Presentation is loading. Please wait.

Presentation is loading. Please wait.

©The McGraw-Hill Companies, Inc., 2001Irwin/McGraw-Hill Donald Cooper Pamela Schindler Chapter 17 Business Research Methods.

Similar presentations


Presentation on theme: "©The McGraw-Hill Companies, Inc., 2001Irwin/McGraw-Hill Donald Cooper Pamela Schindler Chapter 17 Business Research Methods."— Presentation transcript:

1 ©The McGraw-Hill Companies, Inc., 2001Irwin/McGraw-Hill Donald Cooper Pamela Schindler Chapter 17 Business Research Methods

2 ©The McGraw-Hill Companies, Inc., 2001Irwin/McGraw-Hill Chapter 17 Hypothesis Testing

3  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 1 Approaches to Hypothesis Testing Classical Statistics äsampling-theory approach äobjective view of probability ädecision making rests on analysis of available sampling data Bayesian Statistics äextension of classical statistics äconsider all other available information

4  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 2 Types of Hypotheses Null äthat no statistically significant difference exists between the parameter and the statistic being compared Alternative älogical opposite of the null hypothesis äthat a statistically significant difference does exist between the parameter and the statistic being compared.

5  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 3 Logic of Hypothesis Testing Two tailed test änondirectional test äconsiders two possibilities One tailed test ädirectional test äplaces entire probability of an unlikely outcome to the tail specified by the alternative hypothesis

6  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 4 Decision Errors in Testing Type I error äa true null hypothesis is rejected Type II error äone fails to reject a false null hypothesis

7  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Testing for Statistical Significance äState the null hypothesis äChoose the statistical test Select the desired level of significance äCompute the calculated difference value Obtain the critical value äInterpret the test Slide 17 - 5

8  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 6 Classes of Significance Tests Parametric tests  Z or t test is used to determine the statistical significance between a sample distribution mean and a population parameter äAssumptions: äindependent observations änormal distributions äpopulations have equal variances äat least interval data measurement scale

9  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 7 Classes of Significance Tests Nonparametric tests äChi-square test is used for situations in which a test for differences between samples is required äAssumptions äindependent observations for some tests only änormal distribution not necessary ähomogeneity of variance not necessary äappropriate for nominal and ordinal data, may be used for interval or ratio data

10  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Slide 17 - 8 How to Test the Null Hypothesis Analysis of variance (ANOVA) äthe statistical method for testing the null hypothesis that means of several populations are equal

11  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style Multiple Comparison Tests äMultiple comparison procedures ätest the difference between each pair of means and indicate significantly different group means at a specified alpha level (<.05) äuse group means and incorporate the MS error term of the F ratio Slide 17 - 9

12  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style How to Select a Test äWhich does the test involve? one sample, two samples k samples If two or k samples,are the individual cases independent or related? äIs the measurement scale nominal, ordinal, interval, or ratio? Slide 17 - 10

13  The McGraw-Hill Companies, Inc., 2001 Irwin/McGraw-Hill Click to edit Master title style K Related Samples Test Use when: äThe grouping factor has more than two levels äObservations or subjects are ämatched... or äthe same subject is measured more than once äInterval or ratio data Slide 17 - 11


Download ppt "©The McGraw-Hill Companies, Inc., 2001Irwin/McGraw-Hill Donald Cooper Pamela Schindler Chapter 17 Business Research Methods."

Similar presentations


Ads by Google