Understanding the 2000 NBES Methodology

By Michael J. O'Neil, Ph.D.

The 2000 National Business Ethics survey was conducted by O'Neil Associates, Inc. for the Ethics Resource Center under the following conditions:

• The sample was selected via a Random Digit Dialed sample purchased from Survey Sampling, Inc. The main implication of Random Digit Dialing is that both listed and unlisted telephone numbers are included in the sample.
• We excluded from the survey persons who were not employed at least 20 hours per week in an organization and those who did not work in organizations with at least two employees. These exclusions were made because many of the survey questions would have been inapplicable to such persons.
• Interviews were conducted via Computer Assisted Telephone Interviewing (CATI) from our centralized telephone facility in Tempe, Arizona.
• A total of 1,500 interviews were completed with individuals in residential households in the contiguous 48 states. These interviews were completed between November 12, 1999 and February 14, 2000.

As a result, we achieved a random national sample of that population that excluded only the non-telephone population. The distribution of the sample was validated against a wide range of demographic attributes. The sampling error of a survey conducted under these circumstances is ± 2.5% or less at the 95% confidence level. Using the guidelines of the Council of American Survey Research Organizations (CASRO), the survey response rate was 51%

Throughout the entire project our relationship with the Ethics Resource Center was such that, at every juncture, when we presented the ERC with a set of methodological alternatives they consistently opted for the most comprehensive alternative-whether this be a large sample size, a rigorous sampling methodology (Random Digit Dialing) or other attributes. The result is a survey in which readers should have a high degree of confidence.