Transcript Slide 17

Chapter 17
Business Research
Methods
Donald Cooper
Pamela Schindler
Irwin/McGraw-Hill
©The McGraw-Hill Companies, Inc., 2001
Chapter 17
Hypothesis Testing
Irwin/McGraw-Hill
©The McGraw-Hill Companies, Inc., 2001
Approaches to
Click to edit Master title style
Hypothesis Testing
Slide 17 - 1
Classical Statistics
sampling-theory approach
objective view of probability
decision making rests on analysis of available
sampling data
Bayesian Statistics
extension of classical statistics
consider all other available information
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
ClickTypes
to editof
Master
title style
Hypotheses
Slide 17 - 2
Null
that no statistically significant difference exists
between the parameter and the statistic being
compared
Alternative
logical opposite of the null hypothesis
that a statistically significant difference does
exist between the parameter and the statistic
being compared.
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to of
editHypothesis
Master titleTesting
style
Logic
Slide 17 - 3
Two tailed test
nondirectional test
considers two possibilities
One tailed test
directional test
places entire probability of an unlikely
outcome to the tail specified by the alternative
hypothesis
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to edit Errors
Master in
title
style
Decision
Testing
Slide 17 - 4
Type I error
a true null hypothesis is rejected
Type II error
one fails to reject a false null hypothesis
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Testing for Statistical
Click to edit Master title style
Significance
Slide 17 - 5
State the null hypothesis
Choose the statistical test
Select the desired level of significance
Compute the calculated difference
value
Obtain the critical value
Interpret the test
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to edit
Master title style
Classes
of Significance
Tests
Slide 17 - 6
Parametric tests
Z or t test is used to determine the statistical
significance between a sample distribution
mean and a population parameter
Assumptions:
 independent observations
 normal distributions
 populations have equal variances
 at least interval data measurement scale
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to edit
Master title style
Classes
of Significance
Tests
Slide 17 - 7
Nonparametric tests
Chi-square test is used for situations in which
a test for differences between samples is
required
Assumptions
 independent observations for some tests only
 normal distribution not necessary
 homogeneity of variance not necessary
 appropriate for nominal and ordinal data, may be used
for interval or ratio data
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
ClicktotoTest
edit the
Master
style
How
Nulltitle
Hypothesis
Slide 17 - 8
Analysis of variance (ANOVA)
the statistical method for testing the null
hypothesis that means of several
populations are equal
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to editComparison
Master title style
Multiple
Tests
Slide 17 - 9
Multiple comparison procedures
test the difference between each pair of means
and indicate significantly different group
means at a specified alpha level (<.05)
use group means and incorporate the MSerror
term of the F ratio
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
ClickHow
to edittoMaster
style
Selecttitle
a Test
Slide 17 - 10
Which does the test involve?
one sample,
two samples
k samples
If two or k samples,are the individual cases
independent or related?
Is the measurement scale nominal, ordinal,
interval, or ratio?
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001
Click
to edit Master
titleTest
style
K Related
Samples
Slide 17 - 11
Use when:
The grouping factor has more than two
levels
Observations or subjects are
matched . . . or
the same subject is measured more than
once
Interval or ratio data
Irwin/McGraw-Hill
The McGraw-Hill Companies, Inc., 2001