The Institute of Electrical and Electronics Engineers -- United States of America (IEEE-USA) promotes the career and public policy interests of more than 220,000 U.S. members of the IEEE, the world's largest technical professional society, with a total membership of more than 360,000 electrical, electronics and computer engineers and computer scientists in approximately 165 countries. For more than 30 years, IEEE-USA has been conducting, analyzing, and distributing a salary and fringe benefit survey of IEEE members in the United States. The aim of the project is to provide timely information on current and long-term trends related to the income, salary and benefits of electrotechnology professionals in the United States. The information is critical for accurate understanding of compensation practices in these occupations, including how those practices impact on individual professionals.

 

To conduct the salary surveys, invitations to participate are sent to randomly selected samples of IEEE-USA's non-student grade members. Responding survey participants are compared to the full membership to guard against possible response biases by region and membership grade; it would be feasible to weight responses to correct for such biases, but such steps have never been needed. Design specifications for the project call for general confidence, at the .95 level, that sample estimates will not vary from actual population parameters by more than plus or minus two percent. This standard has been easily met for major statistics generated by the surveys. Each result requires individual assessment -- for example, fewer cases are required when results approach the extremes of distributions, more when outcomes fall in the middle of the range of possibilities -- but as a general rule, results based on at least 1,500 cases meet these tests. In recent years the crucial data subset of engineers who are employed full time in their area of professional specialization have been nearly ten times this size, so the general findings of IEEE-USA's surveys about broad trends in compensation are quite precise. As in all surveys, results become somewhat less robust when a database is carved up into subgroups such as those with specific specialties, levels of experience, and kinds of academic degrees. To provide this kind of detail without sacrificing reliability and precision, IEEE-USA now makes increasing use of estimates generated from regression models. These have the great advantage of supporting highly detailed findings without major losses of statistical confidence. They also have the advantage of being "self-smoothing," that is, they reflect overall tendencies for the respondents as a whole and are not subject to the sampling anomalies that can distort results for small selected subgroups of cases.

 

IEEE-USA's initial compensation survey was conducted in early 1972 and yielded comprehensive information on salaries in 1971. The project was repeated in early 1975, providing results for 1974, and was repeated again in 1977, establishing a pattern of biennial studies that continued until 2003, when the survey was made annual. (In 2005, IEEE-USA plans to conduct an annual survey in the first quarter, with three quarterly updates thereafter.) These surveys have been the most detailed examinations of compensation in the engineering profession, and are among the most thorough, accurate, and rapidly produced studies of compensation of any kind. Comparisons of IEEE-USA's results have been made with many other sources of information on engineering salaries, including the annual employer surveys conducted by the Engineering Workforce Commission of the American Association of Engineering Societies, annual reader surveys conducted by EE Times, the annual surveys of registered professional engineers conducted by the National Society of Professional Engineers (NSPE), and income data from such federal agencies as the Bureau of Labor Statistics and the National Science Foundation. All of these comparisons confirm the validity and reliability of IEEE-USA's data. Differences in results exist among these various sources of intelligence on engineering pay, but all of those differences can be explained by variations in populations surveyed, timing of data collections, definitions of what kinds of income are to be tracked, and other characteristics of each particular project. For an example of an illustrative comparison of IEEE's data with another source, see "Sagging Salaries But More Bonuses," ENGINEERS Vol. 1, No. 3 (July 1995: Washington, D.C.: Engineering Workforce Commission of the American Association of Engineering Societies), pp. 1-9.

 

IEEE-USA's compensation surveys set high standards. In 1991 the society began to test the use of regression-based mathematical models as a means to generate much more robust, specific estimates of pay. This work has led to the web-based individual compensation analyzer in use today, a tool that can generate accurate, detailed estimates for complete ranges of base salaries for literally hundreds of thousands of different employment situations, allowing simultaneously for variations in experience, training, areas of specialization, employer characteristics, geographic locations, and other factors that affect compensation. Conventional tabulation-based treatments of salary data cannot yield such results. Internet-based data collection for the surveys was begun in 1999 and moved entirely to the web in 2001. Now further changes have been made. Distribution of results from the surveys, as well as collection of data, is now a web-based operation, and includes the capacity for extensive custom database queries for users of the surveys. Data collection has moved toward real-time tracking of compensation trends, with new information added quarterly. These systems solidify IEEE-USA's status as the most reliable and innovative source of information in the world on the compensation of technical professionals.