Practice Lab 3_Validity
pdf
keyboard_arrow_up
School
University of South Florida *
*We aren’t endorsed by this school
Course
4433
Subject
Communications
Date
Feb 20, 2024
Type
Pages
10
Uploaded by colo123nix
Assignment created by Dr. Wendy Rote 1 Practice Lab 3 - Validity Analysis
Examine the validity of the Positive Affect Subscale of the Positive and Negative Affect Schedule (PANAS). Here is the Scale: The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988) This scale consists of a number of words that describe different feelings and emotions. Read each item and then list the number from the scale below next to each word. Indicate to what extent you have felt this way over the past week: 1 2 3 4 5 Very Slightly A Little Moderately Quite a Bit Extremely
or Not at All __________1. Interested __________11. Irritable __________2. Distressed __________12. Alert __________3. Excited __________13. Ashamed __________4. Upset __________14. Inspired __________5. Strong __________15. Nervous __________6. Guilty __________16. Determined __________7. Scared __________17. Attentive __________8. Hostile __________18. Jittery __________9. Enthusiastic __________19. Active __________10. Proud __________20. Afraid Scoring Instructions:
Positive Affect Score: Average the scores on items 1, 3, 5, 9, 10, 12, 14, 16, 17, and 19. Scores can range from 1-5, with higher scores representing higher levels of positive affect. Negative Affect Score: Average the scores on items 2, 4, 6, 7, 8, 11, 13, 15, 18, and 20. Scores can range from 1 –
5, with lower scores representing lower levels of negative affect.
Assignment created by Dr. Wendy Rote 2 1.
Open SPSS file: Lab3 Dataset. 2.
Compute a new variable representing individual’s average positive affect (PA) score
a.
Transform →
Compute Variable b.
Give the mean scale a name in the box under “target variable”
c.
In the Numeric Expression box type: mean ( d.
Enter each variable name contributing to the scale separated by “,”s
e.
Type: ) f.
Click “paste” then run the syntax
g.
Make sure that a new variable has appeared in your Data View and that it has numbers in it. Convergent Validity:
Is your scale related to things it should be related to, in the expected directions? 3.
Obtain correlations between PA, depression scores (cesd), and self-esteem scores (esteem) to examine convergent validity.
a.
Analyze →
Correlate →
Bivariate b.
Move all 3 variables (PA, CESD, esteem) into the “variables” box c.
Make sure the “Pearson” box is checked under “correlation coefficients,” the “two
-
tailed” box is checked under “Test of Significance,” and the “flag significant correlations” box is checked
d.
Click “paste” then run the syntax
Assignment created by Dr. Wendy Rote 3 e.
Examine the output Correlations
PosAff Rosenberg Self-Esteem Score CESD: Depression Scale Score PosAff Pearson Correlation 1 .439
**
-.350
**
Sig. (2-tailed) .000 .000 N 492 492 492 Rosenberg Self-
Esteem Score Pearson Correlation .439
**
1 -.556
**
Sig. (2-tailed) .000 .000 N 492 492 492 CESD: Depression Scale Score Pearson Correlation -.350
**
-.556
**
1 Sig. (2-tailed) .000 .000 N 492 492 492 **. Correlation is significant at the 0.01 level (2-tailed). Think about the associations you would expect between positive affect, depression, and self-
esteem. You would expect that people with more positive affect would have more self-esteem (positive correlation) and less depression (negative correlation). Look at the actual correlations. Also consider the size of the effect: What is the correlation between PA and Self-Esteem? .44 Is it significant? Yes, p
< .001
. How large is the effect? Medium-Large.
What is the correlation between PA and Depression?
-.35
Is it significant? Yes, p
< .001
How large is the effect? Medium.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
Assignment created by Dr. Wendy Rote 4 Based on these values: Would you say that the positive affect scale score has satisfactory convergent validity? (does it correlate with related variables in the way you’d expect?):
YES 4.
Obtain a scatterplot between PA and Esteem
to verify that any relationship is linear (rather than curvilinear). The use of Pearson’s r
to demonstrate validity requires that this assumption is met. a.
Graphs →
Chart Builder b.
Under the “Gallery” tab, select “Scatter/Dot” and double
-click c.
From the “Variables” list, d
rag PosAff into the “X
-
Axis?” box. Drag Esteem into the “Y
-
Axis?” box. d.
Click paste then run the syntax.
Assignment created by Dr. Wendy Rote 5 e.
Look for a pattern that is like a straight line (or square) rather than a “u” shape. Ok: Bad: The scatterplot of PA and Esteem looks like:
Assignment created by Dr. Wendy Rote 6 Is it relatively linear?
YES. This means that we can trust the correlation analysis to tell us the whole story about the relationship between PA and Esteem. 5.
Obtain another scatterplot, this time between PA and Depression (CESD)
to verify that any relationship is linear (rather than curvilinear). a.
Repeat steps a-e from #4, except put CESD on the Y-axis instead of Esteem. b.
The syntax will look like: The scatterplot of PA and CESD looks like: Is it relatively linear? YES.
This means that we can trust the correlation analysis to tell us the whole story about the relationship between PA and Depression.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
Assignment created by Dr. Wendy Rote 7 Discriminant Validity
: Is your scale UNRELATED (uncorrelated) to things it shouldn’t be related to? 6.
Obtain correlations between Positive Affect and GPA (catGPA) to examine discriminant validity. f.
Analyze →
Correlate →
Bivariate g.
Move PA and gpa into the “variables” box
h.
Make sure the “Pearson” box is checked under “correlation coefficients,” the “two
-
tailed” box is checked under “Test of Significance,” and the “flag significant correlations” box is checked
i.
Click “paste” then run the syntax
Assignment created by Dr. Wendy Rote 8 j.
Examine the output Correlations
PosAff which is the closest to your actual GPA? PosAff Pearson Correlation 1 .030 Sig. (2-tailed) .513 N 492 492 which is the closest to your actual GPA? Pearson Correlation .030 1 Sig. (2-tailed) .513 N 492 492 Look at the correlation. What is the correlation between PA and GPA? .03 Is it significant? No, p
=.513. How large is the effect? Very small.
Based on these values: Would you say that the positive affect scale score has satisfactory divergent validity? (is it uncorrelated with conceptually unrelated variables?):
YES NOTE: These correlational methods are based on the assumption that the variables included in the analyses are roughly normally distributed. If severely skewed, you will need to use non-
parametric statistics. 7.
Obtain a third scatterplot, this time between PA and GPA (catGPA)
to verify that any relationship is non-linear. a.
Repeat steps a-e from #4, except put catGPA on the Y-axis instead of Esteem. b.
The syntax will look like:
Assignment created by Dr. Wendy Rote 9 The scatterplot of PA and GPA looks like: Is it relatively linear?
YES. *
it just looks a little different because GPA is categorical* This means that we can trust the correlation analysis to tell us the whole story about the relationship between PA and GPA.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help
Assignment created by Dr. Wendy Rote 10 8.
Write up your APA style Results section. Describe how you tested validity (including the scales used and types of validity assessed) and what you found. “
Convergent validity of the Positive Affect Subscale of the Positive and Negative Affect Scale (PA
NAS; Watson, Clark, & Tellegan, 1988) was assessed by correlating participants’ mean positive affect score with their self-esteem and depression. As expected, positive affect was positively correlated with self-esteem, r = .44, p < .001, and negatively correlated with depression, r = -.35, p < .001. These correlations were in the expected direction and of medium to large size, providing good evidence of the convergent validity of the subscale. Discriminant validity of the Positive Affect Subscale was assessed by correlating participants’ mean scores on the positive affect subscale with their GPA, as scored on a categorical 1-5 scale (higher scores equal better grades). As expected, positive affect was not significantly associated with GPA, r = .03, p = .52, and the size of the correlation was very small, providing good evidence of the discriminant validity of the scale.