The survey of government leaders is based on in-person interviews conducted under the direction of Princeton Survey Research Associates between October 1997 and February 1998. The presidential appointees and members of the Senior Executive Service who were contacted for interviews were randomly selected from the Summer 1997 edition of the Federal Yellow Book. Interviews were completed with 98 presidential appointees and 151 members of the Senior Executive Service. In addition, interviews were conducted with 81 members of Congress.
Every person in each of the three groups was sent a letter requesting their participation in the survey. Subsequently, the office of every person in the samples was contacted to schedule an interview. When interviews could not be scheduled as a result of the first call, offices were called back at least two additional times, if necessary, to try to schedule a time for the interview. However, interviews could not be conducted with every person in each sample, both because some leaders were unable to schedule interviews during the period the poll was being conducted, and because some leaders declined to participate. The inability to complete interviews with all of the people in each sample may produce some biases in the survey results, since participation rates may vary for different groups and these groups may vary on questions of substantive interest.
However, the final samples on which these survey results are based are in several respects representative of the full populations. For example, the party breakdown among members of Congress in the sample (58% Republicans and 42% Democrats) closely approximates actual partisan divisions in the House and Senate. Similarly, the percentage of presidential appointees and senior civil servants from each federal department is roughly proportional to the actual distribution across departments among Washington-based presidential appointees and senior civil servants.