ePrivacy and GPDR Cookie Consent by Cookie Consent
The leading association
of public opinion and
survey research professionals
American Association for Public Opinion Research

It’s Election Season for the Country, and for AAPOR

by Dan Merkle, President
Dan_Merkle.jpgAAPOR has been busy this election season providing resources related to polling and the election and is in the middle of holding some of its own voting related to AAPOR’s governance and policies.

In September, AAPOR put on two webinars for journalists focusing on accurately interpreting and reporting pre-election polls. These well-attended webinars were taught by former AAPOR President David Dutwin. And this month AAPOR held another Reddit “Ask Me Anything” event as part of the Voting in America series. It covered a variety of issues concerning 2020 pre-election polling and was hosted by AAPOR members Camille Burge and Lydia Saad.

Also, AAPOR has appointed a Task Force on 2020 Pre-Election Polling to examine the performance of the 2020 pre-election polls. The Task Force is chaired by Josh Clinton and includes 19 academic experts and pollsters. Their first report, the “Performance of the Polls in the Democratic Primaries,” was recently released.

The report found that pre-election polls in the 2020 primaries correctly predicted the winning candidate 81% of the time, similar to the performance of primary polls in recent years. This was the case even though pollsters faced particular challenges this year including late-breaking events prior to both the South Carolina primary and Super Tuesday, as outlined in the report. For example, as a result of these events, only 61% of polls in Super Tuesday states predicted the winning candidate.

The report also found that the absolute error on the margin of the victory was slightly higher for the 2020 primaries than in recent primary contests. But again, this was mainly due to the events prior to the South Carolina primary and Super Tuesday. Among all polls conducted in the primaries, the average absolute error was 10 points on the margin of the victory. However, the average absolute error was only seven points for polls conducted in states other than South Carolina and for Super Tuesday. This is lower than previous years except for 2004 which also had an average error of seven points. The Task Force is now busy working on collecting and analyzing all of the publicly released polls conducted for the 2020 General Election.

AAPOR is also in the midst of a series of membership votes on our governance and policies. In August AAPOR members voted overwhelmingly (94% approve, 2% disapprove, 4% abstain) to approve AAPOR’s revised Conduct Policy. This policy outlines expectations, reporting, review procedures and possible sanctions for those who attend or participate in AAPOR events and activities.

As I write this, members are currently voting on proposed changes to AAPOR’s Bylaws. Every five years, AAPOR reviews its Bylaws to ensure they are up to date and adequately govern the structure and operation of the organization. Similarly AAPOR’s Code of Professional Ethics and Practices is also reviewed and updated every five years. Soon AAPOR members will be asked to comment on the proposed changes to the Code. After the conclusion of the comment period members will vote on the proposed changes to the Code. AAPOR Council greatly appreciates your input on these important documents. And please, don’t forget to vote!