Appraising is equal parts art and science many say, and there is some truth to it. It is common practice to round figures in an appraisal in order to represent that the appraisal is not an exact science, but, how much rounding is too much?
Consider for a moment the concessions line of the adjustment in the most commonly used tool by appraisers, the 1004. If the market informs you that a dollar-for-dollar adjustment would be correct, would you then round that total?
Or perhaps the market conditions line, given the below data from a plan of townhomes, all built around the same time, and with little difference in styles. The linear regression shows a market increase of 1.44% (rounded) per year over the last three years. IF you were to apply a per diem adjustment of 1.5%, would you then round that figure as well? How many roundings is enough? At what point does it stop?
Or in terms of land adjustments, the non-linear nature of land adjustments should be obvious (this is county-wide with no refinement in terms of a subject property), but above this, when applying a calculation of land values, after coming up with a figure, would you then round to the nearest 100, 1000? Why?
Perhaps rounding should be reserved for less precise adjustments, based on more subjective tools such as qualitative analysis, or tools that are less reliable, but when you have the data to support precision, why would an appraiser round?
If our bottom line rounded market value opinion, is the result of rounding, of a rounding of a rounding, added to some other roundings... what is the final figure? It would certainly be difficult to call something of that definition a professional opinion of value. Of course there are more complex assignments that require more subjective tools, but when you can be precise, be precise. I find that it makes the less precise parts of the task far easier.