SOWK 4040 RWC Correlation Analyses Using Google Sheets Lab

1Roberts Wesleyan College
Social Research Methods, SOWK 4040
Statistics Lab #3 –Correlation Analyses using Google Sheets
Assignment Scenario and Instructions
Scenario
You are a social worker who works in a large agency serving adults with developmental
disabilities. Your agency has a therapeutic day rehabilitation (Dayhab) unit, as well as an
employment program where the clients learn job skills to make them ready them for
employment.
Correlation Analysis One
The first analysis you will run involves your work in the Dayhab Unit. You run a social skills
training group for your clients that addresses behaviors such as inappropriate touching,
aggression, inappropriate verbalizations, etc. You have been wondering if the group is effective
in changing these behaviors and whether or not the families of your clients notice any
improvement.
You have a scale which measures client behavior change called the Dayhab Socialization Scale.
In addition to this scale, the agency also routinely collects data from families on the progress
they see in their family member (Family Rating of Client’s Socialization). You believe that if
you can show a significant correlation between the Dayhab Socialization Scale scores and the
Family Rating of Client’s Socialization (measured on a 5-point Likert scale), you will have
support for your assertion that the group is effective. To answer this question, first view the
video “Correlation Analysis in G Sheets” on Moodle. Next, run a correlation analysis with the
variables Dayhab Socialization Scale scores and the Family Rating of Client’s Socialization
using the data in the table on pages 3-4. Run the analysis using the XLMiner ToolPak add-on in
Google Sheets. Instructions are below.
A. Copy and transfer the date from the table on pages 3-4 to a new blank Google Sheets
spreadsheet.
B. Name your Google Sheets file.
C. Open the XLMiner ToolPak add-on in Google Sheets (see previous video on t-tests).
D. Run a correlation using the variables, Dayhab Socialization Scale (Input Y Range) and
Family Rating of Client’s Socialization Scale (Input X Range) by choosing the Linear
Regression option in ToolPak.
E. Now, answer the following questions.
1. What is the correlation coefficient (Multiple R value) of the relationship?
2. What is the p-value (significance) of the correlation?
3. Is the correlation statistically significant (yes or no)?
4. Support or Refute the following hypothesis. Be sure to justify your answer based on the
output values (significance of the relationship).
2
“There is a significant relationship between the client’s behavioral scores on the Dayhab
Socialization Scale and the family ratings of client improvement, as measured on the
Family Rating of Client’s Socialization scale.”
Correlation Analysis Two – Validity Analysis
In your work at the agency, you also work with clients in the vocational program to prepare them
for employment in the community. You have developed a tool to measure job readiness (Job
Readiness Scale), but you’re not sure if it is a valid measure (instrument) to gauge whether your
developmentally disabled clients are ready for work or not.
Being the conscientious social worker that you are, you want to insure that you have a valid
instrument. So, you also decide to validate your new instrument to verify that it is really
measuring job readiness and not some other construct (validity). You do this by comparing it to
another tool which measures job readiness in other populations.
A famous researcher that you know, Professor Leia Jones, has developed such a tool (Jones Job
Readiness Scale). You can demonstrate criterion-related validity for your instrument by
correlating the results of your instrument with Professor Jones’ instrument. You give both the
instruments to your group of vocational clients for this purpose. A strong correlation (.7 or
higher) would show that you have a valid measure of job readiness.
To complete this part of the assignment,
A. Open the XLMiner ToolPak once again.
B. Run a correlation using the variables, Job Readiness Scale (Input Y Range) and Jones Job
Readiness Scale (Input X Range) by choosing the Linear Regression option in ToolPak.
C. Answer the following questions.
5. What is the correlation coefficient (Multiple R value) for this analysis?
6. Does the correlation coefficient indicate that a weak, moderate, or strong relationship
exists between the two scales? (Use .7 as the cut-off for a strong relationship)
7. Is there evidence that your scale (Job Readiness Scale) is a valid measure of job
readiness? Yes or No?
8. How do you know the answer to #7 above?
Scatter Chart Analysis
A. Create a Scatter Chart Graph (with trend line) to graph the results of the validity analysis
above. The trend line option is located under Series in the Chart Editor/Customize menu.
Use the same variables as in the validity analysis. Place the graph on a separate tab in the
Google Sheet. [See video “Creating a Scatter Chart in G Sheets”]
B. To complete the assignment, upload your answers to all the questions in a separate wordprocessed document (Word or PDF only) to the Moodle assignment link. Be sure to number
your answers to correspond to the number of the question on the stat lab instructions.
C. You must also download the Google Sheet output as a Microsoft Excel (.xlsx) file. Go to
File, and click on Download, then choose Microsoft Excel (.xlsx) file. Upload the Excel
file to the Moodle assignment link.
3
Developmental Disabilities Agency Data
Dayhab
Socialization Scale
Score
64
70
59
64
55
71
63
64
63
48
54
64
63
62
57
69
66
55
52
61
60
61
62
74
70
64
63
68
69
63
63
63
64
60
Family Rating of
Client’s
Socialization
Job
Readiness
Scale
1
4
1
4
1
5
1
1
1
2
4
2
1
2
3
2
4
1
1
1
3
3
4
1
4
1
1
4
1
2
3
5
3
4
68
75
61
49
54
71
68
80
56
60
59
66
65
60
59
59
70
64
71
57
77
60
62
77
59
75
67
66
51
80
88
56
63
62
Jones
Job Readiness
Scale
73
82
66
49
59
76
73
85
59
65
63
71
70
64
64
64
76
69
77
62
84
65
68
88
63
80
70
71
57
85
91
60
68
68
4
63
63
80
67
68
69
61
69
68
68
68
69
59
58
57
62
50
62
67
68
65
65
64
68
68
69
74
78
65
73
76
64
79
55
61
66
2
2
1
4
4
1
1
3
1
3
1
3
3
1
2
1
1
3
1
4
2
2
3
2
2
1
4
1
1
3
1
4
3
3
2
1
65
58
63
58
64
64
64
63
65
62
60
72
58
58
63
56
71
63
63
63
63
65
65
63
61
57
67
50
68
64
64
63
64
80
56
58
70
64
68
60
69
61
69
62
70
64
65
75
70
63
66
61
77
68
78
68
72
70
68
61
66
61
72
57
73
67
68
69
66
85
61
63