Assessing Leadership

There are a number of difficulties in assessing leadership. Personal opinions about leadership effectiveness can vary across individuals (i.e. one person’s maxim can vary differently from another person’s). Your idea of an effective leader (or manager) may be different than mine. You may think that a good teacher is easy and gives lots of A’s. I may think that a good teacher really challenges students and doesn’t give any A’s.

It is important for organizations to choose good leaders because the wrong decision could cost stockholders billions of dollars and lead the organization to fail. For this reason, researchers have studied ways to identify leadership talent, both qualitatively and quantitatively.

Qualitative Approach

The most common qualitative approach is the case study. Case studies are in-depth analyzes of leaders’ activities. Case studies can provide leadership practitioners with ideas on what to do in different leadership situations. They can show us what traits are associated with leader effectiveness. The problem with case studies is that there is no objective way to determine whether the actions taken actually caused the results, as we discussed above with regards to maxims.

Quantitative Approach

Quantitative approaches focus on numbers rather than description to measure performance of leaders. The most widely used measure of performance is to use performance appraisal ratings of the individual. In performance appraisal ratings, the leader’s superior would rate performance on several dimensions and then provide a recommendation (or not) for a promotion. In our department head example, we may ask the dean to rate our department head.

A better way to judge leadership success may be through subordinates’ ratings of leadership effectiveness. To do this, we would simply ask the subordinates to rate their level of satisfaction with their leaders. Thus we may ask individual professors to rate their satisfaction with the department head.

Finally, we could use unit performance indices to examine what impact leaders have on the bottom line of the organization. We could judge leadership success by examining profit margins, or win-loss records. We could judge the success of our department head by judging how much money the psychology department raised per year or how many students the department graduated.

Ideally we would combine several of these measures of performance to create a more holistic picture of the leader's potential, in this case our department head.

To assess the relationship between factors like a leaders traits and their performance, we could use quantitative approaches to in correlational studies or experiments. You have probably studied these approaches in other classes, but we will briefly discuss them as an important review for some and as an introduction for the rest.

Correlational studies

Correlational studies help us know the statistical relationship between leaders’ traits, mental abilities, or behaviors and various measures of leadership effectiveness such as subordinates’ satisfaction. For example, a correlational study could help us know if there is a relationship between how satisfied psychology professors are with their department head and the number of psychology students that graduate from that department.

A correlation coefficient can be calculated. Correlation coefficients range from -1.00 to +1.00. Coefficients close to 1.00 would indicate a strong positive relationship and those close to -1.00 would indicate a strong, negative relationship.

Let’s say that we found a correlation of .81 between professor satisfaction with the department head and the number of students who graduated from the department. This means that as professors are more satisfied with their department head, more students graduate from the department. This relationship is positive, and fairly strong.

The main problem with correlational studies is that it is hard to make causal inferences based on correlational data. Thus, we cannot assume that the teacher satisfaction with the department head caused more students to graduate. We only know that there is a relationship between teacher satisfaction and the graduation rate of students. It is possible that higher graduation rates increase professor satisfaction. Remember, correlation does not mean causation. A correlation is simply a description of a relationship between two variables. For all we know it could be a third variable causing both. In this case, maybe easily available research funds for this department make professors satisfied and provide students with ample learning opportunities.

Experiments

Experiments allow researchers to make causal inferences about leadership. Experiments consist of both independent and dependent variables. The independent variable is what the research manipulates. The independent variable could be some type of leadership behavior. The dependent variable is what we are interested in measuring and is usually some measure of leadership effectiveness.

We may want to know if there is a relationship between the amount of time a leader spends with subordinates (independent variable) and subordinate performance (dependent variable). We would design an experiment with two groups. In the first group, leaders spend only 5 minutes a day with each subordinate. In the second group, leaders spend 30 minutes a day with each subordinate. Subordinate performance could be measured with performance on a test. We would look to see if those subordinates who spent 30 minutes daily with their leaders scored higher on the test than those who spent only 5 minutes daily with the leader.