Exercise

# From correlations to factors

At the end of Chapter 1, we printed and visualized correlation matrices to spot "groups" of items in a set of survey responses. Now you know that the objective of that exercise was to identify dimensions of a latent variable. While helpful, more advanced techniques such as parallel factor analysis provide a more rigorous examination of hidden patterns in survey responses.

Let's work with `b_loyal_10`

, a ten-item survey meant to measure brand loyalty. We'll start by using the same techniques as we did in Chapter 1 to explore the item correlation matrix. What "groups" of items are together reflecting dimensions of a latent variable?

The `psych`

and `corrplot`

package have been loaded.

Instructions 1/4

**undefined XP**

- Print the correlation matrix of
`b_loyal_10`

items using the relevant function from`psych`

.