Introduction

The current political climate in the United States is extremely polarized. “Political Polarization” refers to the idea that there is a growing divide between individuals and groups based on their political beliefs (Geiger, 2014). Political parties are more divided than ever, as increased hostility amongst politicians has also led partisans to become distrustful of those in opposing parties and their views. Political polarization leads to a distrust in the government, political instability, and decreased cross-party interactions. As humans, we tend to seek out information that approves or reinforces our own beliefs, and also reject things that go against what we think is right. Therefore, the polarization is exacerbated by the influence of biased news sources, as they further reinforce our own views, and also effectively alienate those in the opposing political party.

Given how important news sources are in shaping people’s political beliefs, this paper will ask: “When reporting on the dissolution of the Department of Education in the United States, is there an overlap in language amongst news articles from different sides of the political spectrum?” The paper will focus on 5 main news sources picked from the Ad Fontes Media Bias Chart, 2 that are hyperpartisan right/left, 2 of which are skewed to the right/left, and 1 that is unbiased. It will then evaluate each source by the level of bias it presents, to see if there is any common language that emphasizes the political divide or any language that unbiased sources tend to use.

Literature Review

America is more politically polarized than ever. This is due to a number of reasons, all of which lead to hostility between political parties. There are increasing divisions amongst government officials when it comes to topics such as abortion or gun control, with each side believing that their view is the “correct one”. Voters thus have misconceptions about the opposing party. In fact, those who are the most politically and civically active often hold the least accurate views of the other side’s beliefs (Kleinfield, 2023). Ultimately, this leads Americans to think that there are fewer shared policy beliefs with the other political party, effectively emphasizing polarization.

This leads to aggression between the main political parties. Humans tend to think of themselves as representing a broad group of socioeconomic and cultural backgrounds, instead of realizing that they each are made up of their own distinctive traits (Iyengar et al., 2019). This type of behavior can be tied into polarization in the United States. When voters choose to identify with a specific party, whether that be Democratic or Republican, they participate in partisanship. In so doing, we divide the world around us into two groups, the “in-group,” which is the party we are in, and the “out-group,” the opposing party (Iyengar et al., 2019). Psychological studies show that we tend to hold more positive feelings for the in-group than the out-group (Iyengar et al., 2019). This effectively illustrates political polarization, as people simply believe that the party they identify with is the better one, leading to increased hostility with others who don’t share the same beliefs that they do (Tajfel & Turner, 1979). In the end, this leads to fewer shared views amongst voters in the United States, emphasizing the divide.

It is also important to note the role that mass media plays a role in polarization (Stroud, 2010). News sources that lean to the far left or right tend to deliberately choose what types of information to report or conceal, in order to portray the preferred party in a better light. Certain language used can paint certain politicians or policies in a better light, which ultimately biases the reader to a certain side. Common strategies used to do this include adjectives such as brutal or heroic, or figurative language such as betrayal to dramatize a certain event. This also underscores the importance of media responsibility and the need for well-educated Americans. Instead, this emphasizes the party divide as viewers of these news channels are only shown one side of the spectrum, further leading them to believe that their party is the more “correct” or “better” one. A study done by Jinhe Yang (2023) found that biased news sources, such as CNN or Fox News, showed their ideological preferences for certain parties when reporting on environmental issues through their framework, theme selection, and object selection during their news segments/reports.

Due to biased news reports, Americans can be led into an “echo chamber” (Pariser, 2011). People tend to seek out information that confirms our own beliefs and ideas, and avoid those that go against what they think is true. Online, users tend to join groups or interact with content that is formed around a shared narrative with others who believe in the same things we do. With this, social media companies know that the human attention span is limited, and feed algorithms will thus only show content based on what we have interacted with in the past or have been previously exposed to. Therefore, we are pushed more content that further reinforces our own ideas and also alienates the opposing party and their views. Echo chambers can act as a mechanism to reinforce an existing opinion within a group, and move the entire group to more extreme positions (Cinelli et al., 2023). An echo chamber can thus be broadly defined as “environments in which the opinion, political leaning, or belief of users about a topic gets reinforced due to repeated interactions with peers or sources having similar tendencies and attitudes” (Cinelli et al., 2023).

With all these factors in mind, the overall result amongst voters is increased hostility between party lines. Ordinary Americans have become extremely distrustful of the opposing party; Democrats and Republicans both say that the other party’s members are hypocritical, selfish, and close minded, which leads to a lack of interaction between party lines (Iyengar et al., 2018). The absence of communication reduces the understanding for those in the out-group. In fact, partisans tend to view those in the opposing party and their views as “immoral” (Iyengar et al., 2018). A study done by Pew Research Center found that about half of Republicans (47%) and slightly more than a third of Democrats (35%) said those in the other party were a lot or somewhat more immoral than other Americans.

Methodology

To conduct the research portion of this study, the Linguistic Inquiry and Word Count (LIWC) tool was used to conduct content analysis. LIWC is a reliable research tool used to systematically analyze written or verbal communication such as articles, speeches, or even social media posts. Therefore, LIWC was picked as it can be used to analyze the amount of times a certain word comes up in a text, the amount of times negative or positive language appears in a text, as well as to analyze sentiment, moral language, and emotional tone. With this, this research paper focused on specific categories that LIWC provides, which would be positive and negative emotion words, emotional tones, and the frequencies of certain words amongst texts from each side of the political spectrum.

Five different news sources have been chosen from the Ad Fontes Media Bias Chart, which is a resource used to determine how biased a source may be based on a number ranking. Each news source on the chart is plotted on a graph, with the horizontal axis representing the bias rating, and the vertical axis representing the reliability rating. During this selection, only written language in the form of news articles were picked as it would be easier for LIWC to analyze. To represent a variety of sources across the political spectrum, five news sources across the chart were picked based on where they fell on the bias and reliability scale.

To have a consistent topic with each source, all articles used in this analysis were covering the dissolution of the Department of Education in 2025. The articles were put into the LIWC tool, specifically into the LIWC Analysis and Word Frequencies tools to analyze their languages. The five news sources picked include: MSNBC (Strong Left, -13.89 bias, 34.60 reliability), CounterPunch (Hyper-Partisan Left, -21.65 bias, 21.10 reliability ), Fox News (Strong Right, 11.07 bias, 35.31 reliability), The Western Journal (Hyper-Partisan Right, 19.50 bias, 21.26 reliability), and USAFacts (Balanced Bias, -0.05 bias, 50.24 reliability). A negative bias score indicating a liberal leaning, and a positive score indicating a conservative one. Further, the higher the reliability score, the more reliable the news source is. Comparison was done in five different categories, which included comparing sources from the same side, as well as comparing them from opposite ends of the political spectrum: Hyper-Partisan Left & Skews Left, Hyper-Partisan Right & Skews Right, Hyper-Partisan Left & Hyper-Partisan Right, Skews Left & Skews Right, and every article grouped together with the unbiased article added. If common language is found, it could indicate that certain linguistic patterns could contribute to political polarization.

Results

This section explores the results of the LIWC tests run on each article. The first section analyzes the word frequencies in each source, which looks at how often certain words show up between each article. Then, the emotional tone of the articles are compared using a numerical score from LIWC, which measures the presence of specific emotions within the text.

Word Frequencies

First, Word Frequencies amongst each source were compared. This includes words that came up most often between the articles compared, followed up with how many times the specific word showed up. Between the CounterPunch and MSNBC articles (Hyper-Partisan Left & Skews Left), words that were used most often included education (19), school (16), students (15), federal (14), department (11), schools (9), and budget (9). Whereas words that only appeared in one of the two articles or less frequently would include Trump (9), states (8), congress (6), America (6), and funding (6).

The next two sources that were compared were the WesternJournal and Fox News articles (Hyper-Partisan Right & Skews Right). Words that were used frequently or came up in both articles included education (35), department (25), scores (10), students (9), trump (9), math (9), and results (8). Words that were used less frequently or only in one of two articles include time (8), company (7), countries (6), student (5), lowest (5), and failing (5).

After this, the CounterPunch and WesternJournal (Hyper-Partisan Left & Hyper-Partisan Right) articles were compared. Words that came up most frequently or in both articles included education (16), federal (14), school (12), doe (10), students (9), and department (7). Words that were only used in one of two articles or came up less frequently include budget (7), funding (6), schools (1), local (1), and Trump (1).

Then, the MSNBC and Fox News (Skews Left & Skews Right) articles were compared. Words that were most frequent or appeared in both articles included education (38), department (29), students (15), Trump (13), states (11), and school (10). Words that were used less frequently or only appeared in one of two articles included scores (8), time (8), children (8), company (7), math (7), congress (6), failing (5), and results (6).

The very last category would be every news article used above, with an unbiased article from USAFacts added in. It is important to compare each source with the unbiased source as it gives a sense of how biased the article truly is. The words used most frequently were education (90), department (74), federal (48), students (28), and programs (14). Words that were used less often or only appeared in certain articles out of the five would include spending (31), budget (25), school (22), Trump (18), office (13), scores (11), costs (5), and resources (5).

Emotional Tone Test

For the second category, the number of emotional words, as well as positive or negative emotional words, were analyzed. In this category, a higher score indicates a stronger presence of a certain trait. For example, an authentic tone score of 50 reflects greater authenticity than a score of 20. The first category was between the CounterPunch and MSNBC articles (Hyper-Partisan left & Skews Left). CounterPunch had an authentic tone of 52.75, whilst MSNBC had a score of 11.33, meaning that the CounterPunch article is perceived as more honest/unfiltered. Both articles had positive tones, with CounterPunch having a score of 1.19 and MSNBC with a score of 3.09. They also had negative tones present, with CounterPunch having a score of 0.64, and MSNBC with a score of 1.39. Both articles had various levels of emotional tones as well. In the overall emotion category, CounterPunch had a score of 0.37, whereas MSNBC had a score of 0.4. When it came to positive emotions, CounterPunch had a score of 0, and MSNBC had a score of 0.3. In the negative emotion category, CounterPunch had a score of 0.27, and MSNBC had a score of 0.1. It is also worth noting that CounterPunch had a 0.18 emotional anger score, and MSNBC had a score of 0 in this category.

The next two articles compared would be WesternJournal and Fox News (Hyper-Partisan Right and Skews Right). Fox News had an authentic tone of 36.09, whilst the Western Journal had a score of 23.13. When it came to positive tones, Fox News had a score of 2.05, and the Western Journal had a score of 2.25. Whereas when it came to negative tones, Fox News had a score of 0.82, and Western Journal had a score of 0.19. Fox News showed an overall emotional score of 0.49, and Western Journal had one of 0.38. Furthermore, Fox News had a positive emotion score of 0.25, and Western Journal had one of 0.19. The negative emotion category showed Fox News to have a score of 0.19, whilst the Western Journal had a score of 0. Lastly, when it came to anger, Fox News had a score of 0.08, whilst the Western Journal had a score of 0.

Afterwards, the CounterPunch and Western Journal Articles were compared (Hyper-Partisan Left & Right). CounterPunch had an authenticity tone of 52.75, and Western Journal had a score of 23.13. Positive tones showed that CounterPunch had a score of 1.19, and Western Journal with a score of 2.25. Negative tones on the other hand gave CounterPunch a score of 0.64, and the Western Journal a score of 0.19. Overall emotion showed that CounterPunch had a score of 0.37, and Western Journal had a score of 0.38. CounterPunch had a score of 0 when it came to positive emotions, and the Western Journal had a score of 0.19. Negative emotions showed CounterPunch to have a score of 0.27, whereas Western Journal had a score of 0. Anger emotion scores thus showed that CounterPunch had a score of 0.18, and Western Journal had a score of 0.

In addition, Fox News and MSNBC (Skews Right & Skews Left) were also compared. Fox News had an authenticity score of 36.09, and MSNBC had a score of 11.33. Next, Fox News had a positive tone score of 2.05, and MSNBC had a score of 3.09. Negative tones showed that Fox News had a score of 0.82, and MSNBC had a score of 1.39. Overall emotion gave Fox News a score of 0.49, and MSNBC a score of 0.4. Fox News had a positive emotion score of 0.25, and MSNBC had a score of 0.3. The negative emotion category showed Fox News to have a score of 0.16, and MSNBC to have a score of 0.1. Fox news had an anger score of 0.08, and MSNBC had a score of 0.

Lastly, every single article was compared together, with the addition of an unbiased article from USAFacts. Fox News had an authenticity score of 36.09, MSNBC had a score of 11.33, CounterPunch had a score of 52.75, Western Journal had a score of 23.13, and USAFacts had a score of 16.06. Fox News had a positive tone score of 2.05, MSNBC had a score of 3.09, CounterPunch had a score of 1.19, Western Journal with a score of 2.25, and USAFacts with a score of 0.21. Negative tones showed that Fox News had a score of 0.82, MSNBC a score of 1.39, CounterPunch a score of 0.64, Western Journal a score of 0.19, and USAFacts a score of 0.21. Overall emotion showed Fox News a score of 0.49, MSNBC a score of 0.4, CounterPunch a score of 0.37, Western Journal a score of 0.38, and USAFacts a score of 0. Fox News had a positive emotion score of 0.25, MSNBC a score of 0.3, CounterPunch a score of 0, Western Journal a score of 0.19, and USAFacts a score of 0. Negative emotions showed Fox News to have a score of 0.16, MSNBC a score of 0.1, CounterPunch a score of 0.27, Western Journal a score of 0, and USAFacts a score of 0. Lastly, Fox news had an anger score of 0.08, MSNBC a score of 0, CounterPunch a score of 0.18, Western Journal a score of 0, and USAFacts a score of 0.

Discussion

When it comes to word frequencies amongst different articles from different sides of the political spectrum, a large overlap of language was present. Whenever two articles were compared, whether that be from the same biases or completely different, there would always be key terms that were found in both sources. Although, it is worth noting that this overlap is likely not due to similar language or opinions, but purely because of the nature of news reporting. When reporting on the dissolution of the Department of Education, every news source must use certain key terms such as dissolution, education, or funding. Strong, emotional words that are often used to inflict political bias such as betrayal or radical were not seen as an overlap between different sources. Therefore, it would make sense that there were several words that were shared amongst each source, but this is not because of shared opinions or linguistic patterns utilized to favor a certain political party.

Frequencies

An interesting discovery made when conducting the word frequencies test on LIWC is that certain words associated with a specific political party were not always used in every article. For example, the word Trump was only used in three out of five articles (MSNBC, Western Journal, and Fox News), and the name only came up eighteen times in total. President Trump is strongly associated with the Republican Party, yet is left out of certain articles regarding a decision that he made whilst in office. This is interesting as this could have been done for two main reasons. The first being that certain news sources choose to leave political figures’ names out of their articles in order to try to keep a positive image for their party when reporting on a policy that is generally disliked by the majority. This allows the preferred party to be distanced from backlash. Alternatively, this inconsistency can be associated with tailoring content to biased audiences. News outlets with opposing views could leave polarizing leaders’ names out of articles in order to downplay their importance and power in specific decisions, thus aligning with their readers’ expectations.

Emotional Language

When it comes to analyzing the emotions in a text, there are large discrepancies between articles from each side of the political spectrum. First, it can be noted that Hyper-Partisan articles tended to have quite high authenticity tones, as seen in the scores presented for CounterPunch and Western Journal. This could be due to the fact that many of these articles tended to be extremely biased, and generally come from one’s personal opinion. It is also noted that these sources do generally contain selective or incomplete information to create unfair persuasion to one side of the spectrum, according to the Ad Fontes Media Bias chart.

When it comes to positive or negative tones, the dissolution of the Department of Education is a policy that the Republican party generally supported, indeed, right leaning sources had slightly less negative tones when compared to the left sources.

All sources had emotional phrases or words present, except USAFacts (unbiased=0), as they all had a numerical score above 0 when it came to their overall emotion score. Lastly, when it came to emotional anger scores, only Fox News and CounterPunch had scores above 0. When it came to the CounterPunch article, this could be due to hostility involving the Republican Party and their decision to cut funding for the Department of Education. On the other hand, when it comes to Fox News, this anger could be directed towards the Democratic Party and how they are trying to keep the Department of Education from being dissolved, as it does not align with their political biases.

In this research alone, the dissolution of the Department of Education is a policy that is generally favored by the Republican Party. Thus, many articles that were leaning to the left did show a higher negativity/anger score in both emotions and tones sections, with the left leaning scores adding up to a mean value of 0.43, whilst the right articles had a value of around 0.21. Therefore, in this specific instance alone, articles with political biases to the left did have a higher rate of negativity present in their language. All of this can be seen as contributions to the high rates of political polarization in the United States.

Conclusion

In general, it can be said that there is overlap in language amongst new sources from different sides of the political spectrum, especially when reporting on the dissolution of the Department of Education. This can be seen when sources from each side use the same words when reporting on the case. But, it is also important to note that this result is expected, as writers must use some similar key words and terminologies when reporting on the same thing. After all, it would be impossible to report on the dissolution of the Department of Education without using words such as funding or school. Yet, emotional words that are used to create political divide or polarization are not seen amongst the articles.

It is also worth noting that certain terminologies that were expected to come up in every article did not show up in all. Terms such as Trump only came up in three out of five articles, which was unexpected as the President was one of the major supporters of the dissolution of the Department of Education. This can be drawn to two conclusions. One such that certain articles may choose to leave political figures associated with a certain party out of their reports in order to give them a better image. The other suggests that it is possible that they are left out in order to give tailored content to biased audiences. News outlets may choose to not report specific names in order to downplay their role or importance in a major decision, thus aligning with their audiences’ biases.

On the other hand, when it comes to emotional language in news articles from different sides of the spectrum, there also happens to be an overlap. All articles analyzed displayed some form of emotional language, whether that be positive or negative, except for the USAFacts article. This shows that there is some sort of positive or negative language present in each piece of writing, which can effectively skew the audience’s perceptions of certain sides of the political spectrum. Thus, it can be shown that there does appear to be some sort of overlap in the way news writers from each side of the political spectrum use written language to shape stories in their favored political party’s favor.