Disentangling the Direction of Causality Between Competence Issue Ownership and Party Preference

Disentangling the Direction of Causality Between Competence Issue Ownership and Party Preference Introduction Issue perceptions, particularly issue ownership, are increasingly used to understand voters’ electoral choices (Lefevere, Tresch, & Walgrave, 2015). Issue ownership refers to the idea that voters perceive some parties as better able to “handle” certain issues. Consequently, parties can be considered as “owning” issues they have a good reputation on (Budge & Farlie, 1983; Petrocik, 1989, 1996). These reputations affect voters’ electoral behavior: voters are more likely to vote for a party owning the issues of importance to them (Bélanger & Meguid, 2008; Lachat, 2014; Petrocik, 1996; Walgrave, Lefevere, & Tresch, 2012). However, voters’ issue ownership perceptions may be endogenous to their party preference (Walgrave, Tresch, & Lefevere, 2015). This criticism is most often raised against the competence dimension of issue ownership, which refers to parties’ competence in handling an issue. For example, Kuechler (1991) found that competence evaluations reflect vote choices to a high degree, which suggests that the direction of causality between competence issue ownership and party preference could be reversed (van der Brug, 2004). Green and Jennings (2012) find that competence evaluations Granger cause party preference. Yet, most of these studies rely on cross-sectional data, which limit their ability for causal inferences, while others rely on aggregate-level data, making it impossible to track changes at the individual level. Panel data offer the opportunity to assess whether competence issue ownership causes party preference, or whether the reverse causal relationship is dominant. We present analyses based on three-wave panel data collected among a sample of Dutch citizens, which polled respondents for their party preference and competence issue ownership perceptions on the economy in each wave. Our findings suggest that voters’ economic issue ownership is driven more by their vote choices, than vice versa. The Endogeneity of Issue Ownership A variety of studies assert that voters’ issue perceptions affect vote choices. Spatial models of electoral choice suggest that voters tend to take parties’ issue positions into account when they cast their ballot (Lewis & King, 1999). Yet, scholars increasingly assert that positions are not the only determinant at play (Green & Hobolt, 2008): parties’ issue reputations also matter. This idea is discussed in multiple theories of voting: Riker’s (1993) dominance theory, the valence theory of voting (Sanders, Clarke, Stewart, & Whiteley, 2011), and issue ownership theory (Budge & Farlie, 1983; Petrocik, 1989, 1996). Empirical studies have shown that such issue evaluations affect voters’ electoral choices—both in the aggregate and at the individual level (Sanders et al., 2011; Walgrave et al., 2015). Yet, the persistent empirical support for the electoral impact of voters’ issue evaluations has been matched by an equally persistent critique on their validity as a determinant of electoral choice. The criticism, in its most basic form, is the possibility of reverse causation. Issue evaluations might well be a consequence of party preference, rather than the other way around. This critique relates to a broad discussion regarding the impact of party preferences on voters’ political attitudes. The roots of this discussion lie with Campbell, Converse, Miller, and Stokes’ (1960) argument that party identification acts as a perceptual screen through which voters perceive the world. Though distinct from party preference, this work on party identification suggests that party preference can affect voter’s political attitudes: partisanship affects which information voters attend to (Garrett, 2009; Stroud, 2008) and how they learn from that information (Gerber & Huber, 2010; Jerit & Barabas, 2012). Yet, a parallel line of literature, based on Downs (1957) and Fiorina (1981), conceptualizes party preference as a running tally of voters’ past experiences, suggesting that attitudes drive party preference (Bartels, 2002). Applied to voters’ issue evaluations, there is ongoing debate regarding the extent to which the above critique holds true. In 2006, Evans and Andersen raised concerns about the impact of party preferences on voters’ economic perceptions, which causes scholars to overstate the impact of such perceptions on political preferences (but see Lewis-Beck, 2006). More specifically, party issue evaluations are part of the valence model of electoral choice (Clarke, Sanders, Stewart, & Whiteley, 2004, 2009): using a panel design, Evans and Chzhen tested whether valence evaluations cause party preference, or vice versa. They concluded that: “assessments of government performance are, to a significant degree, a product of party preference and cannot therefore be used to explain that preference” (Evans & Chzhen, 2016a, p. 213; but see Whiteley, Clarke, Sanders, & Stuart, 2016; Evans & Chzhen, 2016b). Yet, as far as we can tell similar panel data analyses do not exist for measures of issue ownership. We focus here only on the competence dimension of issue ownership, which has been most suspect of being caused by party preference (van der Brug, 2004; Stubager & Slothuus, 2013). It is important to assert whether reverse causation also plagues issue ownership measures specifically because the findings regarding related measures of valence may, or may not, apply to a similar degree to measures of competence issue ownership. To start, parties’ valence evaluations are based on a variety of performance assessments—including both issue and leader affect (Evans & Chzhen, 2016a, b; Whiteley et al., 2016), whereas competence issue ownership focuses solely on parties’ issue reputations. Hence, it is not clear to what extent we can extrapolate these findings based on a mix of attitudes to attitudes focused solely on issues. Indeed, based on cognitive interviews, Wagner and Zeglovits (2014) suggest that respondents’ answers to issue ownership questions seem less influenced by their overall party preference. Moreover, valence measures do not require voters to choose one party over another as is common in issue ownership measures—a party owns an issue if it handles the issue better than its competitors. In contrast, valence is about the evaluation of a single political actor as doing either a good or bad job in handling an issue (Clarke et al., 2004). That said, the empirical evidence on competence issue ownership is at least suggestive of potential endogeneity. Apart from the fact that various scholars have raised concerns regarding reverse causation (van der Brug, 2004; Wagner & Zeglovits, 2014), various question wording experiments demonstrate that respondents’ competence issue ownership perceptions correlate to their party preferences (Lefevere, Walgrave, Stubager, & Tresch, 2016; Walgrave, Van Camp, Lefevere, & Tresch, 2016). Finally, Green and Jennings (2012) show that assessments of issue competence Granger cause party preferences for some parties at an aggregate level. Therefore, the extant literature leads us to formulate two competing hypotheses. On the one hand, issue ownership theory claims that respondents’ perceptions of competence issue ownership determine their party preferences. Thus, our first hypothesis is: H1: Voters’ competence issue ownership perceptions positively affect voters’ party preferences. On the other hand, a number of studies have pointed to the potential endogeneity of the competence dimension of issue ownership (Bélanger & Meguid, 2008; Lefevere et al., 2016; Wagner & Zeglovits, 2014; Walgrave et al., 2016). Hence, we posit that: H2: Voters’ party preferences positively affect voters’ competence issue ownership perceptions. Finally, we expect that while voters’ competence issue ownership perceptions and party preferences have a reciprocal causal relationship (H1 and H2), in line with the assumption underlying most of the empirical studies, we maintain that the dominant causal relationship is from competence issue ownership perceptions to party preferences, rather than the other way around. H3: The dominant causal relationship is from competence issue ownership perceptions to party preference, rather than vice versa. Methods To empirically test our hypotheses, we rely on panel survey evidence testing the nature of the causal relationship between competence issue ownership and party preference. More specifically, we use the results from a three-wave online panel survey conducted by I&O Research, an ISO-certified research company, indicating that it meets a set of quality criteria for data collection and quality set in liaison with among others the World Association of Public Opinion Research (WAPOR). The survey was fielded in the first half of 2015, among a sample drawn from I&O’s ongoing panel of respondents, and aims to be representative for the Dutch voting population. Potential participants are for the most part selected based on random samples from (local) population registers and recruited actively, without the possibility of self-registration. The surveys were held within the context of a larger project that focuses on the effects of media coverage on economic and political perceptions, attitudes, and behavior in the aftermath of the large economic crisis that hit most European Union countries. The survey contained a wide range of questions relating to media exposure, the economy, politics, and consumer behavior. The time lag between each wave was 10 weeks, and dates of fielding were February 23, 2015 (Wave 1), April 20, 2015 (Wave 2), and June 15, 2015 (Wave 3). For the first wave, 22,879 people were invited to participate and 9,112 started the questionnaire (response rate: 39.8%). Of those who started the questionnaire, 6,386 completed the survey (completion rate: 70.1%). Only respondents who completed this survey wave were invited to participate in the second wave. In Wave 2, 4,301 respondents completed the questionnaire (completion rate: 69.0%). From those respondents, 3,270 also completed the third wave (completion rate: 77.0%). In this analysis, we only include those respondents that completed all three waves. Despite the intention to create a representative sample, the final sample deviates from the general Dutch population in that it overrepresents older (M = 61, 45, SD = 11.08), male (66%), and highly educated respondents (50.9% obtained a university degree). These numbers do not differ substantially from those of all participants that participated in Wave 1. The Netherlands is a parliamentary democracy with a system of proportional representation and a wide range of political parties being represented in Dutch parliament: 10 parties after the 2012 parliamentary elections (Lijphart, 1999). It is a good case to test our hypotheses, as electoral volatility in the Netherlands is relatively high—and hence respondents’ party preferences should fluctuate over time (Mair, 2008). Moreover, the fragmented nature of the party system means that issue ownership of parties is likely to be contested and should fluctuate over time (Kleinnijenhuis & Walter, 2014). As a measurement of party preference, we use van der Eijk et al.’s propensity to vote question: in each wave, respondents were asked for all parties (11) that got elected in Dutch parliament during the last parliamentary elections1 to indicate the likelihood they would ever vote for that party on an 11-point scale ranging from 1 (certainly not) to 11 (certainly) (van der Eijk, van der Brug, & Franklin, 2006). This allows us to track minor changes in party preference. To measure issue ownership, we use the formulation used by Walgrave et al. (2009): respondents were asked to indicate how capable they considered each of the 11 political parties to be in dealing with the economy and economic issues. Again, an 11-point scale was used, ranging from 1 (totally incapable) to 11 (very much capable). The questions on party preference were asked consistently before those on competence issue ownership. We cannot assert whether this has affected the results, but because of the fact that a number of other questions were always included in between the party preference and issue ownership question blocks, we have no reason to assume that the effect of question ordering had a large impact here. We measured issue ownership only for a single issue—the economy. Though it would be preferable to assess our hypotheses over different issues, and have some baseline to compare with, the survey data only included issue ownership preferences for this single issue. That said, the economy is an appropriate issue to test our hypotheses: the economy is a typical performance issue on which parties can actively perform for better or worse—so we would expect fluctuations in voter’s assessments of parties over time (Petrocik, 1996). Indeed, ample studies on economic voting suggest that voters’ economic perceptions have tangible electoral consequences (Lewis-Beck & Stegmaier, 2000). Whereas for many issues voters have little evidence of party performance, evidence on economic performance is widely available and reported on in the mass media (Boomgaarden, van Spanje, Vliegenthart, & de Vreese, 2011). As such, especially for the economy, we would expect issue reputations to fluctuate, and to affect party preferences. Moreover, many voters consider the economy to be an important issue. In each of the three waves, an open question was asked about the most important problem the country was facing today. The answers were categorized in a range of policy topics, and while the worst part of the economic recession had passed in 2015, still 67.8% of the respondents indicated at least one time an economic topic as the most important problem—by far the most frequently mentioned one. Table 1 provides more detailed descriptive statistics for both variables, demonstrating that there exists considerable variation across parties, and to a lesser extent, waves. Table 1 Descriptive Statistics Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Note. PVV = Partij voor de Vrijheid. Table 1 Descriptive Statistics Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Note. PVV = Partij voor de Vrijheid. Our unit of analysis is a combination of respondent, party, and wave. We estimate two models: one model that predicts party preference based on party preference in the previous wave and issue ownership in the previous wave; and one model that predicts issue ownership based on issue ownership in the previous wave and party preference in the previous wave. Thus, all our regression models include a lagged dependent variable, which is considered an appropriate analytical approach to capture change in the dependent variable, while also accounting for initial different levels across individuals. Our models do not include any respondent characteristics as control variables, as there is no reason to assume that they would systematically affect our dependent variables irrespective of the party considered. To account for party-level heterogeneity (different levels of party support or attributed issue competence that are not captured by the independent variables in the models), we do include party fixed effects in the models. We use clustered standard errors on the respondent level. To make coefficients comparable, we use standardized variables. The total number of observations is 71,280 (3,240 respondents × 2 waves × 11 parties). To check the robustness of our findings, we also conducted a cross-lagged model using structural equation models using the same variables, and tested the regression model for individual parties as well. Results Table 2 presents the results from our analysis. Both our first and second hypotheses are confirmed: Model 1 shows that perceptions in issue ownership affect party preferences in the subsequent wave, controlling for the previous party preference (H1). The effect is, as expected, positive. Yet, the reverse relationship is also significant. If a party is preferred more, this causes upward shifts in competence on the economic issue (H2). If we compare the coefficients from both models, we have to reject H3. The standardized coefficient of the effect of party preference on issue ownership is considerably larger (.216 compared with .127). The difference is .089 and statistically highly significant (SE = 0.006, p < .001; for calculation, see Gelman & Stern, 2006). This is also clear when we compare the change in explained variance for both instances. While issue ownership adds 0.9% to the explained variance of party preferences, party preferences add >3 times as much (2.8%) to the explained variance of issue ownership. Table 2 Explaining Party Preference and Issue Ownership Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Note. N = 71,280; * p < .0001; standard errors are clustered per respondent. Fixed effects for parties are included but not displayed. †Compared with a model that only includes the lagged dependent variable. Table 2 Explaining Party Preference and Issue Ownership Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Note. N = 71,280; * p < .0001; standard errors are clustered per respondent. Fixed effects for parties are included but not displayed. †Compared with a model that only includes the lagged dependent variable. Our robustness checks confirm the findings described above. The resulting cross-lagged model using structural equation modeling with party fixed effects has a good model fit (χ2 = 3.946, df = 3, p = .267, Root Mean Square Error of Approximation (RMSEA)=.002, Comparative Fit Index (CFI) = 1.000) and shows similar results as the regression analysis. Both effects are significant: the impact of issue ownership on party preference has a standardized coefficient of 0.125, and the impact of party preference on issue ownership of 0.213. A χ2 test suggests that the null hypothesis of nondifference between coefficients clearly has to be rejected (χ2 = 254.49, df = 1, p < .0001). The effect of issue ownership on party preference is thus significantly smaller than the effect of party preference on issue ownership. H3 is rejected. The party-level analyses, reported in Table 3, confirm this overall finding. Regression analyses for each of the 11 parties show that in all instances, the standardized coefficient for the effect of party preference on issue ownership is larger than the standardized coefficient for the reverse relationship, and the difference between the coefficients is significant in all but two cases (the small Christian party ChristenUnie and the social–liberal D66). The largest difference (and also the largest effect of party preference on issue ownership) exists for the populist right party Partij voor de Vrijheid (PVV) of Geert Wilders. Apparently, economic issue ownership perceptions have a more limited influence on changes in preferences for this party, but a preference for this party does result in significantly higher perceptions of its competence on the economic issue. Table 3 Relationships Between Party Preference and Issue Ownership per Party Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * * p < .0001; standard errors are clustered per respondent. †Indicates significance of difference between the two coefficients. PVV = Partij voor de Vrijheid. Table 3 Relationships Between Party Preference and Issue Ownership per Party Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * * p < .0001; standard errors are clustered per respondent. †Indicates significance of difference between the two coefficients. PVV = Partij voor de Vrijheid. Conclusion and Discussion We analyzed the extent to which Dutch voters’ competence issue ownership perceptions of the economy determined their party preferences, and vice versa. Though issue ownership perceptions are increasingly used to explain vote choices (Lefevere et al., 2015), scholars have pointed to the potential endogeneity of especially competence issue ownership and party preferences (van der Brug, 2004; Wagner & Zeglovits, 2014). Our findings, though limited in nature, will not diminish these criticisms. The “good” news is that we find that competence issue ownership perceptions matter for party preferences. The “bad” is that the reversed causal relationship from party preferences to competence issue ownership perceptions turns out to be much stronger. This finding is in line with recent work by Evans and Chzhen (2016a), who also point out that assessments of competence seem to be driven by, more than being a determinant of, party preferences. Apparently, people update their perceptions about specific ideas about the competence of parties to deal with specific issues to a considerable extent after they decide on a general party preference. That said, our data is not without shortcomings, and we need to add three important caveats to our conclusions. First, we only considered a single issue—the economy—when it came to ownership perceptions. This limitation may be why we find that the reverse causal relationship—from party preference to issue ownership—is dominant: perhaps party preferences themselves are shaped by the concurrent impact of multiple issues, rather than the single issue we studied here. Moreover, the fact that no party held clear ownership over the issue may affect the result, as evidence suggests that stronger issue ownership (in the aggregate) may be more resilient to change (Walgrave & Lefevere, 2017). Therefore, future research should consider a wider variety of issues and test whether the findings hold for other issues as well. That said, the importance of the economic issue in the electoral process (Dassonneville & Hooghe, 2015), as well as the fact that it was mentioned by more than two-third of the respondents in our survey when asked about the most important problem the country was facing, do suggest that the issue is an important one. Moreover, competence evaluations on the economy should be less prone to conditioning by prior party preferences. Voters have straightforward indicators (e.g., changes in the gross domestic product) that allow them to assess how the economy is doing—and thus to assess the track record of the parties on it. Yet, we find evidence that party preferences do affect these perceptions. For more complex issues where clear indicators of performance are absent, we would expect that the conditioning effect of party preference would be even stronger. These factors make us confident that findings for other issues should lead to similar results. A second limitation is the representativeness of our sample, which over represents male, older, and highly educated people. Yet, we do not see any obvious way that these background characteristics might fundamentally alter the relationships we examined. Given the crystal clear picture that arises and the considerable effects we find, we are confident that results are generalizable, at least to the wider Dutch population. Moreover, the Dutch party system is highly fragmented, and most voters have several alternative parties to choose from. We would expect that the conditioning impact of party preference on issue ownership would be at least as strong in systems with less parties. Third, despite the panel nature of our data, experimental designs might be even better suited to tease out the two-way causality. For example, prior work has shown that the impact of parties’ media performances is moderated by party preference (Walgrave, Lefevere, & Tresch, 2014). A modified version of such an approach, triggering (moderated) shifts in issue ownership (and party preference in turn), could then use repeated measures to assert in which direction the influence is strongest. Our findings add to a growing list of studies that urge scholars to use competence evaluations with caution: we find that competence issue ownership perceptions of the economy were caused by, rather than causing, party preferences. This does not automatically invalidate their use as a determinant of electoral choice—we find that issue ownership affects party preference, controlling for lagged party preference—but it does show that perceptions themselves are subject to partisan conditioning, which is troublesome. Recent work on the associative dimension of issue ownership has demonstrated that this dimension is less strongly correlated to party preference: we urge future research to repeat our approach for this associative dimension, as it might prove to be more resilient to endogeneity. Given the potential issues of endogeneity, scholars should also examine findings regarding the impact of different question wordings: new formulations might trigger respondents to consider ‘pure’ competence, rather than an amalgam of competence and party preference. In sum, while competence issue ownership seems subject to similar issues that plague measures of party competence more broadly, we urge scholars to continue work on the concept of issue ownership, as its burgeoning research agenda suggests various avenues to move forward. Rens Vliegenthart is professor for Media and Society at the Amsterdam School of Communication Research (ASCoR) at the University of Amsterdam. His research interests include media–politics interactions, media-effects, and comparative politics. Jonas Lefevere is Assistant Professor of Communications at Vesalius College Brussels. His research focuses on the interplay between media and public opinion and the strategic behavior of political elites. Footnotes 1 Liberals (VVD), Social-Democrats (PvdA), Socialists (SP), Christian-Democrats (CDA), Right-wing populists (PVV), Social-Progressives (D66), Christian Party (CU), Greens (GroenLinks), Animal party (PvdD), and Elderly party (50+). Acknowledgments This study has been made possible by a Vidi grant (project number: 016.145.369) from the Netherlands Organisation for Scientific Research (NWO). References Bartels L. M. ( 2002 ). Beyond the running tally: Partisan bias in political perceptions . Political Behavior , 24 , 117 – 150 . doi:10.1023/A:1021226224601. Google Scholar CrossRef Search ADS Bélanger É. , Meguid B. M. ( 2008 ). Issue salience, issue ownership, and issue-based vote choice . Electoral Studies , 27 , 477 – 491 . doi:10.1016/j.electstud.2008.01.001. Google Scholar CrossRef Search ADS Boomgaarden H. G. , van Spanje J. , Vliegenthart R. , de Vreese C. H. ( 2011 ). Covering the crisis: Media coverage of the economic crisis and citizens’ economic expectations . Acta Politica , 46 , 353 – 379 . doi:10.1057/ap.2011.18. Google Scholar CrossRef Search ADS van der Brug W. ( 2004 ). Issue ownership and party choice . Electoral Studies , 23 , 209 – 233 . doi:10.1016/S0261-3794(02)00061-6. Google Scholar CrossRef Search ADS Budge I. , Farlie D. J. ( 1983 ). Explaining and predicting elections: Issue effects and party strategies in 23 democracies . London : Allen & Unwin . Campbell A. , Converse P. E. , Miller W. E. , Stokes D. E. ( 1960 ). The American voter . Chicago : The University of Chicago Press . Clarke H. D. , Sanders D. , Stewart M. C. , Whiteley P. (Eds.) ( 2004 ). Political choice in Britain . Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Clarke H. D. , Sanders D. , Stewart M. C. , & Whiteley P. (Eds.) ( 2009 ). Performance politics and the British voter . Cambridge : Cambridge University Press . Google Scholar CrossRef Search ADS Dassonneville R. , Hooghe M. ( 2015 ). Economic indicators and electoral volatility: Economic effects on electoral volatility in Western Europe, 1950–2013 . Comparative European Politics . doi:10.1057/cep.2015.3. Downs A. ( 1957 ). An economic theory of democracy . New York, NY : Harper and Row . Evans G. , Chzhen K. ( 2016a ). Re-evaluating the valence model of political choice . Political Science Research and Methods , 4 , 199 – 220 . doi:10.1017/psrm.2013.11. Google Scholar CrossRef Search ADS Evans G. , Chzhen K. ( 2016b ). Beyond valence: Estimating models of party choice without resort to ecological fallacy or unfounded causal assumptions. A reply to Whiteley et al . Political Science Research and Methods , 4 , 241 – 247 . doi:10.1017/psrm.2015.31. Google Scholar CrossRef Search ADS Fiorina M. P. ( 1981 ). Retrospective voting in American national elections . New Haven : Yale University Press . Garrett R. K. ( 2009 ). Politically motivated reinforcement seeking: Reframing the selective exposure debate . Journal of Communication , 59 , 676 – 699 . doi:10.1111/j.1460-2466.2009.01452.x. Google Scholar CrossRef Search ADS Gelman A. , Stern H. ( 2006 ). The difference between “Significant” and “Not Significant” is not itself statistically significant . The American Statistician , 60 , 328 – 331 . doi:10.1198/000313006X152649. Google Scholar CrossRef Search ADS Gerber A. S. , Huber G. A. ( 2010 ). Partisanship, political control, and economic assessments . American Journal of Political Science , 54 , 153 – 173 . doi:10.1111/j.1540-5907.2009.00424.x. Google Scholar CrossRef Search ADS Green J. , Hobolt S. B. ( 2008 ). Owning the issue agenda: Party strategies and vote choices in British elections . Electoral Studies , 27 , 460 – 476 . doi:10.1016/j.electstud.2008.02.003. Google Scholar CrossRef Search ADS Green J. , Jennings W. ( 2012 ). The dynamics of issue competence and vote for parties in and out of power: An analysis of valence in Britain, 1979–1997 . European Journal of Political Research , 51 , 469 – 503 . doi:10.1111/j.1475-6765.2011.02004.x. Google Scholar CrossRef Search ADS Jerit J. , Barabas J. ( 2012 ). Partisan perceptual bias and the information environment . Journal of Politics , 74 , 672 – 1684 . doi:10.1017/S0022381612000187. Google Scholar CrossRef Search ADS Kleinnijenhuis J. , Walter A. S. ( 2014 ). News, discussion, and associative issue ownership: Instability at the micro level versus stability at the macro level . The International Journal of Press/Politics , 19 , 226 – 245 . doi:10.1177/1940161213520043. Google Scholar CrossRef Search ADS Kuechler M. ( 1991 ). Issues and voting in the European elections 1989 . European Journal of Political Research , 19 , 81 – 103 . doi:10.1111/j.1475-6765.1991.tb01178.x. Google Scholar CrossRef Search ADS Lachat R. ( 2014 ). Issue ownership and the vote: The effects of associative and competence ownership on issue voting . Swiss Political Science Review , 20 , 727 – 740 . doi:doi:10.1111/spsr.12121. Google Scholar CrossRef Search ADS Lefevere J. , Tresch A. , Walgrave S. ( 2015 ). Introduction: Issue ownership . West European Politics , 38 , 755 – 760 . doi:10.1080/01402382.2015.1039375. Google Scholar CrossRef Search ADS Lefevere J. , Walgrave S. , Stubager R. , Tresch A. ( 2016 ). Measuring issue ownership: A comparative question wording experiment . Scandinavian Political Studies , 40 , 120 – 131 . doi:10.1111/1467-9477.12074. Google Scholar CrossRef Search ADS Lewis J. B. , King G. ( 1999 ). No evidence on directional vs. proximity voting . Political Analysis , 8 , 21 – 33 . doi:10.1093/oxfordjournals.pan.a029803. Google Scholar CrossRef Search ADS Lewis-Beck M. S. ( 2006 ). Does economics still matter? Econometrics and the vote . Journal of Politics , 68 , 208 – 212 . doi:10.1111/j.1468-2508.2006.00381.x. Google Scholar CrossRef Search ADS Lewis-Beck M. S. , Stegmaier M. ( 2000 ). Economic determinants of electoral outcomes . Annual Review of Political Science , 3 , 183 – 219 . doi:10.1146/annurev.polisci.3.1.183. Google Scholar CrossRef Search ADS Lijphart A. ( 1999 ). Patterns of democracy. Government forms and performance in thirty-six countries . New Haven; London : Yale University Press . Mair P. ( 2008 ). Electoral volatility and the Dutch party system: A comparative perspective . Acta Politica , 43 , 235 – 253 . doi:10.1057/ap.2008.1. Google Scholar CrossRef Search ADS Petrocik J. ( 1989 ). The theory of issue ownership: Issues, agendas, and electoral coalitions in the 1988 elections. Presented at The Annual meeting of the American Political Science Association, Atlanta, Georgia. Petrocik J. ( 1996 ). Issue ownership in presidential elections, with a 1980 case study . American Journal of Political Science , 40 , 825 – 850 . Google Scholar CrossRef Search ADS Riker W. H. ( 1993 ). Rhetorical interaction in the ratification campaigns. In Riker W. H. (Ed.), Agenda Formation (pp. 81 – 120 ). Michigan : University of Michigan Press . Google Scholar CrossRef Search ADS Sanders D. , Clarke H. D. , Stewart M. C. , Whiteley P. ( 2011 ). Downs, stokes and the dynamics of electoral choice . British Journal of Political Science , 41 , 287 – 314 . doi:10.1017/S0007123410000505. Google Scholar CrossRef Search ADS Stroud N. J. ( 2008 ). Media use and political predispositions: Revisiting the concept of selective exposure . Political Behavior , 30 , 341 – 366 . doi:10.1007/s11109-007-9050-9. Google Scholar CrossRef Search ADS Stubager R. , Slothuus R. ( 2013 ). What are the sources of political parties’ issue ownership? Testing four explanations at the individual level . Political Behavior , 35 , 567 – 588 . doi:10.1007/s11109-012-9204-2. Google Scholar CrossRef Search ADS van der Eijk C. , van der Brug W. , Franklin M. N. ( 2006 ). Rethinking the dependent variable in voting behavior: On the measurement and analysis of electoral utilities . Electoral Studies , 25 , 424 – 447 . Google Scholar CrossRef Search ADS Wagner M. , Zeglovits E. ( 2014 ). Survey questions about party competence: Insights from cognitive interviews . Electoral Studies , 34 , 280 – 290 . doi:10.1016/j.electstud.2013.09.005. Google Scholar CrossRef Search ADS PubMed Walgrave S. , Lefevere J. ( 2017 ). Long-term associative issue ownership change. A panel study in Belgium . Journal of Elections, Public Opinion and Parties . Advance Online Publication. doi:10.1080/17457289.2017.1285305. Walgrave S. , Lefevere J. , Nuytemans M. ( 2009 ). Issue ownership stability and change. How political parties claim and maintain issues through media appearances . Political Communication , 26 , 153 – 172 . doi:10.1080/10584600902850718. Google Scholar CrossRef Search ADS Walgrave S. , Lefevere J. , Tresch A. ( 2012 ). The associative dimension of issue ownership . Public Opinion Quarterly , 76 , 771 – 782 . doi:10.1093/poq/nfs023. Google Scholar CrossRef Search ADS Walgrave S. , Lefevere J. , Tresch A. ( 2014 ). The limits of issue ownership dynamics: The constraining effect of party preference . Journal of Elections, Public Opinion and Parties , 24 ( 1 ), 1 – 19 . doi:10.1080/17457289.2013.811245. Google Scholar CrossRef Search ADS Walgrave S. , Tresch A. , Lefevere J. ( 2015 ). The conceptualisation and measurement of issue ownership . West European Politics , 38 , 778 – 796 . doi:10.1080/01402382.2015.1039381. Google Scholar CrossRef Search ADS Walgrave S. , Van Camp K. , Lefevere J. , Tresch A. ( 2016 ). Measuring issue ownership with survey questions. A question wording experiment . Electoral Studies , 42 , 290 – 299 . doi:10.1016/j.electstud.2015.12.001. Google Scholar CrossRef Search ADS Whiteley P. , Clarke H. , Sanders D. , Stuart M. ( 2016 ). Hunting the Snark: A reply to “Re-evaluating valence models of political choice” . Political Science Research and Methods , 4 , 221 – 240 . doi:10.1017/psrm.2015.30. Google Scholar CrossRef Search ADS © The Author 2017. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Public Opinion Research Oxford University Press

Disentangling the Direction of Causality Between Competence Issue Ownership and Party Preference

Loading next page...
 
/lp/ou_press/disentangling-the-direction-of-causality-between-competence-issue-gcsAuWW1Wp
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved.
ISSN
0954-2892
eISSN
1471-6909
D.O.I.
10.1093/ijpor/edx017
Publisher site
See Article on Publisher Site

Abstract

Introduction Issue perceptions, particularly issue ownership, are increasingly used to understand voters’ electoral choices (Lefevere, Tresch, & Walgrave, 2015). Issue ownership refers to the idea that voters perceive some parties as better able to “handle” certain issues. Consequently, parties can be considered as “owning” issues they have a good reputation on (Budge & Farlie, 1983; Petrocik, 1989, 1996). These reputations affect voters’ electoral behavior: voters are more likely to vote for a party owning the issues of importance to them (Bélanger & Meguid, 2008; Lachat, 2014; Petrocik, 1996; Walgrave, Lefevere, & Tresch, 2012). However, voters’ issue ownership perceptions may be endogenous to their party preference (Walgrave, Tresch, & Lefevere, 2015). This criticism is most often raised against the competence dimension of issue ownership, which refers to parties’ competence in handling an issue. For example, Kuechler (1991) found that competence evaluations reflect vote choices to a high degree, which suggests that the direction of causality between competence issue ownership and party preference could be reversed (van der Brug, 2004). Green and Jennings (2012) find that competence evaluations Granger cause party preference. Yet, most of these studies rely on cross-sectional data, which limit their ability for causal inferences, while others rely on aggregate-level data, making it impossible to track changes at the individual level. Panel data offer the opportunity to assess whether competence issue ownership causes party preference, or whether the reverse causal relationship is dominant. We present analyses based on three-wave panel data collected among a sample of Dutch citizens, which polled respondents for their party preference and competence issue ownership perceptions on the economy in each wave. Our findings suggest that voters’ economic issue ownership is driven more by their vote choices, than vice versa. The Endogeneity of Issue Ownership A variety of studies assert that voters’ issue perceptions affect vote choices. Spatial models of electoral choice suggest that voters tend to take parties’ issue positions into account when they cast their ballot (Lewis & King, 1999). Yet, scholars increasingly assert that positions are not the only determinant at play (Green & Hobolt, 2008): parties’ issue reputations also matter. This idea is discussed in multiple theories of voting: Riker’s (1993) dominance theory, the valence theory of voting (Sanders, Clarke, Stewart, & Whiteley, 2011), and issue ownership theory (Budge & Farlie, 1983; Petrocik, 1989, 1996). Empirical studies have shown that such issue evaluations affect voters’ electoral choices—both in the aggregate and at the individual level (Sanders et al., 2011; Walgrave et al., 2015). Yet, the persistent empirical support for the electoral impact of voters’ issue evaluations has been matched by an equally persistent critique on their validity as a determinant of electoral choice. The criticism, in its most basic form, is the possibility of reverse causation. Issue evaluations might well be a consequence of party preference, rather than the other way around. This critique relates to a broad discussion regarding the impact of party preferences on voters’ political attitudes. The roots of this discussion lie with Campbell, Converse, Miller, and Stokes’ (1960) argument that party identification acts as a perceptual screen through which voters perceive the world. Though distinct from party preference, this work on party identification suggests that party preference can affect voter’s political attitudes: partisanship affects which information voters attend to (Garrett, 2009; Stroud, 2008) and how they learn from that information (Gerber & Huber, 2010; Jerit & Barabas, 2012). Yet, a parallel line of literature, based on Downs (1957) and Fiorina (1981), conceptualizes party preference as a running tally of voters’ past experiences, suggesting that attitudes drive party preference (Bartels, 2002). Applied to voters’ issue evaluations, there is ongoing debate regarding the extent to which the above critique holds true. In 2006, Evans and Andersen raised concerns about the impact of party preferences on voters’ economic perceptions, which causes scholars to overstate the impact of such perceptions on political preferences (but see Lewis-Beck, 2006). More specifically, party issue evaluations are part of the valence model of electoral choice (Clarke, Sanders, Stewart, & Whiteley, 2004, 2009): using a panel design, Evans and Chzhen tested whether valence evaluations cause party preference, or vice versa. They concluded that: “assessments of government performance are, to a significant degree, a product of party preference and cannot therefore be used to explain that preference” (Evans & Chzhen, 2016a, p. 213; but see Whiteley, Clarke, Sanders, & Stuart, 2016; Evans & Chzhen, 2016b). Yet, as far as we can tell similar panel data analyses do not exist for measures of issue ownership. We focus here only on the competence dimension of issue ownership, which has been most suspect of being caused by party preference (van der Brug, 2004; Stubager & Slothuus, 2013). It is important to assert whether reverse causation also plagues issue ownership measures specifically because the findings regarding related measures of valence may, or may not, apply to a similar degree to measures of competence issue ownership. To start, parties’ valence evaluations are based on a variety of performance assessments—including both issue and leader affect (Evans & Chzhen, 2016a, b; Whiteley et al., 2016), whereas competence issue ownership focuses solely on parties’ issue reputations. Hence, it is not clear to what extent we can extrapolate these findings based on a mix of attitudes to attitudes focused solely on issues. Indeed, based on cognitive interviews, Wagner and Zeglovits (2014) suggest that respondents’ answers to issue ownership questions seem less influenced by their overall party preference. Moreover, valence measures do not require voters to choose one party over another as is common in issue ownership measures—a party owns an issue if it handles the issue better than its competitors. In contrast, valence is about the evaluation of a single political actor as doing either a good or bad job in handling an issue (Clarke et al., 2004). That said, the empirical evidence on competence issue ownership is at least suggestive of potential endogeneity. Apart from the fact that various scholars have raised concerns regarding reverse causation (van der Brug, 2004; Wagner & Zeglovits, 2014), various question wording experiments demonstrate that respondents’ competence issue ownership perceptions correlate to their party preferences (Lefevere, Walgrave, Stubager, & Tresch, 2016; Walgrave, Van Camp, Lefevere, & Tresch, 2016). Finally, Green and Jennings (2012) show that assessments of issue competence Granger cause party preferences for some parties at an aggregate level. Therefore, the extant literature leads us to formulate two competing hypotheses. On the one hand, issue ownership theory claims that respondents’ perceptions of competence issue ownership determine their party preferences. Thus, our first hypothesis is: H1: Voters’ competence issue ownership perceptions positively affect voters’ party preferences. On the other hand, a number of studies have pointed to the potential endogeneity of the competence dimension of issue ownership (Bélanger & Meguid, 2008; Lefevere et al., 2016; Wagner & Zeglovits, 2014; Walgrave et al., 2016). Hence, we posit that: H2: Voters’ party preferences positively affect voters’ competence issue ownership perceptions. Finally, we expect that while voters’ competence issue ownership perceptions and party preferences have a reciprocal causal relationship (H1 and H2), in line with the assumption underlying most of the empirical studies, we maintain that the dominant causal relationship is from competence issue ownership perceptions to party preferences, rather than the other way around. H3: The dominant causal relationship is from competence issue ownership perceptions to party preference, rather than vice versa. Methods To empirically test our hypotheses, we rely on panel survey evidence testing the nature of the causal relationship between competence issue ownership and party preference. More specifically, we use the results from a three-wave online panel survey conducted by I&O Research, an ISO-certified research company, indicating that it meets a set of quality criteria for data collection and quality set in liaison with among others the World Association of Public Opinion Research (WAPOR). The survey was fielded in the first half of 2015, among a sample drawn from I&O’s ongoing panel of respondents, and aims to be representative for the Dutch voting population. Potential participants are for the most part selected based on random samples from (local) population registers and recruited actively, without the possibility of self-registration. The surveys were held within the context of a larger project that focuses on the effects of media coverage on economic and political perceptions, attitudes, and behavior in the aftermath of the large economic crisis that hit most European Union countries. The survey contained a wide range of questions relating to media exposure, the economy, politics, and consumer behavior. The time lag between each wave was 10 weeks, and dates of fielding were February 23, 2015 (Wave 1), April 20, 2015 (Wave 2), and June 15, 2015 (Wave 3). For the first wave, 22,879 people were invited to participate and 9,112 started the questionnaire (response rate: 39.8%). Of those who started the questionnaire, 6,386 completed the survey (completion rate: 70.1%). Only respondents who completed this survey wave were invited to participate in the second wave. In Wave 2, 4,301 respondents completed the questionnaire (completion rate: 69.0%). From those respondents, 3,270 also completed the third wave (completion rate: 77.0%). In this analysis, we only include those respondents that completed all three waves. Despite the intention to create a representative sample, the final sample deviates from the general Dutch population in that it overrepresents older (M = 61, 45, SD = 11.08), male (66%), and highly educated respondents (50.9% obtained a university degree). These numbers do not differ substantially from those of all participants that participated in Wave 1. The Netherlands is a parliamentary democracy with a system of proportional representation and a wide range of political parties being represented in Dutch parliament: 10 parties after the 2012 parliamentary elections (Lijphart, 1999). It is a good case to test our hypotheses, as electoral volatility in the Netherlands is relatively high—and hence respondents’ party preferences should fluctuate over time (Mair, 2008). Moreover, the fragmented nature of the party system means that issue ownership of parties is likely to be contested and should fluctuate over time (Kleinnijenhuis & Walter, 2014). As a measurement of party preference, we use van der Eijk et al.’s propensity to vote question: in each wave, respondents were asked for all parties (11) that got elected in Dutch parliament during the last parliamentary elections1 to indicate the likelihood they would ever vote for that party on an 11-point scale ranging from 1 (certainly not) to 11 (certainly) (van der Eijk, van der Brug, & Franklin, 2006). This allows us to track minor changes in party preference. To measure issue ownership, we use the formulation used by Walgrave et al. (2009): respondents were asked to indicate how capable they considered each of the 11 political parties to be in dealing with the economy and economic issues. Again, an 11-point scale was used, ranging from 1 (totally incapable) to 11 (very much capable). The questions on party preference were asked consistently before those on competence issue ownership. We cannot assert whether this has affected the results, but because of the fact that a number of other questions were always included in between the party preference and issue ownership question blocks, we have no reason to assume that the effect of question ordering had a large impact here. We measured issue ownership only for a single issue—the economy. Though it would be preferable to assess our hypotheses over different issues, and have some baseline to compare with, the survey data only included issue ownership preferences for this single issue. That said, the economy is an appropriate issue to test our hypotheses: the economy is a typical performance issue on which parties can actively perform for better or worse—so we would expect fluctuations in voter’s assessments of parties over time (Petrocik, 1996). Indeed, ample studies on economic voting suggest that voters’ economic perceptions have tangible electoral consequences (Lewis-Beck & Stegmaier, 2000). Whereas for many issues voters have little evidence of party performance, evidence on economic performance is widely available and reported on in the mass media (Boomgaarden, van Spanje, Vliegenthart, & de Vreese, 2011). As such, especially for the economy, we would expect issue reputations to fluctuate, and to affect party preferences. Moreover, many voters consider the economy to be an important issue. In each of the three waves, an open question was asked about the most important problem the country was facing today. The answers were categorized in a range of policy topics, and while the worst part of the economic recession had passed in 2015, still 67.8% of the respondents indicated at least one time an economic topic as the most important problem—by far the most frequently mentioned one. Table 1 provides more detailed descriptive statistics for both variables, demonstrating that there exists considerable variation across parties, and to a lesser extent, waves. Table 1 Descriptive Statistics Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Note. PVV = Partij voor de Vrijheid. Table 1 Descriptive Statistics Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Party Wave 1 Wave 2 Wave 3 Party preference Issue ownership Party preference Issue ownership Party preference Issue ownership M SD M SD M SD M SD M SD M SD VVD 3.38 3.31 5.9 2.77 3.75 3.46 6.13 2.95 3.59 3.37 5.87 2.97 PvdA 3.28 3.05 4.94 2.49 3.49 3.09 4.82 2.62 3.26 2.93 4.67 2.64 SP 2.25 2.76 4.45 2.55 2.27 2.78 4.28 2.66 2.24 2.74 4.19 2.66 CDA 3.3 3.14 5.96 2.24 3.69 3.31 6.07 2.42 3.53 3.22 5.86 2.5 PVV 3.39 3.05 2.86 2.42 3.98 3.17 2.59 2.35 3.82 3.12 2.53 2.34 D66 4.3 3.27 6.12 2.5 4.67 3.27 6.04 2.64 4.49 3.22 5.8 2.67 ChristenUnie 2.42 2.52 5.33 2.44 2.9 2.69 5.04 2.58 2.78 2.61 4.85 2.61 GroenLinks 2.96 2.85 4.38 2.4 3.4 2.97 4.16 2.51 3.5 2.99 4.16 2.56 SGP 1.56 1.75 4.39 2.5 1.84 1.99 4.04 2.59 1.83 1.97 3.81 2.59 Partij voor de Dieren 1.88 2.13 2.68 2.07 2.2 2.39 2.53 2.11 2.09 2.29 2.34 1.99 50+ 2.47 2.47 3.41 2.32 2.74 2.65 3.1 2.34 2.61 2.57 2.94 2.28 Mean 2.84 4.58 3.17 4.44 3.07 4.28 Note. PVV = Partij voor de Vrijheid. Our unit of analysis is a combination of respondent, party, and wave. We estimate two models: one model that predicts party preference based on party preference in the previous wave and issue ownership in the previous wave; and one model that predicts issue ownership based on issue ownership in the previous wave and party preference in the previous wave. Thus, all our regression models include a lagged dependent variable, which is considered an appropriate analytical approach to capture change in the dependent variable, while also accounting for initial different levels across individuals. Our models do not include any respondent characteristics as control variables, as there is no reason to assume that they would systematically affect our dependent variables irrespective of the party considered. To account for party-level heterogeneity (different levels of party support or attributed issue competence that are not captured by the independent variables in the models), we do include party fixed effects in the models. We use clustered standard errors on the respondent level. To make coefficients comparable, we use standardized variables. The total number of observations is 71,280 (3,240 respondents × 2 waves × 11 parties). To check the robustness of our findings, we also conducted a cross-lagged model using structural equation models using the same variables, and tested the regression model for individual parties as well. Results Table 2 presents the results from our analysis. Both our first and second hypotheses are confirmed: Model 1 shows that perceptions in issue ownership affect party preferences in the subsequent wave, controlling for the previous party preference (H1). The effect is, as expected, positive. Yet, the reverse relationship is also significant. If a party is preferred more, this causes upward shifts in competence on the economic issue (H2). If we compare the coefficients from both models, we have to reject H3. The standardized coefficient of the effect of party preference on issue ownership is considerably larger (.216 compared with .127). The difference is .089 and statistically highly significant (SE = 0.006, p < .001; for calculation, see Gelman & Stern, 2006). This is also clear when we compare the change in explained variance for both instances. While issue ownership adds 0.9% to the explained variance of party preferences, party preferences add >3 times as much (2.8%) to the explained variance of issue ownership. Table 2 Explaining Party Preference and Issue Ownership Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Note. N = 71,280; * p < .0001; standard errors are clustered per respondent. Fixed effects for parties are included but not displayed. †Compared with a model that only includes the lagged dependent variable. Table 2 Explaining Party Preference and Issue Ownership Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Model 1 Party preference Model 2 Issue ownership β SE β SE Constant 0.042* 0.005 −0.255* 0.007 Party preference (t−1) 0.775* 0.005 0.216* 0.005 Issue ownership (t−1) 0.127* 0.004 0.569* 0.006 R2 0.728 0.609 ΔR2† 0.009 0.028 Note. N = 71,280; * p < .0001; standard errors are clustered per respondent. Fixed effects for parties are included but not displayed. †Compared with a model that only includes the lagged dependent variable. Our robustness checks confirm the findings described above. The resulting cross-lagged model using structural equation modeling with party fixed effects has a good model fit (χ2 = 3.946, df = 3, p = .267, Root Mean Square Error of Approximation (RMSEA)=.002, Comparative Fit Index (CFI) = 1.000) and shows similar results as the regression analysis. Both effects are significant: the impact of issue ownership on party preference has a standardized coefficient of 0.125, and the impact of party preference on issue ownership of 0.213. A χ2 test suggests that the null hypothesis of nondifference between coefficients clearly has to be rejected (χ2 = 254.49, df = 1, p < .0001). The effect of issue ownership on party preference is thus significantly smaller than the effect of party preference on issue ownership. H3 is rejected. The party-level analyses, reported in Table 3, confirm this overall finding. Regression analyses for each of the 11 parties show that in all instances, the standardized coefficient for the effect of party preference on issue ownership is larger than the standardized coefficient for the reverse relationship, and the difference between the coefficients is significant in all but two cases (the small Christian party ChristenUnie and the social–liberal D66). The largest difference (and also the largest effect of party preference on issue ownership) exists for the populist right party Partij voor de Vrijheid (PVV) of Geert Wilders. Apparently, economic issue ownership perceptions have a more limited influence on changes in preferences for this party, but a preference for this party does result in significantly higher perceptions of its competence on the economic issue. Table 3 Relationships Between Party Preference and Issue Ownership per Party Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * * p < .0001; standard errors are clustered per respondent. †Indicates significance of difference between the two coefficients. PVV = Partij voor de Vrijheid. Table 3 Relationships Between Party Preference and Issue Ownership per Party Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * Party IO → PP PP → IO Significance β SE β SE VVD 0.103 0.007 0.186 0.008 * PvdA 0.129 0.008 0.202 0.01 * SP 0.155 0.011 0.255 0.011 * CDA 0.114 0.01 0.186 0.009 * PVV 0.063 0.011 0.273 0.015 * D66 0.183 0.01 0.196 0.01 NS ChristenUnie 0.128 0.008 0.145 0.01 NS GroenLinks 0.18 0.011 0.268 0.011 * SGP 0.073 0.006 0.207 0.015 * Partij voor de Dieren 0.136 0.012 0.255 0.015 * 50+ 0.143 0.012 0.268 0.013 * * p < .0001; standard errors are clustered per respondent. †Indicates significance of difference between the two coefficients. PVV = Partij voor de Vrijheid. Conclusion and Discussion We analyzed the extent to which Dutch voters’ competence issue ownership perceptions of the economy determined their party preferences, and vice versa. Though issue ownership perceptions are increasingly used to explain vote choices (Lefevere et al., 2015), scholars have pointed to the potential endogeneity of especially competence issue ownership and party preferences (van der Brug, 2004; Wagner & Zeglovits, 2014). Our findings, though limited in nature, will not diminish these criticisms. The “good” news is that we find that competence issue ownership perceptions matter for party preferences. The “bad” is that the reversed causal relationship from party preferences to competence issue ownership perceptions turns out to be much stronger. This finding is in line with recent work by Evans and Chzhen (2016a), who also point out that assessments of competence seem to be driven by, more than being a determinant of, party preferences. Apparently, people update their perceptions about specific ideas about the competence of parties to deal with specific issues to a considerable extent after they decide on a general party preference. That said, our data is not without shortcomings, and we need to add three important caveats to our conclusions. First, we only considered a single issue—the economy—when it came to ownership perceptions. This limitation may be why we find that the reverse causal relationship—from party preference to issue ownership—is dominant: perhaps party preferences themselves are shaped by the concurrent impact of multiple issues, rather than the single issue we studied here. Moreover, the fact that no party held clear ownership over the issue may affect the result, as evidence suggests that stronger issue ownership (in the aggregate) may be more resilient to change (Walgrave & Lefevere, 2017). Therefore, future research should consider a wider variety of issues and test whether the findings hold for other issues as well. That said, the importance of the economic issue in the electoral process (Dassonneville & Hooghe, 2015), as well as the fact that it was mentioned by more than two-third of the respondents in our survey when asked about the most important problem the country was facing, do suggest that the issue is an important one. Moreover, competence evaluations on the economy should be less prone to conditioning by prior party preferences. Voters have straightforward indicators (e.g., changes in the gross domestic product) that allow them to assess how the economy is doing—and thus to assess the track record of the parties on it. Yet, we find evidence that party preferences do affect these perceptions. For more complex issues where clear indicators of performance are absent, we would expect that the conditioning effect of party preference would be even stronger. These factors make us confident that findings for other issues should lead to similar results. A second limitation is the representativeness of our sample, which over represents male, older, and highly educated people. Yet, we do not see any obvious way that these background characteristics might fundamentally alter the relationships we examined. Given the crystal clear picture that arises and the considerable effects we find, we are confident that results are generalizable, at least to the wider Dutch population. Moreover, the Dutch party system is highly fragmented, and most voters have several alternative parties to choose from. We would expect that the conditioning impact of party preference on issue ownership would be at least as strong in systems with less parties. Third, despite the panel nature of our data, experimental designs might be even better suited to tease out the two-way causality. For example, prior work has shown that the impact of parties’ media performances is moderated by party preference (Walgrave, Lefevere, & Tresch, 2014). A modified version of such an approach, triggering (moderated) shifts in issue ownership (and party preference in turn), could then use repeated measures to assert in which direction the influence is strongest. Our findings add to a growing list of studies that urge scholars to use competence evaluations with caution: we find that competence issue ownership perceptions of the economy were caused by, rather than causing, party preferences. This does not automatically invalidate their use as a determinant of electoral choice—we find that issue ownership affects party preference, controlling for lagged party preference—but it does show that perceptions themselves are subject to partisan conditioning, which is troublesome. Recent work on the associative dimension of issue ownership has demonstrated that this dimension is less strongly correlated to party preference: we urge future research to repeat our approach for this associative dimension, as it might prove to be more resilient to endogeneity. Given the potential issues of endogeneity, scholars should also examine findings regarding the impact of different question wordings: new formulations might trigger respondents to consider ‘pure’ competence, rather than an amalgam of competence and party preference. In sum, while competence issue ownership seems subject to similar issues that plague measures of party competence more broadly, we urge scholars to continue work on the concept of issue ownership, as its burgeoning research agenda suggests various avenues to move forward. Rens Vliegenthart is professor for Media and Society at the Amsterdam School of Communication Research (ASCoR) at the University of Amsterdam. His research interests include media–politics interactions, media-effects, and comparative politics. Jonas Lefevere is Assistant Professor of Communications at Vesalius College Brussels. His research focuses on the interplay between media and public opinion and the strategic behavior of political elites. Footnotes 1 Liberals (VVD), Social-Democrats (PvdA), Socialists (SP), Christian-Democrats (CDA), Right-wing populists (PVV), Social-Progressives (D66), Christian Party (CU), Greens (GroenLinks), Animal party (PvdD), and Elderly party (50+). Acknowledgments This study has been made possible by a Vidi grant (project number: 016.145.369) from the Netherlands Organisation for Scientific Research (NWO). References Bartels L. M. ( 2002 ). Beyond the running tally: Partisan bias in political perceptions . Political Behavior , 24 , 117 – 150 . doi:10.1023/A:1021226224601. Google Scholar CrossRef Search ADS Bélanger É. , Meguid B. M. ( 2008 ). Issue salience, issue ownership, and issue-based vote choice . Electoral Studies , 27 , 477 – 491 . doi:10.1016/j.electstud.2008.01.001. Google Scholar CrossRef Search ADS Boomgaarden H. G. , van Spanje J. , Vliegenthart R. , de Vreese C. H. ( 2011 ). Covering the crisis: Media coverage of the economic crisis and citizens’ economic expectations . Acta Politica , 46 , 353 – 379 . doi:10.1057/ap.2011.18. Google Scholar CrossRef Search ADS van der Brug W. ( 2004 ). Issue ownership and party choice . Electoral Studies , 23 , 209 – 233 . doi:10.1016/S0261-3794(02)00061-6. Google Scholar CrossRef Search ADS Budge I. , Farlie D. J. ( 1983 ). Explaining and predicting elections: Issue effects and party strategies in 23 democracies . London : Allen & Unwin . Campbell A. , Converse P. E. , Miller W. E. , Stokes D. E. ( 1960 ). The American voter . Chicago : The University of Chicago Press . Clarke H. D. , Sanders D. , Stewart M. C. , Whiteley P. (Eds.) ( 2004 ). Political choice in Britain . Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Clarke H. D. , Sanders D. , Stewart M. C. , & Whiteley P. (Eds.) ( 2009 ). Performance politics and the British voter . Cambridge : Cambridge University Press . Google Scholar CrossRef Search ADS Dassonneville R. , Hooghe M. ( 2015 ). Economic indicators and electoral volatility: Economic effects on electoral volatility in Western Europe, 1950–2013 . Comparative European Politics . doi:10.1057/cep.2015.3. Downs A. ( 1957 ). An economic theory of democracy . New York, NY : Harper and Row . Evans G. , Chzhen K. ( 2016a ). Re-evaluating the valence model of political choice . Political Science Research and Methods , 4 , 199 – 220 . doi:10.1017/psrm.2013.11. Google Scholar CrossRef Search ADS Evans G. , Chzhen K. ( 2016b ). Beyond valence: Estimating models of party choice without resort to ecological fallacy or unfounded causal assumptions. A reply to Whiteley et al . Political Science Research and Methods , 4 , 241 – 247 . doi:10.1017/psrm.2015.31. Google Scholar CrossRef Search ADS Fiorina M. P. ( 1981 ). Retrospective voting in American national elections . New Haven : Yale University Press . Garrett R. K. ( 2009 ). Politically motivated reinforcement seeking: Reframing the selective exposure debate . Journal of Communication , 59 , 676 – 699 . doi:10.1111/j.1460-2466.2009.01452.x. Google Scholar CrossRef Search ADS Gelman A. , Stern H. ( 2006 ). The difference between “Significant” and “Not Significant” is not itself statistically significant . The American Statistician , 60 , 328 – 331 . doi:10.1198/000313006X152649. Google Scholar CrossRef Search ADS Gerber A. S. , Huber G. A. ( 2010 ). Partisanship, political control, and economic assessments . American Journal of Political Science , 54 , 153 – 173 . doi:10.1111/j.1540-5907.2009.00424.x. Google Scholar CrossRef Search ADS Green J. , Hobolt S. B. ( 2008 ). Owning the issue agenda: Party strategies and vote choices in British elections . Electoral Studies , 27 , 460 – 476 . doi:10.1016/j.electstud.2008.02.003. Google Scholar CrossRef Search ADS Green J. , Jennings W. ( 2012 ). The dynamics of issue competence and vote for parties in and out of power: An analysis of valence in Britain, 1979–1997 . European Journal of Political Research , 51 , 469 – 503 . doi:10.1111/j.1475-6765.2011.02004.x. Google Scholar CrossRef Search ADS Jerit J. , Barabas J. ( 2012 ). Partisan perceptual bias and the information environment . Journal of Politics , 74 , 672 – 1684 . doi:10.1017/S0022381612000187. Google Scholar CrossRef Search ADS Kleinnijenhuis J. , Walter A. S. ( 2014 ). News, discussion, and associative issue ownership: Instability at the micro level versus stability at the macro level . The International Journal of Press/Politics , 19 , 226 – 245 . doi:10.1177/1940161213520043. Google Scholar CrossRef Search ADS Kuechler M. ( 1991 ). Issues and voting in the European elections 1989 . European Journal of Political Research , 19 , 81 – 103 . doi:10.1111/j.1475-6765.1991.tb01178.x. Google Scholar CrossRef Search ADS Lachat R. ( 2014 ). Issue ownership and the vote: The effects of associative and competence ownership on issue voting . Swiss Political Science Review , 20 , 727 – 740 . doi:doi:10.1111/spsr.12121. Google Scholar CrossRef Search ADS Lefevere J. , Tresch A. , Walgrave S. ( 2015 ). Introduction: Issue ownership . West European Politics , 38 , 755 – 760 . doi:10.1080/01402382.2015.1039375. Google Scholar CrossRef Search ADS Lefevere J. , Walgrave S. , Stubager R. , Tresch A. ( 2016 ). Measuring issue ownership: A comparative question wording experiment . Scandinavian Political Studies , 40 , 120 – 131 . doi:10.1111/1467-9477.12074. Google Scholar CrossRef Search ADS Lewis J. B. , King G. ( 1999 ). No evidence on directional vs. proximity voting . Political Analysis , 8 , 21 – 33 . doi:10.1093/oxfordjournals.pan.a029803. Google Scholar CrossRef Search ADS Lewis-Beck M. S. ( 2006 ). Does economics still matter? Econometrics and the vote . Journal of Politics , 68 , 208 – 212 . doi:10.1111/j.1468-2508.2006.00381.x. Google Scholar CrossRef Search ADS Lewis-Beck M. S. , Stegmaier M. ( 2000 ). Economic determinants of electoral outcomes . Annual Review of Political Science , 3 , 183 – 219 . doi:10.1146/annurev.polisci.3.1.183. Google Scholar CrossRef Search ADS Lijphart A. ( 1999 ). Patterns of democracy. Government forms and performance in thirty-six countries . New Haven; London : Yale University Press . Mair P. ( 2008 ). Electoral volatility and the Dutch party system: A comparative perspective . Acta Politica , 43 , 235 – 253 . doi:10.1057/ap.2008.1. Google Scholar CrossRef Search ADS Petrocik J. ( 1989 ). The theory of issue ownership: Issues, agendas, and electoral coalitions in the 1988 elections. Presented at The Annual meeting of the American Political Science Association, Atlanta, Georgia. Petrocik J. ( 1996 ). Issue ownership in presidential elections, with a 1980 case study . American Journal of Political Science , 40 , 825 – 850 . Google Scholar CrossRef Search ADS Riker W. H. ( 1993 ). Rhetorical interaction in the ratification campaigns. In Riker W. H. (Ed.), Agenda Formation (pp. 81 – 120 ). Michigan : University of Michigan Press . Google Scholar CrossRef Search ADS Sanders D. , Clarke H. D. , Stewart M. C. , Whiteley P. ( 2011 ). Downs, stokes and the dynamics of electoral choice . British Journal of Political Science , 41 , 287 – 314 . doi:10.1017/S0007123410000505. Google Scholar CrossRef Search ADS Stroud N. J. ( 2008 ). Media use and political predispositions: Revisiting the concept of selective exposure . Political Behavior , 30 , 341 – 366 . doi:10.1007/s11109-007-9050-9. Google Scholar CrossRef Search ADS Stubager R. , Slothuus R. ( 2013 ). What are the sources of political parties’ issue ownership? Testing four explanations at the individual level . Political Behavior , 35 , 567 – 588 . doi:10.1007/s11109-012-9204-2. Google Scholar CrossRef Search ADS van der Eijk C. , van der Brug W. , Franklin M. N. ( 2006 ). Rethinking the dependent variable in voting behavior: On the measurement and analysis of electoral utilities . Electoral Studies , 25 , 424 – 447 . Google Scholar CrossRef Search ADS Wagner M. , Zeglovits E. ( 2014 ). Survey questions about party competence: Insights from cognitive interviews . Electoral Studies , 34 , 280 – 290 . doi:10.1016/j.electstud.2013.09.005. Google Scholar CrossRef Search ADS PubMed Walgrave S. , Lefevere J. ( 2017 ). Long-term associative issue ownership change. A panel study in Belgium . Journal of Elections, Public Opinion and Parties . Advance Online Publication. doi:10.1080/17457289.2017.1285305. Walgrave S. , Lefevere J. , Nuytemans M. ( 2009 ). Issue ownership stability and change. How political parties claim and maintain issues through media appearances . Political Communication , 26 , 153 – 172 . doi:10.1080/10584600902850718. Google Scholar CrossRef Search ADS Walgrave S. , Lefevere J. , Tresch A. ( 2012 ). The associative dimension of issue ownership . Public Opinion Quarterly , 76 , 771 – 782 . doi:10.1093/poq/nfs023. Google Scholar CrossRef Search ADS Walgrave S. , Lefevere J. , Tresch A. ( 2014 ). The limits of issue ownership dynamics: The constraining effect of party preference . Journal of Elections, Public Opinion and Parties , 24 ( 1 ), 1 – 19 . doi:10.1080/17457289.2013.811245. Google Scholar CrossRef Search ADS Walgrave S. , Tresch A. , Lefevere J. ( 2015 ). The conceptualisation and measurement of issue ownership . West European Politics , 38 , 778 – 796 . doi:10.1080/01402382.2015.1039381. Google Scholar CrossRef Search ADS Walgrave S. , Van Camp K. , Lefevere J. , Tresch A. ( 2016 ). Measuring issue ownership with survey questions. A question wording experiment . Electoral Studies , 42 , 290 – 299 . doi:10.1016/j.electstud.2015.12.001. Google Scholar CrossRef Search ADS Whiteley P. , Clarke H. , Sanders D. , Stuart M. ( 2016 ). Hunting the Snark: A reply to “Re-evaluating valence models of political choice” . Political Science Research and Methods , 4 , 221 – 240 . doi:10.1017/psrm.2015.30. Google Scholar CrossRef Search ADS © The Author 2017. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved.

Journal

International Journal of Public Opinion ResearchOxford University Press

Published: Sep 5, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off