Academic - Udayton.edu-Using SPSS For T-Tests
Academic - Udayton.edu-Using SPSS For T-Tests
edu
http://academic.udayton.edu/gregelvers/psy216/spss/ttests.htm
This tutorial will show you how to use SPSS version 12.0 to perform one-sample t-tests, independent samples ttests, and paired samples t-tests.
This tutorial assumes that you have:
Downloaded the standard class data set (click on the link and save the data file)
Started SPSS (click on Start | Programs | SPSS for Windows | SPSS 12.0 for Windows)
One Sample t-Tests
One sample t-tests can be used to determine if the mean of a sample is different from a particular value. In this
example, we will determine if the mean number of older siblings that the PSY 216 students have is greater than 1.
We will follow our customary steps:
1. Write the null and alternative hypotheses first:
H0: 216 Students 1
H1: 216 Students > 1
Where is the mean number of older siblings that the PSY 216 students have.
2. Determine if this is a one-tailed or a two-tailed test. Because the hypothesis involves the phrase "greater
than", this must be a one tailed test.
3. Specify the level: = .05
4. Determine the appropriate statistical test. The variable of interest, older, is on a ratio scale, so a z-score
test or a t-test might be appropriate. Because the population standard deviation is not known, the z-test
would be inappropriate. We will use the t-test instead.
5. Calculate the t value, or let SPSS do it for you!
The command for a one sample t tests is found at Analyze | Compare Means | One-Sample T Test (this is
shorthand for clicking on the Analyze menu item at the top of the window, and then clicking on Compare
Means from the drop down menu, and One-Sample T Test from the pop up menu.):
This output tells us that we have 46 observations (N), the mean number of older siblings is 1.26 and the
standard deviation of the number of older siblings is 1.255. The standard error of the mean (the standard
deviation of the sampling distribution of means) is 0.185 (1.255 / square root of 46 = 0.185).
The second part of the output gives the value of the statistical test:
The second column of the output gives us the t-test value: (1.26 - 1) / (1.255 / square root of 46) = 1.410 [if
you do the calculation, the values will not match exactly because of round-off error). The third column tells
us that this t test has 45 degrees of freedom (46 - 1 = 45). The fourth column tells us the two-tailed
significance (the 2-tailed p value.) But we didn't want a two-tailed test; our hypothesis is one tailed and
there is no option to specify a one-tailed test. Because this is a one-tailed test, look in a table of critical t
values to determine the critical t. The critical t with 45 degrees of freedom, = .05 and one-tailed is 1.679.
6. Determine if we can reject the null hypothesis or not. The decision rule is: if the one-tailed critical t value is
less than the observed t AND the means are in the right order, then we can reject H0. In this example, the
critical t is 1.679 (from the table of critical t values) and the observed t is 1.410, so we fail to reject H0. That
is, there is insufficient evidence to conclude that the mean number of older siblings for the PSY 216 classes
is larger than 1.
If we were writing this for publication in an APA journal, we would write it as:
A t test failed to reveal a statistically reliable difference between the mean number of older siblings that the PSY
216 class has (M = 1.26, s = 1.26) and 1, t(45) = 1.410, p < .05, = .05.
Independent Samples t-Tests
This gives the descriptive statistics for each of the two groups (as defined by the grouping variable.) In this
example, there are 14 people in the 10 AM section (N), and they have, on average, 0.86 older siblings, with
a standard deviation of 1.027 older siblings. There are 32 people in the 11 AM section (N), and they have,
on average, 1.44 older siblings, with a standard deviation of 1.318 older siblings. The last column gives the
standard error of the mean for each of the two groups.
The second part of the output gives the inferential statistics:
The columns labeled "Levene's Test for Equality of Variances" tell us whether an assumption of the t-test
has been met. The t-test assumes that the variability of each group is approximately equal. If that
assumption isn't met, then a special form of the t-test should be used. Look at the column labeled "Sig."
under the heading "Levene's Test for Equality of Variances". In this example, the significance (p value) of
Levene's test is .203. If this value is less than or equal to your level for the test (usually .05), then you can
reject the null hypothesis that the variability of the two groups is equal, implying that the variances are
unequal. If the p value is less than or equal to the level, then you should use the bottom row of the output
(the row labeled "Equal variances not assumed.") If the p value is greater than your level, then you
should use the middle row of the output (the row labeled "Equal variances assumed.") In this example,
.203 is larger than , so we will assume that the variances are equal and we will use the middle row of the
output.
The column labeled "t" gives the observed or calculate t value. In this example, assuming equal variances,
the t value is 1.461. (We can ignore the sign of t for a two tailed t-test.) The column labeled "df" gives the
degrees of freedom associated with the t test. In this example, there are 44 degrees of freedom.
The column labeled "Sig. (2-tailed)" gives the two-tailed p value associated with the test. In this example,
the p value is .151. If this had been a one-tailed test, we would need to look up the critical t in a table.
6. Decide if we can reject H 0: As before, the decision rule is given by: If p , then reject H 0. In this
example, .151 is not less than or equal to .05, so we fail to reject H0. That implies that we failed to observe
a difference in the number of older siblings between the two sections of this class.
If we were writing this for publication in an APA journal, we would write it as:
A t test failed to reveal a statistically reliable difference between the mean number of older siblings that the 10 AM
section has (M = 0.86, s = 1.027) and that the 11 AM section has (M = 1.44, s = 1.318), t(44) = 1.461, p = .151,
= .05.
Independent Samples t-Tests
Cut Point Groups
Sometimes you want to perform a t-test but the groups are defined by a variable that is not dichotomous (i.e., it
has more than two values.) For example, you may want to see if the number of older siblings is different for
students who have higher GPAs than for students who have lower GPAs. Since there is no single value of GPA
that specifies "higher" or "lower", we cannot proceed exactly as we did before. Before proceeding, decide which
value you will use to divide the GPAs into the higher and lower groups. The median would be a good value, since
half of the scores are above the median and half are below. (If you do not remember how to calculate the median
see the frequency command in the descriptive statistics tutorial.)
1. Write the null and alternative hypotheses first:
H0: lower GPA = higher GPA
H1: lower GPA Higher GPA
Where is the mean number of older siblings that the PSY 216 students have.
2. Determine if this is a one-tailed or a two-tailed test. Because the hypothesis involves the phrase "different"
and no ordering of the means is specified, this must be a two tailed test.
3. Specify the level: = .05
4. Determine the appropriate statistical test. The variable of interest, older, is on a ratio scale, so a z-score
test or a t-test might be appropriate. Because the population standard deviation is not known, the z-test
would be inappropriate. Furthermore, different students have higher and lower GPAs, so we have a
between-subjects design. Because of these factors, we will use the independent samples t-test.
5. Calculate the t value, or let SPSS do it for you.
The command for the independent samples t tests is found at Analyze | Compare Means | IndependentSamples T Test (this is shorthand for clicking on the Analyze menu item at the top of the window, and then
clicking on Compare Means from the drop down menu, and Independent-Samples T Test from the pop up
menu.):
Click in the circle to the left of "Cut Point:". Then type the value that splits the variable into two groups.
Group one is defined as all scores
that are greater than or equal to the
cut point. Group two is defined as
all scores that are less than the cut
point. In this example, use 3.007
(the median of the GPA variable) as
the cut point value:
This gives the descriptive statistics for each of the two groups (as defined by the grouping variable.) In this
example, there are 23 people with a GPA greater than or equal to 3.01 (N), and they have, on average,
1.04 older siblings, with a standard deviation of 1.186 older siblings. There are 23 people with a GPA less
than 3.01 (N), and they have, on average, 1.48 older siblings, with a standard deviation of 1.310 older
siblings. The last column gives the standard error of the mean for each of the two groups.
The second part of the output gives the inferential statistics:
As before, the columns labeled "Levene's Test for Equality of Variances" tell us whether an assumption of
the t-test has been met. Look at the column labeled "Sig." under the heading "Levene's Test for Equality of
Variances". In this example, the significance (p value) of Levene's test is .383. If this value is less than or
equal to your level for this test, then you can reject the null hypothesis that the variabilities of the two
groups are equal, implying that the variances are unequal. In this example, .383 is larger than our level
of .05, so we will assume that the variances are equal and we will use the middle row of the output.
The column labeled "t" gives the observed or calculated t value. In this example, assuming equal
variances, the t value is 1.180. (We can ignore the sign of t when using a two-tailed t-test.) The column
labeled "df" gives the degrees of freedom associated with the t test. In this example, there are 44 degrees
of freedom.
The column labeled "Sig. (2-tailed)" gives the two-tailed p value associated with the test. In this example,
the p value is .244. If this had been a one-tailed test, we would need to look up the critical t in a table.
6. Decide if we can reject H 0: As before, the decision rule is given by: If p , then reject H0. In this
example, .244 is greater than .05, so we fail to reject H0. That implies that there is not sufficient evidence
to conclude that people with higher or lower GPAs have different number of older siblings.
If we were writing this for publication in an APA journal, we would write it as:
An equal variances t test failed to reveal a statistically reliable difference between the mean number of older
siblings for people with higher (M = 1.04, s = 1.186) and lower GPAs (M = 1.48, s = 1.310), t(44) = 1.18, p = .244,
= .05.
Paired Samples t-Tests
When two samples are involved and the values for each sample are collected from the same individuals (that is,
each individual gives us two values, one for each of the two groups), or the samples come from matched pairs of
individuals then a paired-samples t-test may be an appropriate statistic to use.
The paired samples t-test can be used to determine if two means are different from each other when the two
samples that the means are based on were taken from the matched individuals or the same individuals. In this
example, we will determine if the students have different numbers of younger and older siblings.
1. Write the null and alternative hypotheses:
H0: older = younger
H1: older younger
Where is the mean number of siblings that the PSY 216 students have.
2. Determine if this is a one-tailed or a two-tailed test. Because the hypothesis involves the phrase "different"
and no ordering of the means is specified, this must be a two tailed test.
3. Specify the level: = .05
4. Determine the appropriate statistical test. The variables of interest, older and younger, are on a ratio scale,
so a z-score test or a t-test might be appropriate. Because the population standard deviation is not known,
the z-test would be inappropriate. Furthermore, the same students are reporting the number of older and
younger siblings, we have a within-subjects design. Because of these factors, we will use the paired
samples t-test.
5. Let SPSS calculate the value of t for you.
The command for the paired samples t tests is found at Analyze | Compare Means | Paired-Samples T Test
(this is shorthand for clicking on the Analyze menu item at the top of the window, and then clicking on
Compare Means from the drop down menu, and Paired-Samples T Test from the pop up menu.):
You must select a pair of variables that represent the two conditions. Click on one of the variables in the left
hand pane of the Paired-Samples t Test dialog box. Then click on the other variable in the left hand pane.
Click on the arrow button to move the variables into the Paired Variables pane. In this example, select
Older and Younger variables (number of older and younger siblings) and then click on the arrow button to
move the pair into the Paired Variables box:
Click on the OK button in the Paired-Samples t Test dialog box to perform the t-test. The output viewer will
appear with the results of the t test. The results have three main parts: descriptive statistics, the correlation
between the pair of variables, and inferential statistics. First, the descriptive statistics:
This gives the descriptive statistics for each of the two groups (as defined by the pair of variables.) In this
example, there are 45 people who responded to the Older siblings question (N), and they have, on
average, 1.24 older siblings, with a standard deviation of 1.26 older siblings. These same 45 people also
responded to the Younger siblings question (N), and they have, on average, 1.13 younger siblings, with a
standard deviation of 1.20 younger siblings. The last column gives the standard error of the mean for each
of the two variables.
The second part of the output gives the correlation between the pair of variables:
The column labeled "Mean" is the difference of the two means (1.24 - 1.13 = 0.11 in this example (the
difference is due to round off error).) The next column is the standard deviation of the difference between
the two variables (1.98 in this example.)
The column labeled "t" gives the observed or calculated t value. In this example, the t value is 0.377 (you
can ignore the sign.) The column labeled "df" gives the degrees of freedom associated with the t test. In
this example, there are 44 degrees of freedom. The column labeled "Sig. (2-tailed)" gives the two-tailed p
value associated with the test. In this example, the p value is .708. If this had been a one-tailed test, we
would need to look up the critical value of t in a table.
6. Decide if we can reject H 0: As before, the decision rule is given by: If p , then reject H0. In this example,
.708 is not less than or equal to .05, so we fail to reject H0. That implies that there is insufficient evidence
to conclude that the number of older and younger siblings is different.
If we were writing this for publication in an APA journal, we would write it as:
A paired samples t test failed to reveal a statistically reliable difference between the mean number of older (M =
1.24, s = 1.26) and younger (M = 1.13, s = 1.20) siblings that the students have, t(44) = 0.377, p = .708, = .05.