University of Michigan needs your feedback to better understand how readers are using openly available ebooks. You can help by taking a short, privacy-friendly survey.
Rationality of Irrationality: Political Determinants and Effects of Party Position Blurring
Citizens in democracies complain that political parties' positions on major issues are too ambiguous for them to confidently understand. Why is party position ambiguity so common? Are party positions ambiguous because political parties fail in forming clear policies or because they deliberately blur their positions? Rationality of Irrationality argues that political parties are motivated to strategically blur their position on an issue when they struggle with a certain disadvantage in the issue. Specifically, political parties present an ambiguous position when their own supporters are divided in their stances on the issue. A political party also blurs position stances when voters do not acknowledge that the party has the ability and integrity to solve problems related to the issue. Political parties blur their position in these cases because ambiguous party positions divert voters' attention from the issue. Voters support a political party whose policy positions on major issues are close to their own stances. However, voters cannot confidently and exactly estimate party positions on an issue when they are only ambiguous.
Fig. 2.1. Party Supporter Division and Issue Emphasis Source: EES (2009), EVS (2008), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. Party supporter division is measured with the standard deviation of party supporters’ responses to survey questions on each issue. Salience scores from the CHES (2010) are used for issue emphasis.
Fig. 2.2. Party Supporter Division and Position Blurring Source: EES (2009), EVS (2008), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. Party supporter division is measured with the standard deviation of party supporters’ responses to survey questions on each issue. The standard deviation (of experts’ estimations) measurement is used for position blurring.
Fig. 2.3. Issue Competence and Issue Emphasis Source: EES (2009), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. Issue competence of a political party is measured with the percentage of people who indicated the party as the best political party to handle the issue. Salience scores from CHES (2010) are used for issue emphasis.
Fig. 2.4. Issue Competence and Position Blurring Source: EES (2009), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. Issue competence of a political party is measured with the percentage of people who indicated the party as the best political party to handle the issue. The standard deviation (of experts’ estimations) measurement is used for position blurring.
Fig. 2.5. Issue Emphasis and Position Blurring when Disadvantaged (Economy) Source: EES (2009), EVS (2008), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. “Divided” indicates that the value of the party supporter division variable is above its median. “Not competent” means that the value of the issue competence variable is below its median. Salience scores from the CHES (2010) are used for issue emphasis, and the standard deviation (of experts’ estimations) measurement is used for position blurring.
Fig. 2.6. Issue Emphasis and Position Blurring when Disadvantaged (Immigration) Source: EES (2009), EVS (2008), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. “Divided” indicates that the value of the party supporter division variable is above its median. “Not competent” means that the value of the issue competence variable is below its median. Salience scores from the CHES (2010) are used for issue emphasis, and the standard deviation (of experts’ estimations) measurement is used for position blurring.
Fig. 2.7. Issue Emphasis and Position Blurring when Disadvantaged (Environment) Source: EES (2009), EVS (2008), CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom. “Divided” indicates that the value of the party supporter division variable is above its median. “Not competent” means that the value of the issue competence variable is below its median. Salience scores from the CHES (2010) are used for issue emphasis, and the standard deviation (of experts’ estimations) measurement is used for position blurring.
Fig. 3.1. Distribution of Experts’ Estimations on the Positions of the National Rally and the Union for a Popular Movement Regarding the Economy Source: CHES (2010) (econlr, “Party position in terms of its ideological stance on economic issues”). The party position variable ranges from 0 (extreme left) to 10 (extreme right). Note: The means of experts’ estimations are 6.5 (the National Rally) and 5.6 (the Union for a Popular Movement), respectively. The difference of the means is not statistically significant at the 0.05 level.
Fig. 3.2. Position-Blurring Measurements from Expert Survey Source: CHES (2010). Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom. While a greater value of standard deviation indicates that the distribution of experts’ estimations is more dispersed (i.e., party positions are more ambiguous), a greater value of unimodality means that the distribution is more unimodal (i.e., party positions are less ambiguous).
Fig. 3.3. Position-Blurring Measurements from Expert Survey and Party Manifesto Source: CHES (2010) (leftright, “Party position in terms of its overall ideological stance”) and MARPOR (Rile, “Right-left position of party”) Note: Data description is presented in the appendix. Each filled circle represents a political party. Included are political parties in the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom. Manifesto data are chosen for election years that are closest to the year 2010 in each country. Both of the position-blurring measurements measure position blurring on overall left/right positions.
Fig. 4.1. Distribution of Western European Social Democratic Party Positions Source: MARPOR for manifesto (Rile); CHES (2006) for expert survey (lrgen); EES (2009) for voters’ estimations (the variable of q47). Note: Party positions from different sources are standardized into the scale from 0 (left-wing) to 10 (right-wing). Included are the following countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden, and the United Kingdom.
Fig. 4.2. Predicted Probabilities of Uncertainty in Party Position Estimation Note: Circles and diamonds indicate predicted probabilities, and lines show their 95 percent confidence intervals. The predicted probabilities and confidence intervals are calculated with results in model 2 in table 4.1. “When blurred” indicates the 90th percentile value of the position-blurring variable, and “When not blurred” indicates its 10th percentile value.
Fig. 4.3. Position Blurring and the Inaccuracy of Party Position Estimation Note: Circles indicate coefficients, and vertical lines show their 95 percent confidence intervals. The coefficients and standard errors are calculated with results in model 2 in table 4.2.
Fig. 4.4. Position Blurring and Issue Taking Note: Solid lines are coefficients, and shaded areas indicate their 95 percent confidence intervals. The coefficients and standard errors are calculated with results in models 4 to 6 in table 4.3.
Fig. 4.5. Predicted Probabilities of Voting Likelihood Note: Circles and diamonds indicate predicted probabilities, and lines show their 95 percent confidence intervals. The predicted probabilities and confidence intervals are calculated with results in model 2 in table 4.4. The probability of “probable to vote” indicates a probability that a person’s answer for the voting likelihood question is 9 on a 1 to 10 scale. “When blurred” and “When not blurred” represent the 90th and the 10th percentile values of the position-blurring variables, respectively.
Fig. 5.1. Party Family and Party Supporter Division Source: EES (2009) for the economy and immigration, and EVS (2008) for the natural environment. Note: Party supporter division is measured with the standard deviation of party supporters’ opinion on each issue. Survey questions for each issue can be found in the appendix. All of the survey responses are normalized to the scale of 1 through 5 for the comparability of standard deviations between issues. Mainstream parties include social democratic, liberal, conservative, and Christian democratic parties, and niche parties comprise radical right-wing parties and ecology parties.
Fig. 5.2. Party Family and Issue Competence Source: EES (2009). Note: Issue competence is measured with the average percentage of survey respondents who indicated a political party belonging to each party family as the best political party at dealing with an issue. Mainstream parties include social democratic, liberal, conservative, and Christian democratic parties, and niche parties comprise radical right-wing parties and ecology parties.
Fig. 5.3. Party Family and Issue Emphasis Source: CHES (2010). Note: Salience scores from CHES (2010) are used for issue emphasis. Mainstream parties include social democratic, liberal, conservative, and Christian democratic parties, and niche parties comprise radical right-wing parties and ecology parties.
Fig. 5.4. Party Family and Position Blurring Source: CHES (2010). Note: Position blurring is measured with the standard deviation of experts’ party position estimations regarding each issue. Mainstream parties include social democratic, liberal, conservative, and Christian democratic parties, and niche parties comprise radical right-wing parties and ecology parties.
x
This site requires cookies to function correctly.