Using the Executive Compensation Assessor > Frequently Asked Questions > 12 Common XA Questions for ERI Expert Witness Voir Dire

In addition to the links and below, please refer to the Executive Compensation Assessor (XA) Methodology.

1. Is XA a survey?

2. Which is the more normative figure: average or median?

When dealing with employee compensation, in most cases, it will be the median. If the salaries paid to employees followed a normal distribution, then the average (mean) and the median would yield the same values, and both would be good normative figures. However, salaries rarely follow a normal distribution as there can be abnormally high salaries paid to a few employees. Averages are affected by outliers, so one abnormally high number will pull up an average even if it is non-representative of the norm and far above the mode. A median is the middle value in a rank-ordered-sequence; it is the central number, unaffected by outliers. The middle remains the middle, regardless of how high the high is or how low the low is.

For related information, please see Means/Medians.

3. What is the difference in cash compensation between public and private companies?

None. Once the predictive factors of size and industry (and profit) have been equalized, no statistically significant difference has been found in cash pay amounts for the same work by private or public companies. This should be no surprise, since both public and private employers share the same labor market and compete for the same executive talents in the same areas. They also frequently have the same members as board directors and employ the same compensation consultants who recommend the same best/newest reward practices. The non-cash reward practices of private enterprises, however, do follow different patterns.

For related information please refer to the Data Plots section of the Methodology and Assessor Series FAQ #34.

4. Does cost of living determine pay?

Pay is employee income, while cost of living measures employee expenses. Family income depends on the amount of combined compensation, savings, investment income and other income sources. Family living costs, whether they be simple renters’ living costs (as measured by the Geographic Assessor) or the more complex lifestyle living costs (as measured by the Relocation Assessor) vary by what they spend and can be strongly affected by tax considerations. It should also be noted that most people do not live where they work. Those who work in Manhattan, for example, usually live elsewhere and commute from their homes, which are located in areas where living costs are typically much lower and where the purchasing power of their income dollars is greater. Even in small communities, employees rarely live within walking distance of their place of employment.

For related information, please see Assessor Series FAQ #3.

5. Are variance statistics provided for all XA survey sources?

The variance statistics are extremely conservative because they are based only on the SEC observations (the only elements that can be accurately counted without risk of data redundancy and/or double-counting).

For related information, please see Data Background FAQ #1.

6. How many incumbents hold each job in XA?

It is impossible to derive an accurate count of incumbent observations. One factor is redundant and/or overlapping data sources. XA compiles data from the SEC Edgar database and multiple survey sources. Another factor is timing; there can be huge gaps between the date of data collection, the date of survey publication and summation and XA's "Data as of" date.

By viewing the Survey Population, you can review ERI's estimates of the number of surveyed job incumbents in the area selected (covering all industries). The Variance Statistics feature displays the conservative but highly precise count of each SEC-reported occurrence of a job incumbent at a public corporation. Please note that many individuals in such cited observations may have changed jobs, areas, employers, industries and salaries since the original surveys were conducted.

For related information please refer to the Data Plots section of the Methodology, Assessor Series FAQ #6 and Using XA FAQ #5.

7. Do the lines on the graph correspond to the dots?

Please refer to Using XA FAQ #4. Also see the Data Plots section of the Methodology.

8. Why doesn’t XA report the 1st or 99th percentiles?

Please refer to the Help topic about Percentiles.

9. Why is the Maximum Reasonable Compensation figure seen in the Proxy Analysis screen view different from the highest percentile you report in the Individual Profile screen view?

One was set by negotiation with the Internal Revenue Service at 2.0 standard errors above the weighted average (approximately from the 97th to 99th percentile), and the other is the highest range of pay (approximately the 90th percentile) reported in surveys.

Please see the Help topic on Maximum Reasonable Compensation.

10. Where can one find proof that XA meets expert witness criteria?

Please refer to the ERI Statement as to the Relevance and Reliability of Data section of the Methodology.

11. How often is XA updated?

Twice each quarter or approximately every six weeks. For related information, please see Assessor Series FAQ #23.

12. How does changing the Planning Date affect the database?

It changes the adjustment percentage applied to the most current consensus rates as of the date of publication of the relevant Assessor.

Please see the Help topic Salary Planning Date and Assessor Series FAQ #18. Also see FAQ #7 for information on setting the salary planning date for the top six executives.