- Joined
- Jun 27, 2017
- Professional Status
- Certified General Appraiser
- State
- California
Bias is inherent to being human, you only what you know and vice versa. I think this is why its critical that the appraiser be a student of the market and provide market based support for everything they do, from comp selection to adjustments. Providing data driven analysis is what can limit the bias. From what I've seen certain appraisers truly think that their opinion matters, I feel that your data matters. As in where did you develop that opinion, did it come in a dream or through analyzing raw data? Far too often I see bias opinion which when unsupported is garbage. I've even seen appraisers quote how many appraisals they have completed in the given year. Who cares? Ego stroking doesnt convince anyone of anything.
Read Hubbard. Some of the most important features of homes are QUALITATIVE. In other words, they are really intangibles. Being good, VERY GOOD, at valuing homes means knowing how to value intangibles. Not only do nearly ALL appraisers not know how to do this, even the so-called experts who write books for the AI, deserve dunce caps.
One of the stupidest things is the grading for Condition and Quality. C1-C6, Q1-Q6. STUPID. You know what mine are? - 0-99!!. Now why do you need so many levels - because when you get into to certain areas of the condition/quality vs price contribution curve, the slope can get very steep - and small changes in quality/condition can result in big changes in value. A good regression program can map these out in fine detail - but only with a more refined scale like 0-99. And, if you have been on this forum for long, you will have come across complaints form many appraisers about these rankings. They have to take up a place in the grid and you have to assign adjustments that are different for the same condition/quality value. It messes up your report big time. "Why did you do <this>?" "Why did you do <that>?"
Clarification: When I say "small changes" can make a big difference in value, I mean "small changes" in the relative ranking of properties on the basis of intangibles such as quality, condition, view and so on. We see this in the Fixers. Maybe only 1-2% of the homes in a neighborhood in a neighborhood of 40 year old homes are in such dire need of repair they would be considered "fixers" Within this 1-2% (of say a population of 300 homes) you might have some that need major reconstruction, others just updating that has never occurred over their lifetime (while the vast majority of homes in the neighborhood have had nearly continual updating over that period) and everything in between. Regression of the residuals can bring out the adjustments to make ... but understand it is based on percentage rank, percent better or worse. The subjective valuation of the appraisers comes into placing the subject within the ranking of comps produced by an ordering of the residuals scored by % better (or worse, take your pick).
Last edited: