Psy 2020 Experimental I: Applied Statistics

Lab 4: t-tests
Dr. Obrecht Lindemann
Welcome! Here you will work through hypothesis testing examples using t-statistics.
You will need to open the Lab 4 t-tests.sav (SPSS datafile) in order to complete this lab.
Please type your answers in this document and submit this file in Blackboard by the end
the week.

Overview
The datafile contains data from two different studies. Each study’s data should be analyzed
using one of the t-tests. First you‘ll need to determine which t-test to use, and then analyze the
data and explain what you found. Please see my lab tutorial for examples of how to do this.
Also, I provide steps for how to run the different types of t-tests at the end of the document (see
SPSS guide for t-tests)

Study descriptions
Dataset 1: College students stated whether they were male or female, and reported the number of
speeding tickets they received in the past year. (Hint: students are in either the male group or the
female group; this tells you whether the variable sex is between or within subjects)

Dataset 2: Participants with a Bachelor’s degree reported their annual income. The mean income
of adults in the United States overall is μ = $41,027.

Dataset 1: Sex and speeding tickets

1) Which t-test should be used to analyze whether sex predicts number of speeding
tickets? (single sample, repeated measures/within subjects, or independent
samples/between subjects)

Independent sample t-test

2) State the independent variable (IV).

Gender or sex i.e. male or female

3) State the dependent variable (DV).

The number of speeding tickets received

4) State the Null hypothesis (H0):

There is no significant difference in the mean number of speeding tickets between males
and females.
5) Run your t-test (see the SPSS guide at the bottom of the document for instructions).
How many speeding tickets did males and females get on average?
a. X males = 1.29
b. X females = 0.39
 6) Next, type your results in APA format (t(df)=x.xx, p=.xx, d=.xx):
(Note, use the “Two-sided p” for your p-value in the Equal variances assumed row)
t(df) = 48
p= 0.006
d= 0.897

7) Explain what your results mean. If you reject the null, be sure to include the
direction of the result, i.e., explain which sex generally speeds more. Mention
whether you have a small, medium, or large effect size (d ≈ .2 or -.2  small; d ≈ .5
or -.5  medium; d ≈ .8 or -.8  large).

Based on the results of the independent samples t-test, we reject the null hypothesis,
indicating a statistically significant difference in the mean number of speeding tickets
between males and females. Specifically, the p-value of 0.006 indicates that the observed
difference is unlikely to have occurred by chance alone. Moreover, the effect size, as
indicated by d of 0.897, suggests a large difference between the two groups. This means
that, on average, males tend to receive significantly more speeding tickets than females.
8) Finally, use Excel to make a graph of your two group means (See instructions at the
bottom of this document, and/or my lab tutorial).

Mean of Gender
1.4 1.29
1.2

0.8
Mean

0.6
0.39
0.4

0.2

0
Males Females

Gender

Dataset 2: Income with a Bachelor’s degree

1) Which t-test should be used to analyze whether people with a Bachelor’s degree
earn a different amount of money compared to the general population (μ =
 $41,027)? (single sample, repeated measures/within subjects, or independent
samples/between subjects)

Single sample t-test

2) State the independent variable (IV).

Education Level
3) State the dependent variable (DV).

Income
4) State the Null hypothesis (H0):

There is no difference in the mean annual income between individuals with a Bachelor's
degree and the general population i.e. μ = $41,027
5) Run your t-test. Run your t-test (see the SPSS guide at the bottom of the document
for instructions). How much money does a person with a Bachelor’s degree make
on average?
a. X income with Bachelor ' s = $48,225.93

6) Next, type your results in APA format (t(df)=x.xx, p=.xx, d=.xx):

(Note, use the “Two-sided p” for your p-value in the Equal variances assumed row)
T(df) = 58, p = 0.02, d = 0.413

7) Explain what your results mean. If you reject the null, be sure to include the
direction of the result. Mention whether you have a small, medium, or large effect
size (d ≈ .2 or -.2  small; d ≈ .5 or -.5  medium; d ≈ .8 or -.8  large).

Since the p-value (0.02) is less than the significance level (usually α = 0.05), we reject the
null hypothesis. This suggests that there is a statistically significant difference in the
mean annual income between individuals with a Bachelor's degree and the general
population. The effect size (d = 0.413) is moderate or medium according to common
conventions. This suggests that although there is a statistically significant difference, the
practical significance may be relatively modest. Therefore, based on these results, we can
conclude that individuals with a Bachelor's degree have a statistically significant
difference in mean annual income compared to the general population, with a small to
moderate effect size.

SPSS guide for t-tests

Single sample t-test:
-AnalyzeCompare Means and ProportionsOne-Sample t-test
-DV goes in the Test Variable box
-Type μ into the Test Value box and hit OK.
Repeated measures/paired-sample/within Ss t-test:
-Data from a given S are given across a single row
-AnalyzeCompare Means and Proportions Paired Samples t-test
-Select and click the arrow button to move both levels of the IV over into the paired
variables box; hit OK

Independent samples/between Ss t-test:

-Levels of the IV are coded in a single column
-AnalyzeCompare Means and Proportions Independent Samples t-test
-The IV (the nominal variable) is the “grouping variable”; after putting it in you must
define the groups (enter in the two values that correspond to the two groups listed in the
data sheet, e.g. 1 and 2)
-Make sure Estimate effect sizes is checked and hit OK

Graph Guide in Excel

Open Excel. Type in your group means and label them; see example:
Men Women
3.5 3.2

(Note: you probably won’t type in Men and Women, but use the labels that make sense for your
variable.)
Highlight all four cells (the two means with their labels), and click the Insert tab. Click on the
picture of the bar graph (2D column, clustered column). Then customize your graph (update
your chart title, remove background lines, add X and Y axes labels). When you click on your
graph you’ll see a + sign; this lets you add axis labels and more.