Scribd
Upload
44 views5 pages

Work Problems Chapter 15

The document discusses a survey given to college students about their concerns for the future regarding the environment, economic security, and social relationships. [1] A factor analysis and reliability analysis were conducted on the survey results to see if students' responses grouped together into meaningful constructs. [2] The analyses identified two solid factors related to the environment and economic security, while the social relationship items did not load strongly onto single factors. [3] Two items cross-loaded onto multiple factors, indicating they did not clearly discriminate between the constructs.

Uploaded by

Nene
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
44 views5 pages

Work Problems Chapter 15

The document discusses a survey given to college students about their concerns for the future regarding the environment, economic security, and social relationships. [1] A factor analysis and reliability analysis were conducted on the survey results to see if students' responses grouped together into meaningful constructs. [2] The analyses identified two solid factors related to the environment and economic security, while the social relationship items did not load strongly onto single factors. [3] Two items cross-loaded onto multiple factors, indicating they did not clearly discriminate between the constructs.

Uploaded by

Nene
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Work Problems Chapter 15

Suppose that I wanted to examine the attitudes of college students regarding their
concerns for the future. I asked a sample of college students to complete a survey that
contained four questions about the environment (e.g., “I am worried about climate
change”), four questions about their own economic security (e.g., “I hope I am able
to find a good job after graduation”), and four questions about their social
relationships (e.g., “I am worried I will lose touch with my friends from college after
I graduate”). 1 I suspect that students will respond similarly to all of the items within
each particular category. In other words, I believe that students who are very
concerned about one aspect of the environment will be concerned about all aspects
of the environment, and will therefore answer all four of the survey items about the
environment similarly. To see whether students’ responses to my survey items
grouped together, I conducted an exploratory principal components factor analysis
and the reliability analysis. Some of the results of these analyses are summarized in
Tables 15.9 and 15.10. Please answer the following questions, some of them based
on the information provided in these tables.
Table 15.9: Eigenvalues, percentage of variance explained, and rotated factor
matrix for work problems.

Rotation Sums of Squared Loadings Rotated Factor Matrix

Factor Total % of Survey 1 2 3 4


(Eigenvalue) Variance items
1 1.92 23.8 Enviro 1 .88

2 1.54 19.2 Enviro 2 .86

3 .83 8.4 Enviro 4 .81

4 .71 6.3 Enviro 3 .53 .36

Money 4 .92

Money 2 .84

Money 3 .74

Money 1 .32 .76

Social 1 .62

Social 2 .42

Social 3 .64

Social 4 .57
Table 15.10: Reliability Statistics for Environmental Items for Work Problem

Cronbach’s Cronbach’s Alpha Based N of Items


Alpha on Standardized Items
.68 .67 4

Corrected Item- Squared Multiple Cronbach’s Alpha If


Total Correlation Correlation Item Deleted
.61 .37 .58
Enviro 1: “I am worried about
climate change.”

.58 .34 .60


Enviro 2: “Rising sea levels are
a serious problem.”

.41 .17 .76


Enviro 3: “My friends and I
volunteer to clean up the
environment.”

.55 .30 .62


Enviro 4: “I am concerned about
how humans are harming the
planet.”
1. What is the purpose of conducting a factor analysis?

The primary purpose of conducting a factor analysis is to see how the


data can be grouped together based on the correlations among the
variables in the dataset. Variables or items that are highly correlated
with each other may represent a unifying construct that is not directly
measured but is indicated by multiple, measured variables. The factors
that emerge from a factor analyses represent these unifying constructs,
often called latent constructs or latent factors.

2. What does a reliability analysis tell you?

A reliability analysis provides information about how well a group of


measured items or variables hold together in a single construct.

3. Using the information about eigenvalues and percentage of variance


explained in Table 15.9, how many solid factors do you think emerged from
the factor analysis of the survey items? Why do you think so?

According to the eigenvalues and percentage of variance explained, it


appears that there are two solid factors in Table 15.9. Two of them have
eigenvalues greater than 1.0 and each explain more than 10% of the
total variance in the items.

4. Using the information from the Rotated Factor Matrix in Table 15.9, which
items are cross-loading on more than one factor? What does this tell you
about these items?

It appears that Enviro 3 and Money 1 each have factor loadings of


greater than .30 on two separate factors. This tells us that these two
items are not doing a good job of discriminating between two factors
and therefore create some confusion about the conceptual clarity of
each of these items.

5. Looking at the factor loadings from the Rotated Factor Matrix in Table 15.9,
what would you predict the Cronbach’s alpha to be for the four Social
items? Why?

Two of the Social items load on Factor 3, and the other two Social items
load on Factor 4. In addition, none of the four Social items load very
strongly (i.e., over .70) on either factor. In addition, the eigenvalues for
Factor 3 and Factor 4 are below 1.0. All of this suggests that these four
items do not go very well together (i.e., are not strongly correlated with
each other), so the Cronbach’s alpha for these items will be quite low.

6. What does the information provided in Table 15.10 suggest regarding which
item, if any, might be eliminated from the Environmental items to create a
good scale?

The “Enviro 3” item, if eliminated, would raise the Cronbach’s alpha


from .68 to .76. This suggests that this item should perhaps be
eliminated from this scale.

7. From a conceptual standpoint, how is the Environmental survey item with


the weakest contribution to the Cronbach’s alpha different from the other
three Environmental items? And how is this difference reflected in the
cross-loading of this item on two factors?

Remember that items should not be eliminated from scales simply


because they weaken its statistical properties. Conceptual fit among the
items should also be considered. Notice that “Enviro 3” is quite
different, conceptually, from the other three items. This item is about
behavior (i.e., volunteering) but the other three items are about
concerns.

You might also like