Today, I start from the beginning, earlier than 101, to explain how real estate data is collected and what it means.
When you donít have enough data, or data that is too heterogeneous, averages and medians do not tell you anything that is helpful.
An average is derived when you add all the data points together and divide by the number of data points. So, averages can be changed dramatically by one or a few very high or very low figures in the data. The average of 1,2,3,4,5,6,7,8,9,10 is 5.5 (add them to 55. Divide by 10 to get 5.5) That seems right; itís sort of in the middle of these numbers. But, if there is an outlier, high or low, look what happens: 1,2,3,4,5,6,7,8,9,10,74. (Adds up to 129. Divide by 11 equals 11.7.) That number, 11, does not reflect the vast majority of numbers in the series. Thatís why, in real estate, a single million-dollar sale in a modest town makes the average sale price figure for that town higher than it should be.
A median is derived when you choose the figure that is in the middle of the data set. Using the two examples above, the median is 5.5 in the first group (somewhat reasonable) and 6 in the other (also fairly accurate.) Median is a better figure since a few very high and very low data points donít mess it up so much when you have a larger group of numbers than the example above. But, median figures donít tell you enough about how high the highs are or how low the lows are. They also donít tell the concentration of property prices. Are there a lot at the high or low end?
You see derisive comments on this blog (and elsewhere) when pundits start quoting average or median house prices. It is generally justified. The problem is that if the data-set is too small or too varied, averages and medians do not give an accurate picture.
At best, a median price for all houses of a specific size in a specific town can give you a guide to what you may need to pay for a house like that in that town. But, a median price for all the houses in the town doesnít tell you much.
Year over year median or average prices can go up or down because all real estate is getting more expensive or cheaper. It can also go up because a town has a new development of expensive housing at the same time that modest housing prices are going down. You cannot know which is true by looking at median prices alone.
Got it? Thursday, I will write more about valuation systems.
The author is solely responsible for the content.