Test of Difference (t-test)

Test of Difference
 one sample z-test
 one sample t-test

 Independent samples t-test

 Dependent samples t-test
 ANOVA (one-way and two-way)
 Repeated Measures ANOVA
 Two Way Mixed-Design ANOVA
Test of Difference
I. Independent samples t-test

 applying null hypothesis testing when you have two

samples (treated differently) Grouping Variable

 two samples must be completely:

○ Separate;
*individuals in one sample are not in any
○ Independent; way connected or related to individuals in
the other sample
○ Random
Test of Difference
 Between-participants

 Unrelated design

 Unmatched t-test

 Unpaired t-test
Test of Difference
 Assumptions of the t Test for Two
Independent Samples

 Random Sampling
 Normal Distribution
 Measured on an interval or ratio scale
 Homogeneity of Variance (HOV)
Test of Difference
 Homogeneity of Variance (HOV)

 “same variance” (for pooled-variance t-test)

○ both samples are quite large

○ two samples are the same size
○ two sample variances are very similar
*one sample variance is no more than twice as large as
the other, you can safely assume HOV in the population.
Test of Difference
 Note:
○ If your samples are:

 not very large,

 not equal in size,
 and one sample variance is more than twice
the other,
Heterogeneity of Variance
*consider the possibility of performing a
separate-variances t test.
Test of Difference
 Alternative method to determine the use pooled
and separate variance t-test?

 Levene’s Test provided by SPSS

Sig. > .05

Equal variances assumed
= pooled-variance t-test

Sig. < .05

Equal variances not assumed
= separate-variance t-test
Test of Difference
 Analysis by SPSS

 Twenty-four people were involved in an experiment to

determine whether background noise (music, slamming
of doors, people making coffee, etc.) affects short-term
memory (recall of words). Half (12) of the sample were
randomly allocated to the NOISE condition, and half
(12) to the NO NOISE condition.
The participants in the NOISE condition tried to memorize
a list of 20 words in two minutes, while listening to pre-
recorded noise through earphones. The other participants
wore earphones but heard no noise as they attempted to
memorize the words. Immediately after this, they were tested
to see how many words they recalled.
Test of Difference
 Analysis by SPSS
Test of Difference
 Analysis by SPSS
Test of Difference
 Analysis by SPSS
Test of Difference
 If the assumption of:
○ random sampling,
○ Normal Distribution,
○ interval or ratio level,
○ and HOV
are not met,

*Some alternatives can be performed

Test of Difference
1. Resampling Procedure (random sampling)
Bootstrapping Method

2. Data Transformation (Normal Distribution)

Removal of Outliers

3. Nonparametric Tests (interval/ratio level)

o Mann Whitney U Test (for ordinal)
o Chi-Square (for nominal)

4. Separate-Variance t-test
Test of Difference
II. Dependent samples t-test

• Related t-test
• Paired t-test
• Matched t-test
• Whitin-participants
Test of Difference
 Dependent samples t-test

 used when the same participants perform

under both conditions

ex. All 24 participants recall words under

the NOISE and also, under NO NOISE
condition.
Test of Difference
 Dependent samples t-test

 Commonly uses a before and after design

(pre-test and post-test).

Ex. A group of students was administered a test

anxiety before taking a Psychological Statistics
FINAL EXAM and another after they took the
exam. The difference on their test anxiety
before and after the exam was computed.
Test of Difference

 Computation is almost the same to that

of independent samples t-test.

 However, it is more sensitive because it

uses data from the same sample.
Test of Difference
 Analysis by SPSS
Test of Difference
 Analysis by SPSS
Test of Difference
 Analysis by SPSS

 A study was carried out comparing the verbal IQ

(VIQ) and the performance IQ (PIQ) of people
with chronic illness. In a normal population you
would expect the two IQ measures to be similar.
The population mean IQ is 100.
Test of Difference
 Analysis by SPSS
Test of Difference
 Analysis by SPSS
Test of Difference
 Nonparametric alternatives for
Dependent Samples t-test

 Wilcoxon T test for related samples

○ For Ordinal data

 McNemar Test
○ For Nominal Data
Test of Difference

ACTIVITY
Identify what statistical test (Independent Samples
t-test or Dependent Samples t-test) is appropriate to
perform from the following situations.
Test of Difference
1. A neuropsychologist believes that right-handed
people will recognize objects placed in their
right hands more quickly than objects placed in
their left hands when they are blindfolded. The
following scores represent how many objects
each participant could identify in 2 minutes with
each hand.
Test of Difference

2. What statistical test should be used to

determine whether there is a significant
tendency for verbal GRE scores to improve on
the second testing?
Test of Difference
3. A developmental psychologist would like to
compare attitudes toward death in which
adolescents were compared with adults.
The Death Anxiety Scale, where high scores
indicate high anxiety, was administered to
both groups.
Test of Difference

4. The stress levels of 30 unemployed

laborers were measured by a
questionnaire before and after a real job
interview.
Test of Difference

5. A comparative study on the attitudes

and uses of music by adults with visual
impairments and those who are
sighted.
Test of Difference
6. Some people spend a lot of time actively
engaged in problem solving. Others do not.
There are large individual differences in
people’s tendency to engage in and enjoy
this ‘cognitive activity’. This individual
difference dimension is called need for
cognition (NEEDCOG). In one study, men
and women were compared on this
dimension.
Test of Difference
7. A cognitive psychologist is testing the theory that
short-term memory is mediated by subvocal
rehearsal. This theory can be tested by reading
aloud a string of letters to a participant, who must
repeat the string correctly after a brief delay. If the
theory is correct, there will be more errors when the
list contains letters that sound alike (e.g., G and T)
than when the list contains letters that look alike
(e.g., P and R). Each participant gets both types of
letter strings, which are randomly mixed in the same
experimental session. The number of errors for each
type of letter string for each participant are recorded.
Test of Difference
8. A teacher wanted to test if there is a
significant difference between his men
and women students in their scores on
their Psychological Statistics Exam
scores.
Test of Difference
9. The data below represents the no. of students
who attended the classes of Graduate Student
Instructor and Full Professor. What statistical test
is more appropriate to use to conduct a
difference test?

Graduate Student Instructor:

9, 11,14,14,16, 19, 37

Full Professor:
13, 15, 15, 16, 18, 23, 28, 31, 31
Test of Difference
10.Twenty people agree to follow a rigorous
regimen of exercise for 2 weeks. They agree
to be weighed both before and after the 2-
week “treatment.” The data for this
experiment are shown.
LET’S CHECK !
Test of Difference
1. A neuropsychologist believes that right-handed
people will recognize objects placed in their
right hands more quickly than objects placed in
their left hands when they are blindfolded. The
following scores represent how many objects
each participant could identify in 2 minutes with
each hand.
Dependent Samples t-test
Test of Difference

2. What statistical test should be used to

determine whether there is a significant
tendency for verbal GRE scores to improve on
the second testing?
Dependent Samples t-test
Test of Difference
3. A developmental psychologist would like to
compare attitudes toward death in which
adolescents were compared with adults.
The Death Anxiety Scale, where high scores
indicate high anxiety, was administered to
both groups.
Independent Samples t-test
Test of Difference

4. The stress levels of 30 unemployed

laborers were measured by a
questionnaire before and after a real job
interview.
Dependent Samples t-test
Test of Difference

5. A comparative study on the attitudes

and uses of music by adults with visual
impairments and those who are
sighted.
Independent Samples t-test
Test of Difference
6. Some people spend a lot of time actively
engaged in problem solving. Others do not.
There are large individual differences in
people’s tendency to engage in and enjoy
this ‘cognitive activity’. This individual
difference dimension is called need for
cognition (NEEDCOG). In one study, men
and women were compared on this
dimension.
Independent Samples t-test
Test of Difference
7. A cognitive psychologist is testing the theory that
short-term memory is mediated by subvocal
rehearsal. This theory can be tested by reading
aloud a string of letters to a participant, who must
repeat the string correctly after a brief delay. If the
theory is correct, there will be more errors when the
list contains letters that sound alike (e.g., G and T)
than when the list contains letters that look alike
(e.g., P and R). Each participant gets both types of
letter strings, which are randomly mixed in the same
experimental session. The number of errors for each
type of letter string for each participant are recorded.
Dependent Samples t-test
Test of Difference
8. A teacher wanted to test if there is a
significant difference between his men
and women students in their scores on
their Psychological Statistics Exam
scores.
Independent Samples t-test
Test of Difference
9. The data below represents the no. of students
who attended the classes of Graduate Student
Instructor and Full Professor. What statistical test
is more appropriate to use to conduct a
difference test?

Graduate Student Instructor:

9, 11,14,14,16, 19, 37

Full Professor:
13, 15, 15, 16, 18, 23, 28, 31, 31
Independent Samples t-test
Test of Difference
10.Twenty people agree to follow a rigorous
regimen of exercise for 2 weeks. They agree
to be weighed both before and after the 2-
week “treatment.” The data for this
experiment are shown.
Dependent Samples t-test