The decision of the Commission on Presidential Debates to use the combined survey results from five national polls to assess the viability of a candidacy as a condition for including him or her in the presidential debates raises critical questions. We believe that the Commission should establish criteria for including only comparable public polls in its determination. We also would like to know what the Commission means by "use the combined survey results." The National Council on Public Polls believes the Commission should make public its answers to these questions well in advance of using the results of public polls as a criterion for debate participation.
NCPP believes that for the polls to be comparable they should meet requirements established by the Commission. As a minimum, the polls should:
- Be conducted within a fixed period of time;
- Include the same segment of the voting population (all adults, or registered voters, or likely voters, defined in a similar manner);
- Be asked essentially the same question or questions in the same questionnaire context;
- Recalculate the candidate percentages after eliminating undecided voters or refusals, if it has not been done by the survey organization.
- Use scientific survey methodology.
Any substantial methodological or procedural differences among the five polls could call their credibility into question. We believe this might provoke criticism of the Commission’s decision.
Whether or not a candidate is included in the presidential debates is obviously an important decision. The National Council on Public Polls does not want unreliable or inappropriate poll results to play a role in that decision
For more information about this and other polling issues, contact the NCPP Polling Review Board Members.