Chapter 5 -- Correlating Personality & Behavior

This exercise illustrates correlation and temporal aggregation. It uses the PERS dataset, consisting of 90 cases and 968 variables. The variables represent measures of traits and relevant behaviors for the dimensions of extraversion (outgoingness) and conscientiousness, reported each week for three weeks by a group of undergraduate psychology students.

How well do you think you could predict behavior from a relevant personality trait? If you gave a group of people a personality test of conscientiousness, for example, how strong do you think it would correlate with specific behaviors, such as neatness or thoroughness? Do you think you could predict the behaviors as well for one specific day or week, compared to a larger span of time?

This exercise will examine the relationship between personality trait measures and behavioral measures, specifically to predict behavior from personality. Behaviors are unreliable when measured at one point in time, so measures of behaviors aggregated over time should correlate better with stable, general characteristics such as personality traits.

Unreliability of single-occasion measurements.

How well can personality traits predict behavior? The empirical counterpart to this question is "How strongly do personality trait measures correlate with behavioral measures?" For example, if the trait of interest was outgoingness, or "extraversion" as it is termed by researchers, a list of relevant behaviors which would illustrate this trait might be measured for a group of people. This might include sociability indicators such as the number of people you have contact with or interact with, indicators about the nature of your social interactions, such as whether you are dominant, assertive, and energetic, and perhaps the quality and quantity of talking that you do in various situations. Behavioral indicators might be assessed in terms of things you have the opportunity to do every day, or things you might only have opportunities to do a few times in a week. The assumption would be that outgoing people have a general personality trait which predisposes them to exhibit these behaviors to a greater degree than people who are less outgoing. A person high in outgoingness should interact with more people, be more assertive in their interactions, talk more, etc., while a less outgoing person should do these all to a lesser degree. Historically, studies would measure these various behaviors at one point in time—ignoring the inherent unreliability of such a measure, since these behaviors may vary from day to day or week to week in the same person, even if the personality of such a person was relatively consistent and "fixed."

If you completed Chapter 4, you have already seen that these behaviors correlate better with each other when they are averaged over time, rather than measured on a single occasion. Not surprisingly, this exercise will show that personality trait measures predict (correlate with) aggregated behaviors better than they predict single-occasion behaviors. Personality traits do not predict with great accuracy what you will do on one day in one situation; they are useful in predicting your average level of behavior over a span of time, however.

This exercise will use weekly behaviors for outgoingness, as an example. Look at these measures in the codebook. We will use the first ten of these, ranging from "ate breakfast w/others" to "went out of way to talk to stranger." There are two personality inventories that were used, the Jackson Personality Inventory (JPI) and the California Psychological Inventory (CPI). First, we will correlate the JPI Social Participation scale (JPISPT) and the CPI Sociability scale (CPISY) with each of the first ten weekly behaviors for outgoingness, using only the measures from Week#1.

Use Analyze>Correlate>Bivariate, and select these variables:

jpispt, cpisy, w1wbo1, w1wbo2, …. w1wbo10 (a total of 12 variables)

Look at the correlation output. Find the highest, lowest, and median correlation for each personality scale across the 10 behaviors.

Now repeat the correlations, this time using the 3-week averages instead of the Week#1 measures (e.g., wbo1, wbo2, etc.)

Look at the correlation output. Find the highest, lowest, and median correlation for each personality scale across the 10 behaviors. Compare the two sets of results.

It should be the case that the aggregated behaviors correlate better. Try this with other behaviors, or even with the other personality measures (JPIRSY and CPIRE, which measure responsibility, i.e., conscientiousness). Don’t forget to recode any reverse-scored items (see codebook suggestions for recoding reversed items; see also Chapter 1).