Chapter Ten Introduction to Hypothesis Testing
Dec 24, 2015
Chapter Ten
Introduction to Hypothesis
Testing
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 2
New Statistical Notation
• The symbol for greater than is >.
• The symbol for less than is <.
• The symbol for greater than or equal tois ≥.
• The symbol for less than or equal tois ≤.
• The symbol for not equal to is ≠.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 3
The Role of InferentialStatistics in Research
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 4
Sampling Error
Remember:
Sampling error results when random
chance produces a sample statistic that
does not equal the population parameter
it represents.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 5
Parametric Statistics
• Parametric statistics are procedures that require certain assumptions about the characteristics of the populations being represented. Two assumptions are common to all parametric procedures:– The population of dependent scores forms a
normal distribution
and– The scores are interval or ratio.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 6
Nonparametric Procedures
• Nonparametric statistics are
inferential procedures that do not
require stringent assumptions about the
populations being represented.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 7
Robust Procedures
Parametric procedures are robust. If the
data don’t meet the assumptions of the
procedure perfectly, we will have only a
negligible amount of error in the
inferences we draw.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 8
Setting up Inferential Procedures
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 9
Experimental Hypotheses
Experimental hypotheses describe the
predicted outcome we may or may not
find in an experiment.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 10
Predicting a Relationship
• A two-tailed test is used when we predict that there is a relationship, but do not predict the direction in which scores will change.
• A one-tailed test is used when we predict the direction in which scores will change.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 11
Designing a One-Sample Experiment
To perform a one-sample experiment, we
must already know the population mean
under some other condition of the
independent variable.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 12
Alternative Hypothesis
The alternative hypothesis describes
the population parameters that the
sample data represent if the predicted
relationship exists.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 13
Null Hypothesis
The null hypothesis describes the
population parameters that the sample
data represent if the predicted relationship
does not exist.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 14
A Graph Showing the Existence of a Relationship
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 15
A Graph Showing That a Relationship Does Not Exist
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 16
Performing the z-Test
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 17
The z-Test
The z-test is the procedure for computing
a z-score for a sample mean on the
sampling distribution of means.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 18
)( X
Assumptions of the z-Test
1. We have randomly selected one sample
2. The dependent variable is at least approximately normally distributed in the population and involves an interval or ratio scale
3. We know the mean of the population of raw scores under some other condition of the independent variable
4. We know the true standard deviation of the population described by the null hypothesis
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 19
Setting up for a Two-Tailed Test
1.Choose alpha. Common values are 0.05 and 0.01.
2.Locate the region of rejection. For a two-tailed test, this will involve defining an area in both tails of the sampling distribution.
3.Determine the critical value. Using the chosen alpha, find the zcrit value that gives the appropriate region of rejection.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 20
A Sampling Distribution for H0 Showing the Region of Rejection for = 0.05 in
a Two-tailed Test
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 21
• In a two-tailed test, the null hypothesis states that the population mean equals a given value. For example, H0: = 100.
• In a two-tailed test, the alternative hypothesis states that the population mean does not equal the same given value as in the null hypothesis. For example, Ha: 100.
Two-Tailed Hypotheses
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 22
X
Xz
obt
NX
X
Computing z
• The z-score is computed using the same formula as before
where
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 23
Rejecting H0
• When the zobt falls beyond the critical value, the statistic lies in the region of rejection, so we reject H0 and accept Ha
• When we reject H0 and accept Ha we say the results are significant. Significant indicates that the results are too unlikely to occur if the predicted relationship does not exist in the population.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 24
Interpreting Significant Results
• When we reject H0 and accept Ha, we do not prove that H0 is false
• While it is unlikely for a mean that lies within the rejection region to occur, the sampling distribution shows that such means do occur once in a while
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 25
Failing to Reject H0
• When the zobt does not fall beyond the critical value, the statistic does not lie within the region of rejection, so we do not reject H0
• When we fail to reject H0 we say the results are nonsignificant. Nonsignificant indicates that the results are likely to occur if the predicted relationship does not exist in the population.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 26
Interpreting Nonsignificant Results
• When we fail to reject H0, we do not prove that H0 is true
• Nonsignificant results provide no convincing evidence—one way or the other—as to whether a relationship exists in nature
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 27
1.Determine the experimental hypotheses and create the statistical hypothesis
2.Compute and compute zobt
3.Set up the sampling distribution
4.Compare zobt to zcrit
Summary of the z-Test
X
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 28
The One-Tailed Test
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 29
One-Tailed Hypotheses
• In a one-tailed test, if it is hypothesized that the independent variable causes an increase in scores, then the null hypothesis is that the population mean is less than or equal to a given value and the alternative hypothesis is that the population mean is greater than the same value. For example:– H0: ≤ 50
– Ha: > 50
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 30
A Sampling Distribution Showing the Region of Rejection for a One-tailed Test of Whether
Scores Increase
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 31
One-Tailed Hypotheses
• In a one-tailed test, if it is hypothesized that the independent variable causes a decrease in scores, then the null hypothesis is that the population mean is greater than or equal to a given value and the alternative hypothesis is that the population mean is less than the same value. For example:– H0: ≥ 50
– Ha: < 50
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 32
A Sampling Distribution Showing the Region of Rejection for a One-tailed Test of Whether Scores
Decrease
[Insert Figure 10.8 here]
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 33
Choosing One-Tailed Versus Two-Tailed Tests
Use a one-tailed test only when confident
of the direction in which the dependent
variable scores will change. When in
doubt, use a two-tailed test.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 34
Errors in StatisticalDecision Making
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 35
Type I Errors
• A Type I error is defined as rejecting H0 when H0 is true
• In a Type I error, there is so much sampling error that we conclude that the predicted relationship exists when it really does not
• The theoretical probability of a Type I error equals
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 36
Type II Errors
• A Type II error is defined as retaining H0 when H0 is false (and Ha is true)
• In a Type II error, the sample mean is so close to the described by H0 that we conclude that the predicted relationship does not exist when it really does
• The probability of a Type II error is
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 37
Power
• The goal of research is to reject H0 when H0 is false
• The probability of rejecting H0 when it is false is called power
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 38
Possible Results of Rejecting or Retaining H0
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 39
14 14 13 15 11 15
13 10 12 13 14 13
14 15 17 14 14 15
Example
• Use the following data set and conduct a two-tailed z-test to determine if = 11 if the population standard deviation is known to be 4.1
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 40
764.2966.0
1167.13obt
X
Xz
966.018
1.4
NX
X
Example
1.H0: = 11; Ha: ≠ 11
2.Choose = 0.05
3.Reject H0 if zobt > +1.965 or if zobt < -1.965.
Copyright © Houghton Mifflin Company. All rights reserved. Chapter 10 - 41
Example
Since zobt lies within the rejection region,
we reject H0 and accept Ha. Therefore, we
conclude that ≠ 11.