Question

# Refer to the Journal of Applied Psychology (Jan. 2011) study of the determinants of task performance. In addition to x1 = conscientiousness score and

Normal distributions

Refer to the Journal of Applied Psychology (Jan. 2011) study of the determinants of task performance. In addition to $$x_{1} =$$ conscientiousness score and $$x_{2} = \{1 \text{if highly complex job, 0 if not}\}$$, the researchers also used $$x_{3} =$$ emotional stability score, $$x_{4} =$$ organizational citizenship behavior score, and $$x_{5} =$$ counterproductive work behavior score to model y = task performance score. One of their concerns is the level of multicollinearity in the data. A matrix of correlations for all possible pairs of independent variables follows. Based on this information, do you detect a moderate or high level of multicollinearity? If so, what are your recommendations?

$$x_{1}\ x_{2}\ x_{3}\ x_{4}$$

Conscientiousness $$(x_{1})$$

Job Complexity $$(x_{2}).\ 13$$

Emotional Stability $$(x_{3}).\ 62.\ 14$$

Organizational Citizenship $$(x_{4}).\ 24.\ 03.\ 24$$

Counterproductive Work $$(x_{5})-\ .23-\ .23-\ .02-\ .62$$

There is a, moderate level of multicollinearity if there is a correlation between a pair of independent variables that is between 0.2 and 0.8 in absolute value. We note that the correlation between emotional stability $$x_{3}$$ and conscientiousness 7 is 0.62, which is between 0.2 and 0.8. We then detect a moderate level of multicollinearity.