How the Chronicle’s Pay Analysis Was Done

The Chronicle’s analysis of compensation paid to chief executives and all other workers is based on data

that nonprofit organizations provided on the informational returns they file with the Internal Revenue Service.

The Chronicle examined data provided by 7,474 groups that paid at least $1-million in wages and provided information on the number of employees on their payrolls. The data examined were for the 2003 tax year, the most recent year for which complete tax data are available; to analyze trends over five years, the study also included figures for the 1998 tax year.

The information was obtained from GuideStar, a nonprofit group in Williamsburg, Va., that has created an electronic database of information from the tax forms.

The Chronicle calculated the average wages at each organization by dividing the total wages paid by the number of employees listed on the tax form. However, because the tax returns do not identify whether employees work full time or part time, an average wage for full-time employees cannot be calculated.

The analysis assumes that the proportions of full-time and part-time workers at each organization remained relatively constant.

Hundreds of groups that showed large increases or decreases in either what they paid their chief executive or workers were contacted by The Chronicle to verify their numbers.

In many of those cases, the figures had been skewed by unusual circumstances, and those were dropped from the analysis. For example, some groups that showed large increases in their chief executives’ pay in 2003 did so because in 1998 they had hired a new leader and not paid that person for a full year.

Conversely, other organizations showed decreases in executive compensation because their top executives had left at some point in 2003, so that salary figure represented only a partial year’s pay.

Employee wages also were distorted in some cases.

Many organizations that provide job training, particularly for the disabled, are required by the IRS to count their clients as employees because they are paid small amounts while they receive training. In some instances, the majority of those listed as employees by such organizations are, in reality, clients of the organizations. Those groups were not considered in The Chronicle’s analysis of compensation trends.