The word "Average" in the improvements section - possible bias?

[RCA]

Average is an objective term. It's a statistical constant that represents a single value within a range of data that describes the whole. Objective terms are based on facts that are not influenced by personal beliefs or biases.

Average can be considered objective if you use it in reference to a set of data according to its statistical definition. But note in this context it can also be called a relative term, as it is relative to a specific population. However, the term is often used in ambiguous ways when describing features that don’t have an exact numerical measure. This often happens in appraisal because so many thing such as features are neither nor - they are in a grey area. In such cases “average” is definitely very subjective.

GSE appraisal standards (or guidelines as they call them) are dead poor at the core. Both Appraisal Institute and GSE will often tell you that the C1-C6 and Q1-Q6 ratings are “absolute” definitions. However if you look at the definitions of these ratings they are replete with undefined imprecise adjectives such as “many” and “significant.” Such people have told me that what they mean by “absolute” is that C4 in the Silicon Valley is C4 in Jackson, Mississippi. Well the problem is that is based on terms such as “many” and “significant” that really produce different results with boundary cases in these two quite different areas. Aside from the fact that UAD definitions are often useless for valuation, they are, generally speaking rubbish terms. They are garbage.

So much better to say: My market area is The City of Pacifica, CA and is represented by the MLS sales from 1/1/2015 to 8/1/2024, downloaded to file X. The Condition-Quality-Appeal Scores are 0.00 - 10.00, representing the percentage of properties in this data set or market area with lower scores. I get this score by ranking the properties by residual from lowest to highest and then assigning them a score that represents the percentage of properties lower in the ranking based on the premise that less appealing properties sell for less than predicted by the
MARS regression model and more appealing properties sell for more than predicted by the regression model.

 

[Tom D]

RCA. Yes, a lof of garbage discussed about words. But just looking at your 'condition-quality-appeal scores' worded term, it scream hidden racism at first glance. A hidden coded phrase.
Let alone understanding the process, you just went over the head of the general public. That's just my opinion being a white cracker appraiser.
Perception is enough to make you guilty of bias. Even when using science and math terms now.

 

[RCA]

RCA. Yes, a lof of garbage discussed about words. But just looking at your 'condition-quality-appeal scores' worded term, it scream hidden racism at first glance. A hidden coded phrase.
Let alone understanding the process, you just went over the head of the general public. That's just my opinion being a white cracker appraiser.
Perception is enough to make you guilty of bias. Even when using science and math terms now.

Sorry you aren’t reading carefully. Maybe you read several words and started assuming many things.

This CQA score is strictly mathematical and calculated by an R script. It is based on the residual. I.e. the difference between the MARS regression model predicted sale price and the actual net sale price.

 

[Tom D]

I understand, I wasn't talking about reality. I was saying the perception of that specific phrase/wording you used 'condition-quality-appeal scores' could be interpreted bias by the average civilian reader who saw it, let alone understand it.

Actually, i do some regression, but your mind is above mine in that field. I was just cherry picking a phrase out of your comment to file a bias claim against you. maybe you would like to pre settle it.
Maybe not any bias, but the perception to a public reader, they assume systemic bias. Race baiters don't care about math and statistics. just a thought about wording used in any comment by us.

 

[alebrewer]

The introduction of the C and Q ratings were most certainly a step forward in encouraging appraisers to quantify their opinions. What is not discussed, however, is that - possibly even more importantly to the GSE's - the introduction of the C and Q ratings allowed them to quantify condition and quality for their valuation algorithms. IOW - while there are those who espouse the notion that the GSE's 'just' did it to force objectivity from the appraiser, that was only one of the reasons for the introduction - and, again, possibly not even the most important reason. I personally believe the C & Q ratings to be much superior to the old GAF (good/average/fair) terms. It is, however, a double-edged sword. Those ratings also allow for more accurate automated valuations by CU and LCA - the end result of which will be the reduction (and eventual elimination) of human appraisers.

 

[djd09]

B. No person is allowed to influence or attempt to influence the development, reporting, result, orreview of an appraisal through coercion, extortion, collusion, compensation, inducement,intimidation, bribery, or any other manner including, but not limited to:

 

[RCA]

Well, I never used GAF or CQ ratings for regression. They are worthless - since most of your comps are likely in the same class, and often in borderline areas. The difference between C3 and C4 could be miniscule or it could be major. At the lower and upper ends of the price curve, market reaction, i.e. estimated sale price, can change drastically with relatively small relative percentage changes in condition and quality. It depends on the neighborhood and market. In Pacifica, you can walk down a street and find major differences in many features from one house the next, - because of age and updates. People with throw $1-1.5M to upgrade a house surrounded essentially by borderline fixers. Why? We are on the ocean. One house can have a fantastic view because it sets 10 feet higher in elevation than the house below it, it can also as a consequence be upgraded with expensive applieances, new floors and ceilings, new this and new that. It is just the way it is.

 

[alebrewer]

The fact that you don't use them, Bert, has no correlation whatsoever with whether the GSE's use them...

 

[RCA]

The fact that you don't use them, Bert, has no correlation whatsoever with whether the GSE's use them...

I said, "Use for regression." Also, I wasn't talking about any correlation. How can you be an objective appraiser when you dream up such stuff? Are you the kind of person who walks into a home, sees some kind of picture hanging on the wall, and starts to imagine all kinds of weird things? It would seem so.

 

[Terrel L. Shields]

The introduction of the C and Q ratings were most certainly a step forward in encouraging appraisers to quantify their opinions.

If quantify is the basis, then having say 5 categories suggests 100% 80% 60% 40% 20% 0% - that's a pretty broad range, especially when you are striving for 5% accuracy range... So, do we use C1-C20? Do we do like school grades and grade on the curve? Do we say 50% is failing and tighten the designations up to 10% within each range. I mean all this is subjective no matter how you slice the cake. At least Bert's suggestion is based upon a numerical calculus.

The difference between C3 and C4 could be miniscule or it could be major.

Which is exactly the problem.