data bubbles continued

Following up on my previous post on mint.com’s data on average check sizes at different dining establishments in RVA, I recently played around on rather cool site called bundle which also tracks spending. Bundle has some nifty little flash bubbles that you can customize to see how people spend money across the country on different categories each month (food, travel, home, entertainment etc). Since this is a “food blog” (sic), I only looked at some Richmond data, specifically, how much the average Richmonder (a nebulous concept - one can also get data from Midlothian or Henrico county) spends each month on food, dining and the like.
In addition to location, you can specify age group, income level (the rich spend more - surprise !!) and even household types (for example, in an interesting bit of socioeconomic behavior, average spending on restaurants & bars in Richmond – single male, no kids - $303, single female, no kids - $197 and married, no kids - $269). Overall average across all groups - $253 on eating out and $248 on groceries. Depending on your fascination (mine is obviously high) with this kind of silly information, you can spend quite a bit of time trying different permutations and combinations.
Avg monthly spending on restaurants and bars in other cities: Washington DC - $473, San Francisco, CA - $419, New York, NY - $287, Boston, MA - $335

Again, mathematically speaking, despite the claimed 20 million data points, it is easy to get misled by averages (vs. medians) and statistical errors including demographics represented, location and such. But looking at the distributions is probably more telling, although I do wonder about that as well. The graph below shows the average distribution on dining out across Richmond.



acknowledging the excellent work of RVAfoodie in redesigning the eatingrichmond.com aggregator - what has become a fascinating glimpse into so many musings across town.