If an appraiser is reviewing her data and finds an outlier, what should she do?

• A. Delete outlier
• B. Re-examine the population of data
• C. Collect more data
• D. Apply transformation

Answer: (B)

When you see an outlier, the first step is to investigate the data source and context to determine whether this point belongs in the same population or signals a different subgroup or data issue. Re-examining the population of data means checking the data's origin, collection method, units, time frame, and any potential entry errors. If the outlier is a genuine observation that reflects a distinct submarket or measurement condition, it helps explain why values diverge and may warrant separate analysis or segmentation rather than discarding it. Only after this vetting should you consider other actions like correcting errors, collecting more data to see if the pattern persists, or applying transformations. Deleting the outlier or transforming data without understanding its role in the population can bias conclusions, so the responsible course is to first re-examine the data population to determine the proper treatment.