Bmgt 311 chapter_13
0 likes414 views
This document discusses techniques for testing differences between groups, including t-tests, z-tests, and analysis of variance (ANOVA). It explains that differences must be statistically significant, meaningful, stable, and actionable to be useful for market segmentation. Examples are provided to demonstrate how to test differences between percentages and means for two independent groups using t-tests or z-tests. The document also discusses how ANOVA can be used to test differences between three or more groups and post-hoc tests can identify which specific groups differ. Paired sample tests are introduced to analyze differences within the same sample. Proper reporting of results to clients is also emphasized.
Bmgt 311 chapter_13
- 1. BMGT 311: Chapter 13 Implementing Basic Differences Tests 1
- 2. Learning Objectives • To learn how differences are used for market segmentation decisions • To understand when t tests or z tests are appropriate • To be able to test the differences between two percentages or means for two independent groups • To know what a paired samples difference test is and when to use it • To comprehend ANOVA and how to interpret ANOVA output 2
- 3. 3
- 4. Why Differences are Important • Market segmentation is based on differences between groups of consumers. • One commonly used basis for market segmentation is the discovery of differences that are the following: • Statistically significant • Meaningful • Stable • Actionable differences 4
- 5. Why Differences Are Important Market Segmentation • Differences must be statistically significant: the differences found in the sample(s) truly exist in the population(s) from which the random samples are drawn. • Differences must be meaningful: one that the marketing manager can potentially use as a basis for marketing decisions. • Differences should be stable: one that will be in place for the foreseeable future. • Differences must be actionable: the marketer can focus various marketing strategies and tactics, such as product design or advertising, on the market segments to accentuate the differences between segments. 5
- 6. Testing for Significant Differences Between Two Groups • Statistical tests are used when researcher wants to compare the means or percentages of two different groups or samples. • t Test: statistical inference test to be used with small sample sizes (n ≤ 30) • z Test: statistical inference test to be used when the sample size is 30 or greater • Note: Most computer statistical programs report only the t value because it is identical to the z value with large samples. 6
- 7. Differences Between Percentages with Two Groups (Independent Samples) • Independent samples are treated as representing two potentially different populations. • Null hypothesis: the hypothesis that the difference in the population parameters is equal to zero • With a differences test, the null hypothesis states that there is no difference between the percentages (or means) being compared. • Significance of differences between two percentages: alternative to the null hypothesis is that there is a true difference between the population parameters. 7
- 8. 8
- 9. Differences Between Percentages with Two Groups (Independent Samples) • Formula for Standard Error of Percentages: • Sp1-p2 = Square Root ((p1 x q1)/n1 + (p2 x q2/n2)) 9
- 10. How Do You Know When the Results Are Significant? • If the null hypothesis is true, we would expect there to be n0 differences between the two percentages. • Yet we know that, in any given study, differences may be expected due to sampling error. • If the null hypothesis were true, we would expect 95% of the z scores computed from 100 samples to fall between +1.96 and −1.96 standard errors. 10
- 11. How Do You Know When the Results Are Significant? • If the computed z value is greater than +1.96 or −1.96, it is not likely that the null hypothesis of no difference is true. Rather, it is likely that there is a real statistical difference between the two percentages. 11
- 12. Example: Page 330 • Last year a Harris Poll showed 40% of surveyed companies were coming to college campuses to hire seniors (n = 300 companies surveyed). • This year, the Harris Poll reported the percentage is 65% (n = 100 companies surveyed). • Is this a significant difference? 12
- 13. Example: Page 330 • Applying the formula: P1 = 65 and P2 = 40, n1 = 100, n2 = 300 • z = 4.51 • Since the z value is greater than +1.96, the difference between the two percentages is significant. 13
- 14. In Class Example #1 • Is there a significant difference? • Last year a Point Park Poll showed 75% of surveyed companies were coming to college campuses to hire seniors (n = 200 companies surveyed). • This year, the Point Park Poll reported the percentage is 70% (n = 150 companies surveyed). • Is this a significant difference? 14
- 15. In Class Example #1 (Refer to Page 330) • p1 = • Q1 = • n1 = • p2 = • Q2 = • n2 = • z = • Last year a Point Park Poll showed 75% of surveyed companies were coming to college campuses to hire seniors (n = 200 companies surveyed). • This year, the Point Park Poll reported the percentage is 70% (n = 150 companies surveyed). • Is this a significant difference? 15
- 16. In Class Example #2 16
- 17. In Class Example #2 • Is there a significant difference? • Last year a Point Park Poll showed 80% of surveyed companies were coming to college campuses to hire seniors (n = 400 companies surveyed). • This year, the Point Park Poll reported the percentage is 65% (n = 300 companies surveyed). • Is this a significant difference? 17
- 18. In Class Example #2 • Is there a significant difference? • Last year a Point Park Poll showed 80% of surveyed companies were coming to college campuses to hire seniors (n = 400 companies surveyed). • This year, the Point Park Poll reported the percentage is 65% (n = 300 companies surveyed). • Is this a significant difference? • p1 = • Q1 = • n1 = • p2 = • Q2 = • n2 = • z = 18
- 19. Testing the Difference Between Means • Differences between two means from independent samples • Differences between three or more means from independent samples • Differences between paired means 19
- 20. Differences Between Means with Two Groups (Independent Samples) • The procedure for testing the significance of difference between two means from two different groups is identical to the procedure for testing two percentages. • Equations differ due to the use of a metric (interval or ratio) scale. 20
- 21. Differences Between Means with Two Groups (Independent Samples) 21
- 22. An Example: Testing the Difference Between Two Means (Page 333) • Do male teens and female teens drink different amounts of sports drinks? 22
- 23. An Example: Testing the Difference Between Two Means • The difference between males (9 bottles) and females (7.5 bottles) is significant; z =6.43. 23
- 24. Example 3: Testing the Difference Between Two Means • x1 = 7 • x2 = 7.5 • s1 = 1.5 • s2 = 1 • n1 = 200 • n2 = 200 • z = 24
- 25. Example 4: Testing the Difference Between Two Means • x1 = 9 • x2 = 9.5 • s1 = 2 • s2 = 2.5 • n1 = 300 • n2 = 300 • z = 25
- 26. Analysis of Variance • Analysis of variance (ANOVA): used when comparing the means of three or more groups • ANOVA is an investigation of the differences between the group means to ascertain whether sampling errors or true population differences explain their failure to be equal. • ANOVA will “flag” when at least one pair of means has a statistically significant difference, but it does not tell which pair. • Green flag procedure: If at least one pair of means has a statistically significant difference, ANOVA will signal this by indicating significance 26
- 27. 27
- 28. 28
- 29. ANOVA Advantages • ANOVA has two distinct advantages over performing multiple t tests of the significance of the difference between means. • Immediately notifies researcher if there is any significant difference • Arranges the means so the significant differences can be located and interpreted easily 29
- 30. Post Hoc Tests: Detect Statistically Significant Differences Among Group Means • Post hoc tests: options that are available to determine where the pair(s) of statistically significant differences between the means exist(s) • Duncan’s multiple range test: provides output that is mostly a “picture” of what means are significantly different • The Duncan multiple range test’s output is much less statistical than most other post hoc tests and is easy to interpret. 30
- 31. Differences Between Two Means Within the Same Sample (Paired Sample) • You can test the significance of the difference between two means for two different questions answered by the same respondents using the same scale. • Paired samples test for the differences between two means: a test to determine if two means of two different questions using the same scale format and answered by the same respondents in the sample are significantly different. 31
- 32. Reporting Group Differences Tests to Clients • Differences may not be obvious to the client, especially if the researcher does not take care to highlight them. • Group comparison table: summarizes the significant differences in an efficient manner • Reporting of findings has a significant ethical burden for marketing researchers, as they cannot choose to report only “good news” to clients. 32