1 Statistical Analysis: Description, Evaluation, and Estimation
1.1 Relationship between Statistics and Management—A Conceptual Model
2 What’s the Problem? Problem Identification, Variables, and Measurement
2.1 What Are the Units of Analysis? How Many Units of Analysis Have Been Observed?
2.2 What Kind of Problem Is Presented?
2.3 What Are the Variables? How Many Variables Are in the Problem?
2.4 What Is the Level of Measurement for Each Variable?
3 Who, What, When, and How Much? One-Variable Description
3.1 One Variable—Nominal Level of Measurement
3.2 One Variable—Ordinal Level of Measurement
3.3 One Variable—Interval or Ratio Level of Measurement
4 What About It? One-Variable Evaluation—Nominal Level of Measurement
4.1 Two Categories
4.2 More than Two Categories
4.3 Assumptions
5 Ranks and Scales: One-Variable Evaluation—Ordinal and Interval Measures
5.1 Ordinal Level of Measurement
5.2 One Variable, Interval Level of Measurement (n > 30)
6 Confidence: One-Variable Estimation
6.1 Confidence Intervals
6.2 Estimation of Sample Size
7 Tables and Graphs: Two-Variable Description
7.1 The Data File
7.2 Crosstabulations
7.3 Scatterplots
8 Two by Two: Two-Variable Evaluation—Nominal—Nominal Measures
8.1 Both Variables Have Two Categories, Unpaired Data, n > 30
8.2 Unpaired Data, 26 < n £ 250
8.3 Unpaired Data, n ≤ 26
8.4 Paired Data
8.5 More Than Two Categories
8.6 Matched Data
9 Order within Groups: Two-Variable Evalutaion—Nominal–Ordinal
9.1 The Nominally Measured Variable Has Two Categories, Unpaired Data
9.2 The Nominally Measured Variable Has Two Categories, Paired Data
9.3 The Nominally Measured Variable Has More than Two Categories
10 A Tale of Two Ranks: Two-Variable Evaluation—Ordinal–Ordinal and Ordinal–Interval Measures
10.1 Spearman’s Rank Correlation Coefficient
10.2 Goodman and Kruskal’s Gamma Statistic
10.3 Somers’ d Statistic
10.4 Kendall’s Tau Statistics
11 t Time with a Bit of ANOVA: Two-Variable Evaluation—Nominal–Interval Measures
11.1 The Independent Variable Has Two Categories
11.2 The Independent Variable Has More than Two Categories
12 Going Straight: Two-Variable Evaluation—Interval–Interval
12.1 Regression Analysis
12.2 Strength of Association—Correlation Analysis
12.3 Correlation is Not Causality
12.4 Assumptions
13 Line-Up: Two-Variable Estimation—Interval–Interval Measures
13.1 Regression Analysis—Example
13.2 Population Regression Coefficient
13.3 Population Coefficient of Determination and Correlation Coefficient
13.4 Confidence Interval for Estimating an Individual Value of Y
14 The Flat Earth Society: More than Two Variables
14.1 Multiple Regression Analysis
Procedures
Glossary of Terms
Rules for Rounding
Glossary of Symbols
Appendix A The Binomial Probability Distribution
Appendix B The Proportion of the Area under the Normal Curve
Appendix C Critical Values of the Z Statistic
Appendix D Critical Values for the Chi-Square Statistic
Appendix E Critical Values of D in the Kolmogorov–Smirnov One-Variable Test
Appendix F Critical Values of the t Statistic
Appendix G Critical Values of the F Statistic
Appendix H Strength of Association Thermometer
Appendix I Critical Values of C in Fisher’s Exact Probability Test
Appendix J Critical Values of U in the Mann–Whitney Test
Appendix K Critical Values of D for the Kolmogorov–Smirnov Two-Variable One-Tailed Test
Appendix L Critical Values of W in the Wilcoxon Test
Appendix M Critical Values of Spearman’s Rank Correlation Coefficient
Appendix N Random Numbers
Back to top