Hypothetical Claim

​

​

What the People Would Think Under Good Conditions For Thinking About the Issue

Public Opinion: Three Problems

• Rational Ignorance

​

• Phantom Opinions

​

• Selectivity of Sources

Key Components

• Balanced information

​

• Small group deliberation

​

• Expert Q & As

​

• Random sampling/representative samples

​

• Control groups who do not deliberate

​

Is Deliberation an Antidote for Extreme Partisan Polarization?

26 of the 49 Policy Proposals Exhibit “Extreme Partisan Polarization”

​

Two Criteria for inclusion:

​

1. A Majority of the Members of Each Party Take Positions on Opposite Sides of the Scale

​

• At least 15% of the Members of Each Party Take the Most Extreme Position on opposite sides of the scale (0 or 10 on the 0 to 10 scales)

​

​

Two Kinds of Polarization

1. Issue-Based Polarization

​

​

2. Affective Polarization

Our Question: Is Deliberation an Antidote to Both?

​

​

Research Question 1: On issues of extreme partisan polarization, will cross-party dialogue lead members of each party to move toward or away from the mean position of the other party?

Why We Might Expect Unstructured Dialogue to produce More Polarization on the Issues

​

1. Motivated reasoning that is “directional” or partisan rather than “accuracy” based (Kunda 1990), Taber and Lodge (2006), Druckman (2012) will produce change in the direction initially preferred by that party

​

• “Imbalance in the argument pool” (fostered by selective exposure and filter bubbles) will produce change in the direction initially favored by most members Sunstein (2002, 2009), Sunstein and Hastie 2013.

​

• Social comparison effect Asch (1955), Sunstein (2002, 2009).

​

​

​

Why We Might Expect Structured Deliberation (Deliberative Polling) to Produce Less Polarization on the Issues

​

1. Motivated Reasoning that is Accuracy Based rather than Directional (or Partisan)

​

• Balance in the Argument Pool

​

• Confidential questionnaires (secret ballots) which limit the Social Comparison Effect

​

​

�Research Question 2: On issues of extreme partisan polarization, will cross party dialogue lead the most extreme members of each party to move toward or away from the most extreme members of the other party?�

​

Same Dynamics for those taking the most extreme positions on the 0 to 10 scales.

​

For unstructured dialogue, we posit more polarization.

​

For structured deliberation (as in the Deliberative Poll) we posit less polarization

​

Research Question 3: After deliberating on issues of extreme partisan polarization, will cross party dialogue increase or decrease affective polarization?

�

For unstructured discussion, we expect cross party dialogue may trigger the dynamics of in-group/out-group identity producing increased hostility and hence negative affect toward the outgroup. It would increase affective polarization.

​

For structured deliberation (Deliberative Polling) we expect that cross party dialogue will decrease polarization because of intergroup contact theory (contact produces liking).

​

The structured design of the Deliberative Poll fits the four criteria of Allport 1954. See also Pettigrew 1988; Pettigrew and Tropp 2006; Hewstone and Swart 2011.

​

​

Deliberative Polling is a structured design that

1. Focuses participants on common goals
2. Emphasizes a cooperative environment
3. Fosters equal status
4. Provides institutional support for the contact

​

Allport hypothesized that these four factors would work best together in a “structured design.”

​

What Happened?

��������� ��

​

​

Percentage of Republicans before discussion with these Views on Immigration

A selection of positions that were most altered by discussions with people of other viewpoints.

��������� ��

​

​

Percentage of Democrats  before discussion with these views on Economy

Democrats lost enthusiasm for some potentially costly government programs.

Summary of Changes for Democrats and Republicans on the 26 Polarized Issues

Affective Polarization Declines with Deliberation