Averages are Awesome, but Medians May Be More Meaningful!

I’m pretty sure we are all aware of what an average is. Averages have a pretty significant weakness, though, from a data analysis perspective: They are easily skewed by “outlier” data. One extremely large gift, for example, can make your average significantly higher than it would otherwise be. There are numerous things you can do to minimize this impact. For example, statisticians will often remove outlier data on both ends (high and low) before calculating the average. But what constitutes outlier data? If you are planning this route, you then have to start worrying about additional calculations.

At ResultsPlus, we’re all for meaningful, accurate, and EASY. Enter the “median.”

Many of you already know what a median is, but for those of you who don’t, here’s a brief description: A median, in the case of a list of gifts, for example, is the amount of the gift in the middle of the list. So, if I have a list of 11 gifts, sorted in order from lowest to highest (or in reverse, it doesn’t matter in this particular case), the amount of the 6th gift in the list is the median. What this tells me, from an analysis perspective, is that 50% of my gifts were lower than this amount, and 50% of my gifts were over than this amount.

For organizations attempting to raise their “average gift amount”, it’s quite likely that this statement isn’t “the whole story.” While organizations definitely want this to occur, the reasoning behind this, from what I’ve heard during conversations with customers, is that organizations want to increase donor investment in mission, so that the budget will grow to support expanding the amount of good the organization can do while meeting its mission. A one-time bump in the average, due to an outlier gift, looks good on paper for one year, but it doesn’t provide a context for action. Focusing on increasing the median gift will, by its nature, increase the amount of all gifts in the “over the median” portion of the list of gifts. That, in all likelihood, means that you are also increasing your average gift. But, unlike with averages, you can reasonably estimate that you are increasing loyalty and/or capacity for future gifts in a good portion of your list.

Better yet, use averages, medians, minimums, maximums, and total counts in conjunction to get a MUCH better picture of your fundraising activities, donor involvement, and roadmap for future efforts!

Check back next week where my blog post will provide step-by step instructions for generating averages, medians, minimums, maximums, etc. in ResultsPlus and other tools you already know and use.