PSYC 355
SPSS HOMEWORK: CORRELATION ASSIGNMENT INSTRUCTIONS
OVERVIEW
This assignment is designed to increase your statistical literacy and proficiency in conducting
and interpreting the Pearson correlation coefficient. You will be completing two Pearson
correlation coefficient analyses in SPSS, using data that are related to specific research scenarios
in the behavioral sciences, such as psychology, social work, and counseling. You will also be
completing one Cumulative Knowledge Question that will use an analysis learned in a previous
week. Behind the scenes knowledge of how this test is conducted is fundamental to being able to
understand and apply research in your related field to your practice. Additionally, SPSS skills are
professionally valuable, as it is one of the most commonly used statistical software packages in
behavioral science settings, both academic and professional.
INSTRUCTIONS
This assignment includes three problem sets that contain research scenarios and related
questions.
For each scenario, you will run an analysis in SPSS. The required product will include
the SPSS output, an APA-style Results section describing the results, and the appropriate
graph inserted as a Figure in APA style.
For each scenario, you will create a new SPSS data file with the data from the problem
scenario. ALL variable names in your SPSS data files must include your initials, so that
they show up in your SPSS output and graph. For example, for someone with the initials
ABC creating a variable of memory scores, the variable could be named
“ABC_mem_scores” or “Mem_Scores_ABC”, etc.
For all problems, interpret results based on an alpha level of a = .05.
Please review the Watch: SPSS Homework Tutorial: The Pearson Correlation in this module
for directions on how to run the statistical test, as well as the Watch: Results Sections for the
Pearson Correlation Coefficient Analysis in APA Style, which includes a template for
completing the APA-style Results sections for the Pearson correlation coefficient. The scenarios
begin on the next page.
PSYC 355
Fake Positive
Reviews
Total Number
of Returns
67
19
23
78
12
52
26
30
10
87
41
40
5
10
2
21
3
9
8
8
1
11
11
6
1. Paste SPSS output. (10 pts)
Correlations
LR_Fake_Posi
tive_Reviews
LR_Total_Nu
mber_Returns
LR_Fake_Positive_Revie
ws
Pearson
Correlation
1 .649*
Sig. (2-tailed) .022
Problem Set 1: Pearson Correlation Coefficient Analysis CHANGE NUMBERS
Research Scenario: Fake reviews of online products have been on the rise over the last
decade, where individuals or firms are paid to provide positive or negative reviews of
products they have not actually purchased to sway future consumers. An industrial/
organizational psychologist at a large online retailer wants to determine whether there is a
relationship between fake positive reviews (which present a product as probably better than it
really is) and the number of customer returns. He records the number of positive reviews from
unverified purchasers as well as the total number of times each product was returned over the
course of 6 months. The data for twelve of these products are entered into the table below.
Using this table, enter the data into a new SPSS data file and run a Pearson correlation
coefficient analysis to test whether there is a relationship between the number of fake
positive reviews for a product and the number of customer returns. Create a scatterplot
to display the relationship between the variables. Remember to put your initials within
any and all variable names. Follow the directions below the table to complete the
homework.
PSYC 355
N12 12
LR_Total_Number_Retur
ns
Pearson
Correlation
.649*1
Sig. (2-tailed) .022
N12 12
*. Correlation is significant at the 0.05 level (2-tailed).
2. Write an APA-style Results section based on your analysis. Include your scatterplot as an
APA-style Figure, with figure number and title, as demonstrated in the Results Section
presentation. (Results = 12 pts; Figure = 8 pts)
We conducted a Pearson correlation coefficient to determine whether fake positive
reviews are associated with customer returns. A positive correlation was found
between the two variables r (11)=.649, p=.022, indicating a relationship between the
two variables. The null hypothesis is rejected because the p-value is less than the
commonly used significance level of .05. Therefore, the number of fake positive
reviews and customer returns are statistically significant. There is evidence to
suggest that fake positive reviews are influencing customer decisions and increasing
returns.
PSYC 355
Age CCAS Score
47 45
52 23
60 30
61 26
20 66
39 54
39 28
59 40
40 50
26 98
57 33
23 38
60 25
37 29
19 83
27 62
55 31
60 30
32 60
30 73
46 27
24 88
1. Paste SPSS output. (10 pts)
Problem Set 2: Pearson Correlation Coefficient Analysis
Research Scenario: Recent research into the effects of climate-based events on mental health
suggests that climate anxiety is on the rise, especially among younger people. A clinical
psychologist would like to determine whether there is a relationship between age and climate
anxiety. She administers the Climate Change Anxiety Scale (CCAS; Clayton, S., & Karazsia, B.
T., 2020) to 22 adults between the ages of 18 and 65. Scores on the CCAS range from 22 to
110, with higher scores indicating higher climate anxiety and cognitive and functional
impairment related to the anxiety. The data are listed in the table below.
Using this table, enter the data into a new SPSS data file and run a Pearson correlation
coefficient analysis to test whether there is a relationship between age and scores on the
CCAS. Create a scatterplot to display the relationship between the variables. Remember
to put your initials within any and all variable names. Follow the directions below the
table to complete the homework.
PSYC 355
Correlations
LR_Age
LR_CCAS_Sc
ore
LR_Age Pearson
Correlation
1 -.771**
Sig. (2-tailed) <.001
N22 22
LR_CCAS_Scor
e
Pearson
Correlation
-.771** 1
Sig. (2-tailed) <.001
N22 22
**. Correlation is significant at the 0.01 level (2-tailed).
2. Write an APA-style Results section based on your analysis. Include your scatterplot as an
APA-style Figure, with figure number and title, as demonstrated in the Results Section
presentation. (Results = 12 pts; Figure = 8 pts)
We conducted a Pearson correlation coefficient analysis to assess the association between
age and climate anxiety. There was a significant negative relationship between age and
climate change anxiety (r (19) = -.771, p = .001), indicating a strong inverse linear
relationship. As a result, we reject the null hypothesis that there is no correlation between
age and climate change anxiety.
PSYC 355
Problem Set 3: Cumulative Knowledge Question
Research Scenario: In response to recent Gallup and Pew Center polls of Americans’ trust in
their political leaders (https://news.gallup.com/poll/512651/americans-trust-local-
government-congress-least.aspx; https://www.pewresearch.org/politics/2022/06/06/public-
trust-in-government-2/), a social psychologist decides to study people’s trust in politicians
among different political parties and regions in his state. The psychologist administers a scale-
level questionnaire with possible scores ranging from 1-30, with higher scores indicating
higher trust in politicians, and lower scores indicating lower trust. The data from this scale-
level questionnaire are shown in the table below.
Using this table, enter the data into a new SPSS data file. Choose the correct analysis (it
will be one learned in a previous module) to determine whether there is a difference in
trust scores depending on political party and region. Choose and create the correct
graph that corresponds with this analysis. Remember to put your initials within any and
all variable names. Follow the directions below the table to complete this part.
PSYC 355
1. Urban 2. Rural
1. Republican 16
15
29
18
19
28
29
30
2. Democrat 30
24
28
15
19
14
16
20
3. Libertarian 21
12
10
9
9
18
15
11
4. Independent 15
23
6
21
15
9
15
13
1. Paste SPSS output. (10 pts)
Descriptive Statistics
Dependent Variable: LR_Score
Parties Regions Mean
Std.
Deviation N
Republican Urban 19.50 6.455 4
Rural 26.50 5.066 4
Total 23.00 6.547 8
Democrat Urban 24.25 6.652 4
Rural 17.25 2.754 4
Total 20.75 6.018 8
Libertarian Urban 13.00 5.477 4
Rural 13.25 4.031 4
Total 13.13 4.454 8
Independen
t
Urban 16.25 7.632 4
Rural 13.00 2.828 4
Total 14.63 5.605 8
Total Urban 18.25 7.298 16
Rural 17.50 6.583 16
Total 17.88 6.847 32
PSYC 355
Tests of Between-Subjects Effects
Dependent Variable: LR_Score
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Partial Eta
Squared
Corrected
Model
758.500a7 108.357 3.742 .007 .522
Intercept 10224.500 1 10224.500 353.076 <.001 .936
Parties 541.250 3 180.417 6.230 .003 .438
Regions 4.500 1 4.500 .155 .697 .006
Parties *
Regions
212.750 3 70.917 2.449 .088 .234
Error 695.000 24 28.958
Total 11678.000 32
Corrected
Total
1453.500 31
a. R Squared = .522 (Adjusted R Squared = .382)
Multiple Comparisons
Dependent Variable: LR_Score
Tukey HSD
(I) Parties (J) Parties
Mean
Difference (I-
J)
Std.
Error Sig.
95% Confidence Interval
Lower
Bound
Upper
Bound
Republican Democrat 2.25 2.691 .837 -5.17 9.67
Libertarian 9.88*2.691 .006 2.45 17.30
Independe
nt
8.38*2.691 .023 .95 15.80
Democrat Republican -2.25 2.691 .837 -9.67 5.17
Libertarian 7.62*2.691 .043 .20 15.05
Independe
nt
6.12 2.691 .132 -1.30 13.55
Libertarian Republican -9.87*2.691 .006 -17.30 -2.45
Democrat -7.62*2.691 .043 -15.05 -.20
Independe
nt
-1.50 2.691 .944 -8.92 5.92
Independe
nt
Republican -8.37*2.691 .023 -15.80 -.95
Democrat -6.12 2.691 .132 -13.55 1.30
PSYC 355
Libertarian 1.50 2.691 .944 -5.92 8.92
Based on observed means.
The error term is Mean Square(Error) = 28.958.
*. The mean difference is significant at the .05 level.
2. Write an APA-style Results section based on your analysis. Include your graph as an
APA-style Figure, with figure number and title, as demonstrated in the Results Section
presentation. (Results = 12 pts; Figure = 8 pts)
A two-factor analysis of variance revealed significant differences in trust scores by political
party affiliation, F(2, 20) = 28.958, p .001. A Libertarian (M = 9.88) had a significantly
higher mean trust score than a Republican (M = 2.25, p = .006), while an Independent (M =
8.38) had a significantly higher mean trust score than a Republican (p = .002). A significant
difference was also found in the mean trust score for Libertarians (M = 7.52) compared to
Democrats (p = 0.043). The interaction effect between political party and region, however,
was not statistically significant, F(2, 20) = 1.50, p = .944. There is therefore no difference in
trust scores between political parties, rejecting the null hypothesis.
3.
Powered by TCPDF (www.tcpdf.org)
