1. Describe two (2) research scenarios where cluster
analysis may be used as the data analysis strategy.
* When a business wants to introduce a new product, it must first
determine its target audience. The corporation can discover unique
groups of customers that may react differently to the new product by
performing a cluster analysis on its customer data (taking factors like
age, income, historical purchasing behavior, geographic location, etc.
into consideration). For instance, a segment may include high-earning
young adults who are early adopters of technology, and they can be
targeted with certain marketing techniques.
* Cluster analysis can be used by educational institutions to
classify students according to factors including performance, learning
preferences, and interests. This can aid in developing effective
academic programs, personalizing teaching strategies, and identifying
students who may require further support.
2. Describe two (2) research scenarios where one way manova
may be used as the data analysis strategy.
* The two dependent variables are “short-term memory recall” and
“long-term memory recall,” while the independent variable is “lecture
duration,” which has four independent groups: “30 minutes,” “60
minutes,” “90 minutes,” and “120 minutes.” You could use a one-way
MANOVA to determine whether there were differences in students’
short-term and long-term recall of facts based on three different
lecture lengths.
* 33 participants are divided into three groups at random by the
researcher. The first group receives interactive technical dietary
information from an online website. Group 2 receives the same
information from a nurse practitioner, but group 3 receives it from a
video recording produced by the same nurse practitioner. To see if there
is a difference between the presentation types, the researcher examines
three different ratings of the presentation: difficulty, usefulness, and
importance. The researcher is particularly interested in whether the
interactive website is better because that is the most economical way to
convey the information.
3. Describe two (2) research scenarios where linear
discriminant analysis may be used as the data analysis
strategy.
* LDA is extremely useful in the medical profession for categorizing
patient diseases based on a variety of factors relating to the patient’s
health and the current medical treatments. It categorizes disease as
mild, moderate, or severe based on these criteria. The doctors can alter
the treatment’s pace by using this classification to their
advantage.
* Currently, LDA is being used for customer identification. It
implies that we may quickly find and choose the features that can
specify the demographic of clients who are more likely to buy a
particular product in a mall with the aid of LDA. This can be useful if
we want to identify a specific demographic of shoppers who tend to buy a
particular item in a mall.
4. Describe two (2) research scenarios where factor analysis
may be used as the data analysis strategy.
* To find “factors” that can explain a range of test outcomes,
factor analysis is performed. For instance, intelligence studies have
shown that persons who perform well on verbal aptitude exams also
perform well on other verbal aptitude tests. Researchers used factor
analysis to identify one factor, commonly referred to as verbal
intelligence, which reflects an individual’s capacity to solve
challenges involving verbal skills.
* In disciplines including finance, biology, psychology, marketing,
operational research, and the like, factor analysis is applied. For
instance, when asked about a product’s utility, cost, and durability
during surveys regarding consumer satisfaction, respondents may give
similar answers.
5. Mr. Smith wants to determine the dimensions, if any,
along which certain attitudinal variables may be related to certain
health characteristics. The attitudinal variables are attitude toward
the role of women (attrole), toward locus of control (control), toward
current marital status (attmar), and toward self (esteem). Larger number
indicate increasingly conservative attitudes about the proper role of
women, increasing feelings of powerlessness to control one’ fate
(external as opposed to internal locus of control), increasing
dissatisfaction with current marital status, and increasingly poor
self-esteem). The health variables are mental health (menheal), physical
health (phyheal), number of visits to health professionals (timedrs),
attitude toward the use of medication (attdrug), and frequency-duration
measure of the use of psychotropic drugs (druguse). Large numbers
reflect poorer mental and physical health, more visits to doctors,
greater willingness to use drugs, and more use of them. Use the data
‘canondat.sav’ for this problem. Screen the data for normality and
multivariate outliers. Perform necessary transformation of the
variables, if necessary. Conduct a canonical correlation analysis and
determine what dimension(s) of attitude and health variables may be
related. Report your results in 2-3 paragraphs, including summary
tables.