What Dashboards Are Telling You: Descriptive Statistics and Correlations
“Most often when we see illogical behavior, the fault is in the measurement system, not in the employees!”
Today, most HR organizations have some method of reporting data on their activities. These methods can be as simple as an Excel spreadsheet kept on a user’s desktop or may involve specialized or custom-written tools involving databases or business intelligence. The reports that come out of these data sources can be called descriptive statistics. The reports might be a few tables or graphs from a spreadsheet, dashboard-like reporting mechanisms, or other specialized tools such as SAS Enterprise BI Server. The goal of a descriptive statistic is to provide a quick indicator of what is going on in the organization, such as: What is the headcount? Who completed what training? How much is payroll? The information from these sources would generally be presented in some kind of dashboard and/or scorecard. The word correlation can be used in various ways. In our case, correlation is when various key performance indicators tend to vary according to the circumstances. For example, if we see that sales are higher in the Eastern region, we say that the sales are correlated with region.
This book has hinted at preferences for “heavy statistics” of the sort that require some sort of specialized mathematical knowledge to fully implement. So, why talk about the simpler things? ...