## People Aren’t Normal

There was a fascinating journal article published in Personnel Psychology (“The Best and the Rest: Revisiting the Norm of Normality of Individual Performance”) that showed personal performance across many industries didn’t fit a normal distribution (bell curve) but instead fit a Paretian curve (power law).  The Paretian curve is similar to the 80/20 rule which states that 80% of the work is done by 20% of the population and this revelation flies in the face of traditional corporate thought with regard to evaluating employee performance.

Before we get into the meat of the journal article, let’s go over some high level statistical information.

Anyone who is in statistics, engineering or manufacturing is familiar with a Normal Curve (also called the Bell Curve).  This distribution graph (shown below) has the probability of a certain variable being encountered represented on the y-axis and the continuous range of variable possibilities represented on the x-axis.  The  Normal Curve predicts that over 68% of the population will be within +/- 1 standard deviation of the mean and only 0.2% of the population will reside at the tails (outside of +/- 3 standard deviations).

The following picture shows the difference between the Normal Distribution and the Paretian curve.

You’ll notice that the right tail of the Paretian curve (areas of exceptional performance) is thicker which means a larger percentage of the population performs at the exceptional level than what would be predicted by a Normal Curve.  The Paretian curve also shows that a larger percentage of the population performs below the mean than what is predicted from the Normal Curve.

When talking about variables related to “things” (manufacturing process variables such as temperature, pressure, flow rate, etc. or characteristics such as height, weight, distances, etc.) then the Normal Curve is a good approximation of the population but this journal article concludes that the Normal Curve does not approximate the performance of individuals.

Now let’s get to the results of the journal article.  Here is a quote from the journal article introduction:

“We revisit a long-held assumption in human resource management, organizational behavior, and industrial and organizational psychology that individual performance follows a Gaussian (normal) distribution. We conducted 5 studies involving 198 samples including 633,263 researchers, entertainers, politicians, and amateur and professional athletes. Results are remarkably consistent across industries, types of jobs, types of performance measures, and time frames and indicate that individual performance is not normally distributed—instead, it follows a Paretian (power law) distribution. Assuming normality of individual performance can lead to misspecified theories and misleading practices. Thus, our results have implications for all theories and applications that directly or indirectly address the performance of individual workers including performance measurement and management, utility analysis in preemployment testing and training and development, personnel selection, leadership, and the prediction of performance, among others.”

Anyone who works in a company that has more than a dozen employees has to go through an annual performance review period and they know that the ratings for their team are forced to fit into a normal distribution – very few low/high performers and the vast majority of the population must fit into the “average” label.  It’s a frustrating process and this journal article sheds light on why it is so difficult –Personal performance doesn’t fit the Normal Distribution.

This article covered 5 studies – The number of published articles for researchers, the number of Emmy nominations for actors/actresses, the number of times someone was re-elected to the state and national office, positive college/professional sports statistic and negative college/professional sports statistics.  Here is a summary of the conclusions from the article (emphasis mine):

“We conducted five separate studies to determine whether the distribution of individual performance more closely follows a Paretian curve than a Gaussian curve. In all studies, the primary hypothesis was that the distribution of performance is better modeled with a Paretian curve than a normal curve. For each of the five studies, we used the chi-square (χ2) statistic to determine whether individual performance more closely follows a Paretian versus a Gaussian distribution. The chi-square is a “badness of fit” statistic because higher values indicate worse fit (Aguinis & Harden, 2009). That is, the greater the degree of divergence of an empirically derived performance distribution from a Gaussian or Paretian distribution, the higher the chi-square. Accordingly, for each of the samples we studied we first forced the data to conform to a normal distribution and then forced the same data to conform to a Paretian distribution. For each comparison, a smaller chi-square value indicates which of the two theoretical distributions describes the data better.”

In all areas that this study investigated, the distribution of personnel performance followed a Paretian distribution instead of a Normal distribution.  Here are a few of the distributions that are presented in the paper that prove this conclusion.

So what does this tell us?

Managers have been trying to fit their employees into a Normal Curve for decades and the reason this has been a struggle is because this is not reality!

There are a vast number of employees who perform below “average” when measured against the Normal Curve but “below average” on the Paretian curve doesn’t carry the same stigma as “below average” on the Normal curve.  If a company chose to terminate all those who perform below average on a Paretian curve then a majority of its employees would be sent packing.   Companies need a new grading scale that recognizes that a large, but important, percentage of employees actually performs “below average” but not so low to warrant disciplinary action.

There are also a significant proportion of employees who perform “above average” and this number is vastly greater than is predicted by the Normal Curve.  Companies need a new grading scale that accounts for better differentiation of employees that perform above expectations since the right hand tail of the Paretian curve is thicker than the Normal Curve predicts.  Companies need to spend more time differentiating their top performers, rewarding them accordingly and incentivizing those “below average” to improve their performance.

This study should prompt companies to radically reform their performance metrics.