QUESTION $5$

The primary database, consisting of $100$ observations on $18$ separate variables, is based on a market

segnentation study of HBAT customers. HBAT sels paper products to two market segments: the

newsprint industry and the magazine industry. Also, paper products are soid to these market

segnents elther directy to the customer or indirectly through a broker. Two types of infomaton were

collected in the surveys. The first type of information was perceptons of HBAT's performance on $13$

STA$4820/015/0/2024$

attributes. These atributes, developed through focus groups, a pretest, and used in previous studes,

are the most infuertial in the selection of supplers in the paper industry. Respondents included

purchasing managers of fims buying from HBAT, and they rated HBAT on each of the $13$ attributes

using a $0-10$ scale, with $10$ beling "Excellent" and $0$ belng "Poor". The second type of information

relates to purchase outcomes and business relatonshlps $($e$.$g$.$ satistaction with HBAT and whether

the firm would consider a strateglc allance$/$partnership with HBAT$).$ A third type of information is

avallable from HBAT's data warehouse and includes information such as slze of customer and length

of purchase relationship.

Exploratory factor analysis can identify the structure of a set of variables $36$ well $36$ provide a process

for data reduction. In our example, the perceptions of HBAT on $13$ attributes $($X$6$ to X$18)$ are examined

for the following reasons:

Data Summarization with Interpretation $-$ Understand whether these perceptions can be

"grouped". Even the relatively small number of perceptions examined here presents a complex

plcture of $78$ separate correlations. By grouping the perceptions and then engaging in the steps

of factor interpretation, HBAT will be able to see the big plcture in terms of understanding its

customers and what dmensions the customers thinik about HBAT.

Data Reduction $-$ Reduce the $13$ varlables to a smaler number of composite factors. If the $13$

variables can be represented in a smaller number of composite variables, then the other

muitivariate techniques can be made more parsimonious. Of course, this approach assumes

that a certain degree of underiying order exists in the data belng analysed.

a$.$ Using the Table $1$ and Figure $1$ below how many factors should the researcher keep and

why? $(3)$

Table $1$