Sampling of Assessments
The BOE conducts a sampling of assessments on a five-year cycle for the ten largest counties and cities and counties, and on either a random or as needed basis for the remaining counties.
The purpose is to review a representative sampling of the assessments making up the local assessment rolls, both secured and unsecured, to determine how effectively the Assessor is identifying those properties subject to revaluation and how well the Assessor is performing the valuation function. The primary use of the assessment sampling is to determine an Assessor's eligibility for the cost reimbursement authorized by section 75.60.
A representative random sampling is drawn from both the secured and unsecured local assessment rolls of the county or city and county. These assessments are stratified into 18 value strata (nine secured and nine unsecured.) From each stratum a random sampling is drawn for investigation, sufficient in size to reflect the assessment level within the county or city and county.
For purposes of analysis, after the sample is drawn, the items are identified and placed into one of the five assessment categories listed below:
- Base year properties
- Transferred properties
- New construction
- Non-Proposition 13 properties
- Unsecured properties
A simple nonstratified random sampling would cause the sample items to be concentrated in those areas with the largest number of properties and might not adequately represent all assessments of various types and values. Because a separate sample is drawn from each stratum, the number of sample items from each category is not in the same proportion to the number of assessments in each category. This method of sample selection causes the raw sample, that is, the "unexpanded" sample, to over-represent some assessment types and underrepresent others. "Expanding" the sample data eliminates this apparent distortion in the raw sampling; that is, the sample data in each stratum are multiplied by the ratio of the number of assessments in the particular stratum to the number of sample items selected from the stratum. Once the raw sampling data are expanded, the findings are proportional to the actual assessments on the assessment roll. Without this adjustment, the raw sampling would represent a distorted picture of the assessment practices. This expansion further converts the sampling results into a magnitude representative of the total assessed value in the county or city and county.
The results of the field investigations are reported to the Assessor, and conferences are held to review individual sample items whenever the Assessor disagrees with the conclusions. The results of the sample are then expanded as described above. The expanded results are summarized according to the five assessment categories and by property type and are incorporated into the published assessment practices survey report.
In addition, the results of the assessment sampling are used to determine whether an Assessor has achieved an average assessment level in the county of at least 95 percent of the assessment level required by statute; and whether the sum of all the differences between the BOE staff's appraisals and the Assessor's values (without regard to whether the differences are under assessments or overassessments), expanded statistically over the Assessor's entire roll, is less that 7.5 percent of the assessment level. A county must achieve both of these requirements to be eligible to receive reimbursement for administrative costs pertaining to the supplemental roll.