Thursday, March 28, 2013

Distance to 99 Ranch vs. School Ratings


Someone on Facebook asked me to do a scatter plot of city distance to 99 Ranch and School Rating. For those who do not know what that is, 99 Ranch Market is an Asian American supermarketchain owned by Tawa Supermarket Inc.







BTW I saw this recently from an internet humor site (I didn't create this):


Out of curiosity, I used Google Maps to find the distance of Walmart and Whole Foods from these areas. Suppose the followings: dm=closest distance to Walmart, df=closest distance to Whole Foods, then Snob Score is defined as (dm-df / average(dm, df)). The size of the circle is defined by how expensive a home is. The above humor site has a point, but unfortunately Saratoga and Los Altos ruined the best of line fit in the scatter:



I'm not sure how this is helpful to anyone especially given the fact that the cities here are hand-picked and subject to selection bias. At any rate, what do you think, is there anything you can make out from these plots?

Tuesday, March 26, 2013

Home Prices vs. School Ratings

As my son turns 2, we are starting to realize the importance of location. In particular, location will in large determine how good a school/district he will get into. Here are some of the scatter plots I generated. I picked the cities that many of my friends/family live in. I got all the data from Redfin (older Feb 2013 data) and Google Maps. The scatter plot below shows a strong correlation between the $/sqft of home vs the average school rating (taking an average of all the schools in the area, listed on Redfin-- make sure to read the disclaimer below). Note that when taking area into account, there is a big distinct best fit line between the Bay Area and SoCal. BTW the size of the circles throughout the scatter plots below correspond to median home prices:




Below is a similar plot based on the median home price vs average school rating:








I was curious to see how the average number of offers drive up the listing price. For example, does a city with an average of 12.3 offers/home end up with a higher-than-original listing price than the city with average of just 1.5 offer/home? Look at the scatter plot below and judge for yourself (poor Milpitas, with so many competing offers, it is still significantly lower than listing price):




I was curious to see how strong the correlation is between the median price of a home and the $/sqft, and I find it interesting that the more expensive homes also happen to cost more $/sqft. In another word, you can pretty much interchange the median price of a home and the $/sqft on any of the scatter plot here and see very similar results:




Here's the last plot, and let's play devil's advocate here. Countless university grads look at education as a must-spend-at-all-cost expense. But there are many other people who look at education in terms of ROI (return on investment). For example, someone who goes to a vocational school isn't thinking about his career but instead thinking about getting a job. He is trying to justify spending [limited] money on certifications/AA degrees that will hopefully help him get higher paying jobs. Assuming all high school degrees are nearly the same, then degrees from the left side (San Ramon, Simi Valley), are much cheaper than degrees from the right side (Palo Alto, Santa Monica).

In another word, nobody ever cares about what elementary, middle, high school you went to... what is the point of bidding up home prices? Why spend so much in Palo Alto when cheaper San Jose homes will give your kids the same degrees? Is the housing rat-race really worth the hype?





Are you still convinced that good school districts are worth every single dime? You may be surprised when you read the Two-Income Trap: Why Middle-Class Parents Are Going Broke by E. Warren. Piaw has a good review on this book as well:


Please share this URL with your friends/family members if this link is useful and/or entertaining. Thanks!


-Kevin



Disclaimers:
I understand there are many flaws to the methodology, including: "a few bad apples ruined my area", a big city will tend to incorporate many other neighboring (and sometimes less desirable) schools, student population is not taken into account, loss of precision in Redfin when they mapped API scores to integer values, typos/mistakes in data entry, so on so forth. I recognize that this blog is subject to all of the above problems and more, and that they need to be addressed to make a more accurate report.

On the side, I find it interesting to hear a lot of strong emotion-based criticisms about my "research", as if a lower score on where a person already invested in becomes a personal attack! All I can say is that everyone has a valid point, so please calm down and take these scatter plots/data with a grain of salt. If you want to see raw data, I've shared it publicly here.