18-1

Chapter 18
Hypothesis Testing

18-2

Hypothesis Testing

Inductive Reasoning

Deductive Reasoning

18-3

Statistical Procedures

Inferential Statistics

Descriptive Statistics

Exhibit 18-1 Hypothesis Testing and the Research Process

18-4

18-5

Approaches to Hypothesis Testing

Classical statistics Objective view of probability Established hypothesis is rejected or fails to be rejected Analysis based on sample data Bayesian statistics Extension of classical approach Analysis based on sample data Also considers established subjective probability estimates

18-6

Statistical Testing Procedures

State null hypothesis Interpret the test Choose statistical test

Stages

Obtain critical test value

Compute difference value

Select level of significance

18-7

Tests of Significance

Parametric

Nonparametric

18-8

Assumptions for Using Parametric Tests

Independent observations

Normal distribution

Equal variances

Interval or ratio scales

18-9

Advantages of Nonparametric Tests

Easy to understand and use Usable with nominal data

Appropriate for ordinal data

Appropriate for non-normal population distributions

18-10

How To Select A Test

How many samples are involved?

If two or more samples are involved, are the individual cases independent or related? Is the measurement scale nominal, ordinal, interval, or ratio?

18-11

Exhibit 18-7 Recommended Statistical Techniques

Two-Sample Tests
__________________________________________ __

k-Sample Tests
__________________________________________ __

Measureme nt Scale Nominal

One-Sample Case Binomial x2 one-sample test KolmogorovSmirnov onesample test Runs test

Related Samples McNemar

Independent Samples Fisher exact test x2 twosamples test Median test Mann-Whitney U KolmogorovSmirnov WaldWolfowitz t-test Z test

Related Samples Cochran Q

Independent Samples x2 for k samples

Ordinal

Sign test Wilcoxon matched-pairs test

Friedman twoway ANOVA

Median extension Kruskal-Wallis one-way ANOVA

Interval and Ratio

t-test Z test

t-test for paired samples

Repeatedmeasures ANOVA

One-way ANOVA n-way ANOVA

18-12

Questions Answered by One-Sample Tests Is there a difference between observed frequencies and the frequencies we would expect? Is there a difference between observed and expected proportions? Is there a significant difference between some measures of central tendency and the population parameter?

18-13

Parametric Tests

Z-test

t-test

18-14

One-Sample t-Test Example

Null Statistical test Significance level Calculated value Critical test value Ho: = 50 mpg t-test .05, n=100 1.786 1.66 (from Appendix C, Exhibit C-2)

18-15

One Sample Chi-Square Test Example

Expected Frequencies (percent x 60) 27

Living Arrangement

Intend to Join

Number Interviewed

Percent (no. interviewed/200)

Dorm/fraternity
Apartment/rooming house, nearby Apartment/rooming house, distant Live at home

16

90

45

13

40

20

12

16 15
_____

40 30
_____

20 15
_____

12 9
_____

Total

60

200

100

60

18-16

One-Sample Chi-Square Example

Null Statistical test Significance level Calculated value Critical test value Ho: 0 = E One-sample chi-square .05 9.89 7.82 (from Appendix C, Exhibit C-3)

18-17

Two-Sample Parametric Tests

18-18

Two-Sample t-Test Example

A Group

B Group

Average hourly sales Standard deviation

X1 = $1,500 s1 = 225

X2 = $1,300 s2 = 251

18-19

Two-Sample t-Test Example

Null Statistical test Significance level Calculated value Critical test value Ho: A sales = B sales t-test .05 (one-tailed) 1.97, d.f. = 20 1.725 (from Appendix C, Exhibit C-2)

18-20

Two-Sample Nonparametric Tests: Chi-Square

On-the-Job-Accident Cell Designation Count Expected Values Smoker Heavy Smoker

Yes 1,1 12, 8.24 2,1 9 7.73 3,1 13 18.03 34

No 1,2 4 7.75 2,2 6 7.27 3,2 22 16.97 32

Row Total 16

Moderate

15

Nonsmoker

35

Column Total

66

18-21

Two-Sample Chi-Square Example

Null There is no difference in distribution channel for age categories. Statistical test Chi-square Significance level .05 Calculated value 6.86, d.f. = 2 Critical test value 5.99 (from Appendix C, Exhibit C-3)

18-22

Two-Related-Samples Tests

Parametric

Nonparametric

18-23

Exhibit 18-9 Sales Data for Paired-Samples t-Test

Company GM GE Exxon IBM Ford AT&T Mobil DuPont Sears Amoco Total Sales Year 2 126932 54574 86656 62710 96146 36112 50220 35099 53794 23966 Sales Year 1 123505 49662 78944 59512 92300 35173 48111 32427 49975 20779 Difference D 3427 4912 7712 3192 3846 939 2109 2632 3819 3187 D = 35781 D2 11744329 24127744 59474944 10227204 14971716 881721 4447881 6927424 14584761 10156969 D = 157364693
.

18-24

Paired-Samples t-Test Example

Null Statistical test Significance level Calculated value Critical test value Year 1 sales = Year 2 sales Paired sample t-test .01 6.28, d.f. = 9 3.25 (from Appendix C, Exhibit C-2)

18-25

Related-Samples Nonparametric Tests: McNemar Test

Before

After Do Not Favor

After Favor

Favor

Do Not Favor

18-26

An Example of the McNemar Test

Before
After Do Not Favor After Favor

Favor

A=10

B=90

Do Not Favor

C=60

D=40

18-27

k-Independent-Samples Tests: ANOVA Tests the null hypothesis that the means of three or more populations are equal One-way: Uses a single-factor, fixedeffects model to compare the effects of a treatment or factor on a continuous dependent variable

18-28

Exhibit 18-12 ANOVA Example

__________________________________________Model Summary_________________________________________ Source Model (airline) Residual (error) Total d.f. 2 57 59 Sum of Squares 11644.033 11724.550 23368.583 Mean Square 5822.017 205.694 F Value 28.304 p Value 0.0001

_______________________Means Table________________________ Count Delta Lufthansa KLM 20 20 20 Mean 38.950 58.900 72.900 Std. Dev. 14.006 15.089 13.902 Std. Error 3.132 3.374 3.108

All data are hypothetical

18-29

ANOVA Example Continued

Null Statistical test Significance level Calculated value Critical test value A1 = A2 = A3 ANOVA and F ratio .05 28.304, d.f. = 2, 57 3.16 (from Appendix C, Exhibit C-9)

18-30

Post Hoc: Scheffes S Multiple Comparison Procedure Crit. Diff. 11.400 11.400 11.400

Verses Delta Lufthansa KLM Lufthansa KLM

Diff 19,950 33.950 14.000

p Value .0002 .0001 .0122

18-31

Exhibit 18-13 Multiple Comparison Procedures

Test Fisher LSD Bonferroni Tukey HSD Tukey-Kramer Games-Howell Tamhane T2 Scheff S Brown-Forsythe Newman-Keuls Duncan Dunnets T3 Dunnets C X X Complex Comparisons X X X X X X X X X X X X X X X X X X X X X X X Pairwise Comparisons Equal ns Only Unequal ns X X X X X X Equal Variances Assumed X Unequal Variances Not Assumed

18-32

Exhibit 18-14 ANOVA Plots

18-33

Exhibit 18-15 Two-Way ANOVA Example

__________________________________________Model Summary_________________________________________ Source Airline Seat selection Airline by seat selection Residual d.f. 2 1 2 54 Sum of Squares 11644.033 3182.817 517.033 8024.700 Mean Square 5822.017 3182.817 258.517 148.606 F Value 39.178 21.418 1.740 p Value 0.0001 0.0001 0.1853

__________Means Table Effect: Airline by Seat Selection___________ Count Delta economy Delta business Lufthansa economy 10 10 10 Mean 35.600 42.300 48.500 Std. Dev. 12.140 15.550 12.501 Std. Error 3.839 4.917 3.953

Lufthansa business
KLM economy KLM business

10
10 10

69.300
64.800 81.000

9.166
13.037 9.603

2.898
4.123 3.037

All data are hypothetical

18-34

k-Related-Samples Tests More than two levels in grouping factor Observations are matched

Data are interval or ratio

18-35

Exhibit 18-17 RepeatedMeasures ANOVA Example

__________________________________________________________Model Summary_________________________________________________________

Source
Airline Subject (group) Ratings Ratings by air....... Ratings by subj.....

d.f.
2 57 1 2 57

Sum of Squares
3552735.50 15067.650 625.633 2061.717 2490.650

Mean Square
17763.775 264.345 625.633 1030.858 43.696

F Value
67.199

p Value
0.0001

14.318 23.592

0.0004 0.0001

___________________________________Means Table by Airline _________________________________________________________________________ Count Rating 1, Delta Rating 1, Lufthansa Rating 1, KLM Rating 2, Delta Rating 2, Lufthansa Rating 2, KLM 20 20 20 20 20 20 Mean 38.950 58.900 72.900 32.400 72.250 79.800 Std. Dev. 14.006 15.089 13.902 8.268 10.572 11.265 Std. Error 3.132 3.374 3.108 1.849 2.364 2.519

______________________________________Means Table Effect: Ratings_________________________________________________________________ Count Mean Std. Dev. Std. Error

Rating 1
Rating 2

60
60

56.917
61.483

19.902
23.208

2.569
2.996

All data are hypothetical.