Evidence suggests that politically right-leaning individuals are more likely to be closed-minded. Whether this association is inherent or subject to change has been the subject of debate, yet has not been formally tested. Through a meta-analysis, we find evidence of a changing association between conservatism and facets of closed-mindedness in the U.S. and international context using 341 unique samples, over 200,000 participants, and 920 estimates over 71 years. In the U.S., data ranging from 1948 to 2019 revealed a linear decline in the association between social conservatism (SC) and closed-mindedness, though economic conservatism (EC) did not vary in its association with closed-mindedness over time. Internationally across 18 countries, excluding the U.S., we observed a curvilinear decline in the association between SC and closed-mindedness over that same time, but no change in ECs association. We also tested variation over time for attitudinal measures of conservatism ranging between 1987 to 2018. In the U.S., we observed a linear increase in the association between right-wing authoritarianism (RWA) and closed-mindedness, with a similar linear increase in the association between social dominance orientation (SDO) and closed-mindedness. Internationally, there was a curvilinear increase in the association between RWA and closed-mindedness, but no change in the association with SDO. We discuss the changes to the political landscape that might explain our findings.
Compared to liberals, conservatives tend to score higher on individual difference measures of closed-mindedness. That is, conservatives tend to describe themselves as less tolerant of uncertainty, more likely to rely on intuition rather than effortful thinking, and as avoiding rather than embracing change. Previously, scholars have argued that traditionalism and mainstream ideas have an inherent appeal to people who are more closed-minded. Conservatives often adopt a traditional stance on a host of issues, whereas liberals are more likely to embrace a non-traditional stance. That is, self-described differences between conservatives and liberals in closed-mindedness seem to demonstrate a preference for certain types of political beliefs.
We asked whether conservatives and liberals always differ in their self-reported closed-mindedness, or whether the size of this difference may have changed over time. We tested the hypothesis that the differences have increased against the hypothesis that the differences have shrunken in size. The hypothesis that difference would have increased is based on the observation that, across the world, the public has become more polarized, implying that liberals and conservatives have become more distinct on a host of political values and beliefs. If individual differences in closed-mindedness reflect a preference for matching values and beliefs, then we would expect conservatives and liberals to be more different today than they once were. The hypothesis that the difference between conservatives and liberals would have decreased is based on the notion that once liberal ideas have become much more mainstream; thus, embracing such beliefs may no longer reflect low closed-mindedness. At the same time, conservatism may no longer appeal mainly to people motivated to avoid uncertainty, effortful thinking, and change. Today’s conservatives are more likely to present themselves as proponents of change than was once the case. Given these competing hypotheses, we wanted to test how differences in conservatives’ and liberals’ self-descriptions have evolved over time.
We conducted a meta-analysis over a 71-year period (1948-2019). We obtained all data sets we could find (published or unpublished) that assessed the correlation between political conservatism-liberalism and various self-report measures of closed-mindedness. We found hundreds of such data sets, which we analyzed in various ways, with the bulk of the data coming from the U.S. In, and outside of, the U.S. we confirmed that conservatives score higher on self-reported closed-mindedness compared to liberals. However, in the U.S. we found that the originally modest-sized difference has been steadily declining since 1948, and by 2019 was very small. Across 18 non-U.S. countries, we found a different trajectory of change over time, but the differences between conservatives and liberals were the smallest by 2019. These conclusions primarily held for 1) social conservatism, but not economic conservatism, and 2) self-description of political identity, but not attitudinal measures of conservatism-liberalism.
Especially in the U.S. liberals and conservatives have become more similar in their self-described closed-mindedness over time. Based on our data, we cannot say if liberals have become more or conservatives less closed-minded (or both). We also do not believe that the preference for mainstream ideas on the part of closed-minded individuals has changed. Rather, we suggest that the change in the shrinking differences between liberals and conservatives reflect a changing political environment and political messages as to which kinds of beliefs are seen as more mainstream.
Research on the association between closed-mindedness (CM) and political orientation has often assumed an inherent link between the two, such that higher CM orients people toward embracing conservatism. Based on a comprehensive meta-analysis,
Recently,
We explore the stability and variability of the association between facets of CM and political conservatism over time. We argue that two changes in the political landscape offer competing expectations as to the strengthening and weakening of the association. First, we highlight how a politically polarized public may now, more than ever, distinguish between conservative and liberal ideals and policies, making the “match” between a person’s cognitive needs and their political ideologies much stronger. To the extent that conservative ideology does appease closed-minded needs, a politically polarized public could indicate an
Alternatively, people with a closed-minded orientation tend to adhere closely to familiar and dominant norms of society. We highlight evidence and discourse in the U.S. suggesting that liberalism may be increasing in dominance and liberal ideals may be becoming much more normative. Thus, the orientation of conservatives (relative to liberals) on dimensions of traditionalism and maintaining the status quo may be shifting. This suggests that any latent affinity between conservatism and CM may be
Research supports the notion that political ideology is associated with cognitive styles or preferences.
We focus here on “closed-mindedness” (e.g., Research by
Why people are motivated to adopt a political identity has been the subject of some debate. Political scientists initially assumed that people adopt ideological positions that correspond to their issue stances (see
The substantive difference between liberalism and conservatism is thought to be one of the major reasons why closed-minded individuals tend to prefer conservative over liberal ideals. One dimension of CM is the need for cognitive closure. People scoring high on this dimension are likely to seize and freeze on information over time; exhibiting a strong desire to maintain existing beliefs and circumstances while actively avoiding changes (
Similarly, several studies have demonstrated a relationship between closed-mindedness and attitudinal correlates of political conservatism. Right-wing authoritarianism (RWA) and social dominance orientation (SDO) are thought to be highly emblematic of political conservatism’s characterization of resistance to change and acceptance of inequality (
It is also worth noting that social conservatism and economic conservatism reflect two different dimensions of ideological values and must be carefully distinguished (e.g.,
Politics is an ever-changing institution. Yet, theorizing on the link between closed-mindedness and conservatism often assumes an
First, we consider the possibility that an increase in political polarization may have sharpened the distinction between liberalism, conservatism, and their attendant values, which should make it easier for a match between cognitive needs and political ideology to emerge. Thus, we would expect a strengthening of the association between conservatism and CM over time.
Alternatively, we consider the possibility that at some point conservatism may have been identified with stability and aversion to change. Yet, over time conservatives may have become increasingly advocates of change, and liberals the advocates of prevailing norms and values. Under this hypothesis, one would expect the association between conservatism and CM to have become weaker over time.
Communication from political leaders is theorized to be seminal in influencing political followers. For instance, from 2007 to 2015, U.S. citizens were largely supportive of a border wall between the U.S. and Mexico to reduce illegal border crossings. However, following Trump’s 2016 election and signature campaign promise to build a border wall, support among Democrats fell sharply, but increased among Republicans (
For people to be aware of the conservative orientation on traditional values and the liberal orientation on social change, they must receive these messages from the political information environment (
Research suggests that elite messages have been increasingly coherent with their political alignments. In the U.S., Democratic and Republican congressional officials have become more tightly aligned with their respective parties: Democrat leaders are more liberal than they ever have been in the past, and Republican leaders are more conservative (
Two caveats bear mentioning. Much of the research on political polarization focuses on the United States. Though polarization has been diagnosed in other countries as well (e.g.,
Political change can represent a source of uncertainty and hardship. How an individual responds to changes in the environment can vary as a function of their personal motivations. People with a closed-minded orientation generally resist changes in their environment and tend to prefer what is familiar (
Conservatism in other cultural contexts represents a different set of ideals. In post-communist countries, researchers have uncovered an association between facets of closed-mindedness and
In summary, the historical dominance of political ideals can impact their appeal to closed-minded people. Closed-minded individuals prefer what is familiar due in part to their desire to avoid change. The aforementioned research demonstrates how conservatism and liberalism may vary in their dominance across different cultural contexts. Yet, certain streams of research also suggest that the dominance of any ideology within a single culture can also change.
In “The Emerging Democratic Majority,”
Even rhetoric among conservative leaders appears to echo these changes toward a Democratic majority. Rather than opposing change and endorsing the status quo, conservatives often present themselves as
Above, we provide contrasting perspectives on the possibility of a weakening versus strengthening association between closed-mindedness and political orientation. Yet, we would be remiss to acknowledge that change does not have to occur monotonously in one direction. There may be periods when the association between closed-mindedness and conservatism/liberalism strengthened, whereas at other times it might have weakened again. Although we are unable to make any predictions as to when either type of development may have occurred, this study makes it a priority to not only test for linear change over time, but also allows for the possibility of curvilinear change.
The current research seeks to test two competing hypotheses regarding the relationship between closed-mindedness and political conservatism over time. Informed by the political development of political polarization in the United States, our first hypothesis suggests that the association between closed-mindedness and conservatism has increased over time. We contrast this with the prediction that, with the increasing dominance of liberal ideals in the United States, the association between closed-mindedness and conservatism has decreased over time. In doing so, we carefully distinguish measures of general/social conservatism and economic conservatism. Our focus is on self-descriptions of “conservative” or “liberal,” which have been the focus of most of the research on political orientation, among other things because such simple labels are the most potent predictors of worldviews and attitudes (e.g.,
To examine the generality of any observed changes, we also track the changing association between conservatism and closed-mindedness across the world, i.e. in all other countries for which data are available. Moreover, because RWA and SDO seem to capture critical aspects of
The current meta-analysis focused on self-identified political ideology as an individual difference variable and sought to relate it to measures of closed-mindedness (e.g.,
We used a wide variety of search procedures. First, online searches were conducted for a variety of research domains, including psychology (PsycINFO, PsycARTICLES), sociology (ProQuest Sociology), political science (ProQuest Political Science), and dissertations (ProQuest Dissertations & Theses Global). Google Scholar was also utilized due to differences in proprietary algorithms compared to other databases. When searching for articles, several search term combinations were used. Because the current work requires that studies assess both political ideology and measures of closed-mindedness, terms and phrases associated with both kinds of measures were often searched simultaneously. Political ideology terms included words or phrases such as: political, conservative, liberal, Republican, Democrat, left-wing, right-wing, right-wing authoritarianism (RWA), and social dominance orientation (SDO); which were combined with socio-cognitive words or phrases such as need for cognition, intolerance of ambiguity, cognitive complexity, etc. In all, approximately 120 combinations of search terms and phrases were used when recruiting studies. Second, we sought data by extracting relevant studies from past meta-analyses, such as
Overall, our search produced 234 articles, book chapters, and datasets with 341 studies conducted between 1948 and 2019 (see Appendix A in the Countries (and total coefficients extracted) included: Australia (
One of the current authors extracted information from individual studies and datasets. Firstly, because of underlying differences between general (often “social conservatism”), economic conservatism, and attitudinal dimensions of political ideology, each was coded separately. Because the type of conservatism measured was stated explicitly, there was no ambiguity for coding purposes. We extracted a total of 116 coefficients pertaining to economic conservatism, 533 coefficients pertaining to general conservatism or some subtype of conservatism (i.e., social, cultural, moral, and religious), 173 pertaining to RWA, and 98 coefficients pertaining to SDO. In our analyses involving RWA, we only included coefficients based on a version of
Because a large number of data points were from the United States, we decided to report results separately for the U.S. and all remaining countries. A combined analysis is reported as part of Appendix B (
In our meta-analytical approach, we relied on the R package
Whereas the above approach was suitable for the U.S. data, in the analysis of international (non-U.S.) data, country of origin must be considered another source of interdependence. That is, two randomly chosen data points chosen from the same society can be assumed to be more similar to each other than two randomly chosen data points from two different societies. Because country represented another source of interdependence, this was also modeled as a random effect in a cross-classified model. However, because
In all models, we tested to what extent there were differential associations between closed-mindedness and social and economic conservatism. We examined both linear and curvilinear (quadratic) change over time. To reduce collinearity between the linear and the curvilinear terms over time, we centered the time variable (e.g.,
All analyses were carried out using Fisher
For the U.S., correlations between social conservatism and economic conservatism, on the one hand, and closed-mindedness, on the other hand, were available from 1948 to 2019. An initial analysis revealed that, across all measures, types of conservatism, and years, the estimated
More important, there was a linear decrease in the strength of the association between social conservatism and closed-mindedness,
The interaction between type of conservatism and the linear trend over time was not reliable,
We also introduced a curvilinear term, which we allowed to interact with type of conservatism; however, neither this curvilinear trend itself nor its interaction with type of conservatism was statistically reliable. This conclusion was reached regardless of the year on which we centered—a necessary consequence of the fact that there was no discernible curvilinear trend in the data. Instances where curvilinear trends were similarly excluded from models presented below indicate that their inclusion did not reliably improve our model estimates.
Internationally, correlations between social conservatism and closed-mindedness were available from 1948 to 2019, though correlations between economic conservatism and closed-mindedness were available only from 1959 to 2015. The Year variable was centered on 1977, which was when the association between closed-mindedness and social conservatism was at its maximum. Across all measures, types of conservatism, and years, the average
At its peak in 1977, we observed a substantial correlation between closed-mindedness and social conservatism,
The curvilinear estimate did not appear to differ significantly between social or economic measures of conservatism (
We applied the same approach as applied to the measures of self-reported political ideology to the correlates of our attitudinal measures, RWA and SDO. Again, we provide separate analyses for the U.S. and the rest of the world, simply because the U.S. was again the single largest source of data. Tests of curvilinear trends were performed as well, but are only reported for non-U.S. data. In the U.S., curvilinear trends did not contribute to any model estimates and were excluded from analyses accordingly.
Correlations between measures of closed-mindedness and RWA were available from 1995-2018. The Year variable used in this model was centered on its mean (
Outside of the U.S., correlations between closed-mindedness and RWA were available from 1987-2017. The year variable used in our model was centered on its mean (
Correlations between closed-mindedness and SDO were available from 1998-2018. Year of data collection was centered on 2011, the mean year of the available data. Across all years of available data, the average
Internationally, correlations were collected from 1996-2017. The Year variable was centered on 2008, the mean year of data collection. The average
Our meta-analytic findings indicate that the association between conservatism/liberalism and closed-mindedness has changed over time, such that it decreased in linear and non-linear ways from 1948 to 2019. In our U.S. subset, self-described socially liberal and self-described socially conservative individuals appear to have become more similar over time, as indicated by a decreasing correlation between conservatism and closed-mindedness. Note, however, that the correlation is still statistically reliable and positive, indicating a continued tendency for conservatives to be more closed-minded even when the strength of the association varies over time.
This finding supports the hypothesis of a weakening association, which argues that the rising strength of liberalism and rhetoric calling for change among conservative elites might be contributing to a weakening association over time. Our findings do not support the competing hypothesis that any increase in political polarization over time is paralleled by an increase in the association between closed-mindedness and conservatism. In other words, even though several scholars have documented an increasing polarization of U.S. society (e.g.,
Instead, our findings are more likely the results of the increasing dominance of liberal ideals and changes in political discourse among conservative leaders that might not appeal to people with a closed-minded orientation. We note early on that people with a closed-minded cognitive style prefer familiar and dominant norms of society and prefer to avoid change in general (
Echoing these changes, conservative leaders in the U.S. have become increasingly vocal with their concerns regarding the changing social landscape. Conservatives now put forth change agendas aimed at reinvigorating social policy and values that once dominated (e.g.,
Note that our findings for conservatism-liberalism contrast with our findings obtained with regard to RWA and SDO: Whereas we did observe a linear decline in the association between closed-mindedness and self-described conservatism in the U.S., there was a simultaneous
This, however, seems to produce a contradiction. Both RWA and SDO are known correlates of conservatism-liberalism (e.g., Along with other scholars, we view the use of RWA and SDO as measures of conservatism to be a conceptual error. Conservatism-liberalism is akin to the traditional left-right dimension, although not necessarily identical to it. By contrast, authoritarianism can exist on the left or the right, even the concept of left-wing authoritarianism has received wide-spread attention only recently (e.g.,
Notably, findings based on data from outside the U.S. were quite different. Outside the U.S. the association between closed-mindedness and social conservatism varied over time in a Separate analyses over time for Germany and the U.K. were not carried out because Germany only contributed 12 pre-1985 data points, and a single post-1992 data point, with the corresponding numbers for the U.K. being 24 and 6. Because of the 1985-1992 gap in available data, Appendix C (
Our results for U.S. and non-U.S. data were also somewhat different for measures of RWA and SDO. In the U.S., we observed a curvilinear increase in the association between RWA and closed-mindedness. Internationally, there was support for a linear increase. Notably, the difference in the correlations was largest at the earliest point for which data was collected (1995 for the U.S., and 1987 internationally). In addition, whereas in the U.S. there was evidence of a linear increase in the association between SDO and closed-mindedness, no change emerged for our international data. Again, we cannot exclude that the observed change merely reflects an imbalance of the data from different countries. As with any investigation of the association between conservatism and closed-mindedness, the hope is that, over time, additional data will become available, which allows us to investigate such changes over time in countries other than the U.S. At the very minimum, the observed patterns (i.e. over-time decline for symbolic self-placements in the U.S., but an increase for RWA; curvilinear over-time changes for symbolic self-placements in our international data, paired with an increase for RWA) point to the possible existence of distinct causal processes that do shape the association between closed-mindedness and measures of symbolic (or self-report measures) of political orientation, on the one hand, and attitudinal measures of ideology, on the other hand.
Lastly, the current findings partially support underlying differences between economic and social conservatism and their association with closed-mindedness. Whereas we observed temporal changes in the association between social conservatism and closed-mindedness, at least in the U.S., comparable changes in the association between economic conservatism and closed-mindedness were never observed, neither in our U.S. data nor our international data.
Like any empirical study, our meta-analysis suffers from various limitations. The file-drawer problem (
A thornier issue is the comparatively low availability of data in the mid-1980s in the association between closed-mindedness and conservative identities. We submit that research on closed-mindedness and political orientation did not seem to be particularly popular during this period as we had little trouble securing data from before and after the 1980s. We speculate that one reason for the paucity of data may be that the 1980s saw the emergence of a new construct, Right-Wing Authoritarianism (
Previous assumptions about the link between conservatism and closed-mindedness must be treated with caution. Our findings do not contradict the notion, for instance, that closed-minded cognitive styles dispose individuals to gravitate more toward conservatism over liberalism (e.g.
The Supplementary Materials include all the Appendices referenced throughout the article. Appendix A contains the citations of the articles, books, and datasets used to produce our results. We also reported additional results in Appendix B, which includes the same analyses we report in our results section but reflects the two combined datasets (U.S. and International). Because we had a gap in our data between 1985-1992, Appendix C reports separate analyses for pre-1985 and post-1992 data (for access see
The authors have no funding to report.
The authors have declared that no competing interests exist.
The authors have no additional (i.e., non-financial) support to report.