How to measure employee performance with little information
While those that have a padded resume of related work experiences build employer confidence when it comes to their ability to perform well—one challenge remains.
We still have a difficult time distinguishing the transfer of previous work performance to a new workplace.
To what extent does an employee’s exceptional performance in their education and previous work outcomes translate to good performance and outcomes in their current position—and how do we measure it?
Because we want to extrapolate from employees their best work and performance outcomes without neglecting their strengths and preferences, it’s inevitable we end up measuring success in small sample sizes actually relevant to the role.
The notion of working with smaller sample sizes mirrors that of the Central Limit Theorem whereby it claims that smaller data pools will result in us seeing more extreme outcomes. We know this to be true for a number of reasons.
For instance, you wouldn’t assess one employee’s performance to be successful based on the success of an entire department. Just as you wouldn’t discount another employee’s high performance as a result of a department not reaching collective goals.
Another example of this is where we’re much more likely to see either great outcomes or poor outcomes at the very beginning of an employee’s lifecycle with a particular team or company. Hence, the probation period of course.
So what are we taking away from this?
The theorem itself tells us that we should be cautious when it comes to drawing quick conclusions on employee performance when the data pool is small.
At 7Geese, we avoid this by tracking goals and obtaining ongoing 360 feedback. We do this because we’re working towards placing more emphasis on measuring long-term employee performance with team and company outcomes.
We tend to believe the outcomes drawn from a small sample size is representative of future outcomes in employee performance. Law of Small Numbers
Small sample sizes and a lack of data are one of the key reasons performance reviews have become discredited for effectiveness over the years.
What’s most frustrating for employees is when conclusions are drawn during performance reviews based on a very small sample of data from their recent performance rather than data from across the year.
So how can you gain more data on employee performance?
This is where ongoing feedback and coaching can lighten the strong inferences made from small samples of employee performance and shed light on the importance of measuring the longevity of positive outcomes as a result of exceptional performance.
There’s no magic shortcut in distinguishing the level of employee performance. The lesson is always to obtain more data before drawing conclusions about employee performance and outcomes and resist applying strong inferences from small sample sizes.
In fact, this is why we love our automated performance reviews at 7Geese. We believe that one of the best ways to measure employee performance is to compile data throughout the year. So when it comes time for performance reviews, the performance data is automatically populated and ready to use.
For more information on performance review best practices, check out our latest guide.
Also published on Medium.