When preparing a business interruption calculation for an insurance claim, one of the most common statistical tools I use when evaluating continuing and non-continuing business expenses is the correlation coefficient. The correlation coefficient can be found using analysis tools found in spreadsheet software, using specialized statistics software, or by calculating the values with a pencil and calculator (Yuck!).
The correlation coefficient is significant because it tells whether or not two variables move in relation to one another and it also describes the relative strength of any potential relationship. In this post, I want to address positive correlations. The correlation coefficient is always expressed as a value between 0 and 1 for a positive correlation. If the correlation is 1, a perfect positive correlation, then every time X increases or decreases then Y does the same proportionally. There are no hard and fast rules, but I apply the following general cut offs when reviewing expenses:
0.75 – 1.00 Strong positive correlation
0.40 – 0.75 Weak positive correlation
0.00 – 0.40 Inconclusive or no correlation
For example, if variable X is the number of widgets produced and variable Y is direct production labor hours, I would expect to see both numbers moving up or down in relative tandem. If I calculated the correlation coefficient at 0.92 I would say that if production of widgets was reduced, I would expect to see a closely proportional savings in direct labor hours.
Alternatively, if the coefficient is 0.35, the expense may be saved, but I would suggest that it be evaluated on a basis other than being tied to widget production because there is no identifiable relationship between the two variables.
Need a forensic accountant?
We’re not your typical bean counters. We know how to calculate, present, and defend business interruption insurance claims with off-the-charts winning results for our clients.
We’d love to help you. Please get in touch.
