analysis Techniques for Testing the Influence of New Venture Strategy on Performance.
TABLE 1: Guidelines for Selecting Parametric and Nonparametric Location Parameter Tests

RESEARCH FINDINGS

Normality Assumption
TABLE 2: Test Statistics for Assessing the Normality Assumption
Equal Variance Assumptions
TABLE 3: Equality of Variance Test for Scope Classifications
Continuous Distribution Assumption
Discussion of Variable Transformations
Summary of Findings on Distributional Characteristics
Comparing the Validity and Robustness of Nonparametric and Parametric Techniques: Influence of Scope on New Venture Economic Performance
TABLE 4: Nonparametric and Parametric Analysis of Variance Results for Scope Variable
TABLE 5: Nonparametric and Parametric Multiple Comparison Results for Scope Variable
Comparing the Validity and Robustness of Nonparametric and Parametric Techniques: Influence of Strategic Weapons on New Venture Performance
TABLE 6: Nonparametric and Parametric Pairwise Comparison Results for Scope Variable
TABLE 7: Nonparametric and Parametric Analysis of Variance Results for Strategic Weapons Variable
TABLE 8: Nonparametric and Parametric Multiple Comparison Results for Strategic Weapons Variable
TABLE 9: Nonparametric and Parametric Pairwise Comparison Results for Strategic Weapons Variable

COMPARISON OF NONPARAMETRIC AND PARAMETRIC RESULTS

TABLE 10: Comparison of Nonparametric and Parametric Results for New Venture Strategy Classes

SUMMARY AND CONCLUSIONS
REFERENCES

ABSTRACT

This study examined the validity and usefulness of analogous parametric and nonparametric statistical data analysis techniques for exploring relationships among various new venture strategies and new venture economic performance. This study of nearly 200 new ventures found that (1) the assumptions underlying the theoretical development of parametric statistical data analysis techniques were substantially violated with the data utilized in this study; and (2) nonparametric statistical data analysis techniques produced substantially more statistically significant results than did parametric techniques.

 

Return to 1997 Topical Index


1997 Babson College All Rights Reserved
Last Updated 06/01/98