Political Orientation and Institutional Trust: A Quantitative Analysis of Systematic Differences

This paper investigates whether political orientation influences trust in state institutions like the police, military, and legal system. Through survey data, the researchers demonstrate that right-leaning individuals generally report higher institutional trust than their left-leaning counterparts.

Introduction

This research paper aims to investigate whether or not political orientation is linked to any systematic differences of trust in government institutions, including the police, military, legal systems and the government itself, using statistical analysis based on data collected through a survey. We aim to demonstrate this by displaying how people that see themselves as leaning more towards the right on the political spectrum than left display a higher trust in the aforementioned government institutions compared to their left leaning counterparts.

In our modern society we have seen a swift global rise of populism, growing division of politics. That is why we believe it is more important now than ever to understand how one's political orientation may influence the trust in government institutions. This new political environment has a great impact on the stability of our democracies, and understanding why certain political affiliations lessens peoples trust in said democratic institutions is vital for our ability to protect the democracy.

The data later used in this paper was collected via a survey designed by students of the course “Quantitative Research Methods” at Linnaeus University in Växjö, Sweden. The survey targeted issues and questions regarding Human Security and was tested by the students of the course prior to its release and distribution to the public. In total, the survey consisted of 46 questions, covering a range of issues linked to human security.

The survey was distributed online using Google Forms and targeted any adult living in Sweden or abroad, with hopes of reaching a wide audience representing as much as possible of the population. Despite this large target, due to various limitations, mainly acquaintances, family members and local strangers were involved and able to participate. To take ethical issues into account all participants were informed that their responses were to remain anonymous. This approach resulted in a convenience sampling and is therefore not representative of any broader population, as participants tend to be linked to those that created the survey and its distribution (Bryman, A. 2012, p.201). As such, any results are not to be generalized with confidence. In total, the survey generated a total of 496 responses, with 201 men, 281 women and 14 other/preferred not to answer.

The convenience sampling method is an important limitation to this paper and its data. This is due to the fact that convenience sampling does not generate a perfect sample pool of participants needed in order to make any generalized conclusion, as explained earlier.

The statistical analysis was conducted using SPSS, a statistical program that helps us analyse the findings in order to explore whether people that tend to lean right in the political spectrum trust public institutions more than those who lean towards the left. This approach allowed us to identify whether or not there are systematic differences in institutional trust based on political orientation.

The Analytical Process and its Weaknesses

Out of the 496 participants, the data we collected from the survey shows that 128 men and 225 women respondents view their political orientation as leaning to the left, while 72 men and 53 women tend to lean to the right, totaling 478 responses. 

For the statistical analysis, we decided not to include respondents that marked that they identified as other, or respondents that preferred not to answer. We did this because the sample size of these two categories were too small for analysis, with only 14 out of 496 respondents indicating as “other” or “preferred not to answer”. 

The question was phrased in such a way that would let the survey participants choose a number ranging from 1 to 10, with 1 being far left, and 10 being far right. The results of this survey is notably weighted towards the left as 73,8% of the responses are leaning in that direction according to the data, i.e. they picked 1 through 5. This is likely linked to the weaknesses of the convenience sampling, but may also be a result of poor design of the question, but it is somewhat irrelevant to the hypotheses being tested. This is due to the fact that even though the respondents are mostly left leaning, it is still enough respondents who placed themselves on the right side of the scale to get a reasonable statistical analysis made, even though the left side may have been compromised due to reasons that will be explained in the next paragraph. Furthermore, our hypothesis aims to understand why right leaning people have a higher degree of trust in institutions, this can still be achieved regardless of the fact that more individuals picked the left side as their political orientation.

Graph 1. Univariate analysis of political orientation.

As we got 478 respondents picking their correlating number between 1 and 10 we created two separate groups which we then analysed further. The first group would be the Left Leaning Views, which is represented by those that have chosen the numbers 1 through 5. The second group is the Right Leaning Views, which includes those who selected numbers 6 through 10. This binary grouping allows us to simplify the analysis by making 10 separate categories into only two. However, it is important to acknowledge that this division may overlook the difficulties of political identity, particularly for those who selected numbers closer to the middle, such as 5 or 6. It is also noteworthy that participants may think of 5 as being the natural center between these numbers, which of course is a false truth and could introduce interpretive bias, meaning our interpretation may not be the intended choice done by the participant. To mitigate this we also did a secondary binary grouping where we disregard those individuals that have picked either number 5 or 6, instead focusing on those that clearly have chosen either left or right as their political orientation, creating a Far Left Group and a Far Right Group. This is however limited by the fact that the sample pool is now much smaller, specifically for the Far Right Group, this category now totals at only 79 respondents. The dispersion of nations within this survey is another factor that has to be taken into consideration, since we know that the political scale differentiates from nation to nation. This means that someone who resides in the USA might have a different understanding of where the political “center” lies compared to what a Swedish person might interpret as the “center”.

In order to generate the trust levels, the respondents were asked to rate their respective trust towards each institution separately by choosing a number ranging from 1 through 7, where 1 is regarded as “no trust at all”, and 7 being regarded as “complete trust”. In the original survey “Media” was also included as one of the institutions respondents could indicate their trust towards. It was on the same 7-point scale as the other options. however we made the decision to not include the results from this specific variable in our research paper. The reason why we decided to not include media in our paper was because we saw it as too separated from the other variables. The other variables such as for example police and military are institutions that are state operated, whereas Media generally is privately owned. There are of course state driven media, for example the Swedish SVT, but the question was not formulated in a way that made any distinction between state-owned Media and private owned Media, therefore we made the decision to not include the results from that specific variable, because we were only interested in state and governmental driven institutions.

Literature Review

As stated by another source later in this paper, there has not been any extensive research into our specific question. Much of the literature that has been written before has chosen to tackle specific questions for example political ideology and trust in vaccines, or political ideology and conspiracy theory belief. One article focuses on the link between political ideology and trust towards police (Hansen, M.A and Navarro, J.C, 2023), which is one variable that we also incorporate, but we also include all other variables of government to try to get a more overarching perspective. This is where our research might fill a gap since much of the other papers done on the subject focuses on just a singular part e.g political ideology and trust in police, we try to broaden the perspective by adding more variables for a more comprehensive study.

The Main Findings

One key take-away from the data that we found was that, in line with our hypotheses, the respondents with right leaning views tend to have higher trust levels towards institutions overall, than those with left leaning views. As we add all these respondents together in the first binary grouping and gather these groups’ mean trust level, based on the “1 through 7-system”, towards each individual institution, the results are as as follows:

This clear and significant difference in trust between right and left leaning views on the political spectrum is a clear indication that our hypotheses may be correct, yet we must look into more variables to further strengthen this claim. Therefore, this paper will also analyse the variations between gender and age as our control groups. What is notable is perhaps the group mean result, the participants with left leaning views have a mean of 3,53, indicating an all around lower score than the middle score of 4, i.e. people within this grouping overall tend to distrust these institutions. For the right leaning views however, the mean figure is 4,15, showing that overall, this grouping is above the middle score, i.e. more trusting towards the institutions in question.

The lowest trust scores received by both sides is attributed to the government, 3,14 for the left, and 3,71 for the right. This result does still strengthen our hypotheses, yet as most respondents answering this survey have a current ruling government whose politics and policies are leaning towards the right on the political spectrum, this may skew these numbers in our hypotheses favour.  We will expand on this issue later in our discussion.

Shown here below in graph 2 is the cumulative results more easily understood to see how the sides differentiates, measuring the mean scores of how respondents on the political spectrum rate their trust in the following institutions: government, military, police and the legal system.

Graph 2. Bivariate analysis: Political orientation and trust in institutions.

As is clearly visible in the graph, the Right Leaning Views trust towards each of these institutions are higher than those of the Left Leaning Views. More specifically, 40,9% of the Right Leaning Views have a high trust of these institutions, while for the Left Leaning views, this number is only 29,8%. If we take our secondary binary grouping, the Far Left Groups’ number is lowered even further down to 26,7%, while the number for the Far Right Group remains the same at 40,9%.

Control Variables

Gender

When analysing the data, we need to have control variables. As we take women and men for instance, there are clear differences between the trust levels of institutions. Overall, when analysing all respondents regardless of political orientation the data suggests that men tend to trust these institutions more than women, as indicated by the higher group mean results across the four institutions:

However, when using our binary groupings to further analyze, an interesting deviation occurs. In the Left Leaning View, men continue to show higher trust levels than women (3.77 vs. 3.43). In contrast, in the Right Leaning Views, women now show higher trust levels than men (4.27 vs. 4.01). This suggests that the relationship between gender and trust in institutions is complex and requires further research to fully comprehend.

Another surprising result is displayed in Graph 3a and 3b below. Men with right leaning views tend to distrust the government more than women with similar views. In contrast, women and men on the left leaning side have a highly similar level of indicated trust. Why men on the right indicate a higher level of distrust compared to their women counterparts is something that would require further analysis. It might be due to the smaller sample size of women on the right, making their percentages skewed. 

In graph 3c and 3d we see the indicated trust levels of right leaning men and women towards the military, and also the left leaning men and womens trust towards the military. Again we see a higher degree of trust from the right leaning side, with a notable spike in graph 3d, where 30% of right leaning women indicated a trust level of 6 (high). This again could be due to the aforementioned smaller sample size. 

Graph 3a. & 3b. Bivariate analysis: Political orientation and trust in government separated by gender.

Graph 3c. & 3d. Bivariate analysis: Political orientation and trust in the military separated by gender.

Age Groups

Another control variable we chose to use is the age of the respondents. We divided the ages of the respondents of each of the first binary groupings into 5 separate categories, 20 or below, 21 through 25, 26 through 40, 41 through 53, and finally 54 and above. In graph 4, below, we analyse the Right Leaning Views trust toward government, and we can see that there are no clear indications that either age group has more or less trust than another. One anomaly that we notice may be that the age group of 20 or below tend to have lower trust than those of the other groups.

In Graph 5, in contrast, we analyse the Left Leaning Views, again against trust of the government. The graph shows, in line with our previous data analysis, that overall, the left leaning respondents show similar trust levels towards institutions regardless of age group.

Graph 4. Bivariate analysis: Age groups with right leaning views and trust in government.

Graph 5. Bivariate analysis: Age groups with left leaning views and trust in government.

Trust In People Versus Institutions

One interesting note in the difference between political beliefs is the fact that on average, our data shows that left leaning people tend to have more trust towards other individuals, than they do towards institutions. On the other hand the reverse is true for people on the right, where the feeling of trust is stronger towards institutions than it is towards other people. We see this in our data where 20,2% of far right respondents indicated that they had low trust towards other people whereas only 11,6% of far left respondents indicated low trust. This is visualized in graph 5 below.

Graph 6. Bivariate analysis: Political orientation and trust in people.

These results may be explained by the idea that conservatives believe that the average human is self-interested and competitive by nature, and also believe that other people might be dangerous and in some cases be potentially harmful. Additionally they don’t expect other people to act in such a way to bolster the interests of the collective (Duckitt, J 2001; Duckitt, J and Parra, C 2004). 

Discussion

The findings in the end were consistent with our original hypothesis that Right Leaning people tend to trust government institutions more than their Left Leaning counterparts. In our research we have looked at a number of different studies that in some way relate to our own. We found that there has not been a great deal of academic research devoted to our exact question or hypotheses. This does not however mean that there is none, since there are quite a lot of academic papers written that are closely related to our own, even though they do not pose the same hypotheses that we do, they are at least strolling in the same forest as we are, so to speak. This fact we then feel lends credence to our question since it seems to be a field of studies that has been largely overlooked. The only other study we found that tackles our own in a direct way also mentions the fact that there has been hardly any studies done on this specific question (Fisher, J.A.V.  2011, p. 3). 

An aspect of that has to be accounted for when posing an hypothesis like ours is, the impact the political leanings of any current government with majority control have in any given nation. By this we mean to say that if the ruling government's political ideology overlaps with your personal ideology, your feeling of trust towards said government will increase. J.A.V. Fischer (2011, p. 27) found in her paper that your trust in the government exponentially increases if the government shares your own ideology, and it also shows a drastic decrease in trust if the current government’s political ideology greatly differs from your own. This result is psychologically explainable if we use the example of in-groups versus out-groups. An in-group is a group that you as an individual feel a sense of belonging to, and it also includes the people that you acknowledge or perceive as belonging to your group. You generally treat people you view as part of your group well, and the members share a positive view of each other. The other side of the coin is the out-group, the out-group is anyone you deem to not belong to your group. These people are viewed negatively by your group and will receive poor treatment when compared to the treatment given to the in group (Ashcraft, D and Treadwell, T 2008). So if you count yourself as a conservative voter and the current sitting government is conservative, your trust will automatically increase since you perceive said government to be a part of your in-group. This means that if viewed from an in-group versus out-group perspective that if the sitting government does not share your ideology, your trust will then in this case decrease.

Conclusion

The findings support our hypotheses that individuals with right leaning views on the political spectrum tend to have higher levels of trust towards government institutions compared to those with left leaning views.

Through the survey and a reasonably diverse group of respondents, we were able to create both univariate and bivariate analyses of the findings. As mentioned earlier, there are clear limitations to this data and research. Such as how a current sitting government may affect outcomes of trust, or how our convenience sampling is a limiting factor, but also how the survey itself may have design flaws in itself. These limitations along with others does make it impossible to generalize the findings to a broader population. The results presented do however indicate a pattern that seems to be strong across our control groups, gender and age, although these variables introduced us to some interesting deviations and findings. For instance how men generally across the political spectrum tend to show more trust toward institutions than women, yet while looking specifically at the respondents with right leaning views, this trend is reversed, and women show higher trust levels. 

In conclusion, throughout this research we have found that measuring trust levels between political orientations is not easily understood, nor analysed, as it is extremely complex. To have better data, further research must be done over a longer period of time and with a variety of governments, both left and right leaning, in order to better understand the dynamics and address the limitations discussed earlier. This is needed in order that policymakers and others can address these issues and work towards strengthening the trust levels, thus strengthening the democratic institutions. 

Reference list

Ashcraft, D. and Treadwell, T. (2008). The Social Psychology of Online Collaborative Learning Social Categorization: In-Groups and Out-Groups. [online] Available at: https://www.wcupa.edu/coral/documents/07in-outgroups.pdf.

Balliet, D., Tybur, J.M., Wu, J., Antonellis, C. and Lange, V. (2018). Ingroup Favoritism among Republicans and Democrats during a US National Election. The Journal of Conflict Resolution, [online] 62(4), pp.797–818. doi:https://doi.org/10.2307/48597317.

Bryman, A. (). Social Research Methods . 4th ed. New York: Oxford University Press Inc, p.201.

Duckitt, J. (2001). A dual-process cognitive-motivational theory of ideology and prejudice. Advances in Experimental Social Psychology Volume 33, 33, pp.41–113. doi:https://doi.org/10.1016/s0065-2601(01)80004-6.

Duckitt, J. and Parra, C. (2004). Dimensions of Group Identification and Out-Group Attitudes in Four Ethnic Groups in New Zealand. Basic and Applied Social Psychology, 26(4), pp.237–247. doi:https://doi.org/10.1207/s15324834basp2604_1.

Fischer, J.A.V. (2011). Living Under the ‘Right’ Government: Does Political Ideology Matter to Trust in Political Institutions?. SSRN Electronic Journal. doi:https://doi.org/10.2139/ssrn.1944169.

Hansen, M.A. and Navarro, J.C. (2023). Bad apples or rotten orchards? Public attitudes of interactions with police and the role of political ideology. Policing, 46(5/6), pp.861–877. doi:https://doi.org/10.1108/pijpsm-07-2023-0098.