Author: Jae Yeon Chang Major: Economics 

In the spring semester of my junior year, I took my first experimental economics class with Dr. Li Hao at the University of Arkansas. In this class, I discovered that economics went beyond just learning concepts and memorizing definitions– it was about asking questions and solving problems. Experimental economics is a growing field that utilizes controlled experiments to study economic questions. For my thesis, I wanted to explore the relationship between media slant and beliefs and behavior about undocumented immigration in an experimental setting. 

Controversial topics in our society today are becomingly increasingly politicized and utilized to evoke a reaction from media consumers. We are seeing a huge increase in 21st century “interpretive journalism,” which allows for more opinionated articles. Studies have shown that slanted media is not improved with competition, which is arguably the most critical component for a free market. What does improve media slant in the market for information, however, is heterogeneity in the market. But we are seeing less and less of this today as media sources market themselves to narrower ideological segments of society. I noticed that as the topic of undocumented immigration became more politicized and opinionated, news sources were focusing more on evoking a reaction out of people to attract them to their articles. I wanted to know how this could affect our beliefs towards undocumented immigration, and I took this question to Dr. Andy Brownback, an Assistant Professor of Economics at the U of A. We came up with a plan throughout the fall semester to not only explore these relationships but collect the data ourselves. He encouraged me to apply for the Honors College Research Grant so that I would have funds to give to participants in my experiment, which revolved around the dictator game that I had learned in my experimental economics class. The dictator game involves participants receiving a fixed amount of money to split between themselves and another party. This is a one-person game in the sense that the “dictator” is the only person making a decision. Economic theory says that the “dictator” should maximize profit by taking all of the money for themselves. However, empirical results have often shown otherwise, suggesting that there is an altruistic component of the decision-making process. With the dictator game, we are able to measure the relationship between altruistic behavior towards undocumented immigrants and beliefs towards undocumented immigration. 

During the spring grant term, I designed a survey in Qualtrics that would help me to measure people’s beliefs and behaviors, before and after an “information intervention” where participants would read an article about undocumented immigration. Then, I was able to recruit participants in to the U of A Behavioral Business Research Lab to use my survey and collect data. In the experiment, the subjects were first asked to complete an initial survey measuring the participant’s beliefs and knowledge about undocumented immigration in the United States. Then, I measured their altruistic behavior with the dictator game. Participants were given $10 for each round and could allocate up to $5 to the other party. Then to analyze the effect of media slant on beliefs and behavior towards undocumented immigration, all participants were asked to choose and read an article relating to undocumented immigration. Three articles were used for this information intervention. One article had an explicitly objective title and came from an objective source. The other two articles had politically charged titles related to undocumented immigration and were from well-known slanted sources. After the information intervention, participants took the same survey of beliefs and played the same dictator game as before to measure the change in beliefs and behavior towards undocumented immigration after reading the article.  

After collecting my data, I ran linear regressions to study the relationships between news sources, changes in beliefs towards undocumented immigration, and changes in behavior towards undocumented immigrants. Dr. Brownback guided me through the process of cleaning data, running regressions and summary statistics, and creating visualizations in Stata, a statistical software used by many economists. With this analysis, we found that people do choose to read articles that confirms their current beliefs. Moreover, subjects who were assigned a different article had a smaller magnitude of change compared to subjects who got their chosen article, but the data does not support a statistically significant relationship. This implies that one article may not be enough to change beliefs or behavior. We also found that donations decreased in the second dictator game for all recipient groups, but there was a larger decrease in donations for documented immigrants and U.S. citizens. This suggests that when given a fixed sum of money, people will express their change in beliefs towards undocumented immigrants by decreasing their donations towards documented immigrants and U.S. citizens rather than increase their donations to undocumented immigrants. Understanding the underlying bias that affects decision making is crucial in interpreting controversial topics today such as undocumented immigration in the United States. This study suggests that one article may not be enough of an impact on beliefs and behavior, but I am grateful for the opportunity to start developing an idea about these connections.  

Because of the Honors College Research Grant, I received the opportunity to get hands-on experience conducting research and collecting data in an experimental setting. It turned my research project into an interactive, holistic learning opportunity instead of a project where I was analyzing data that someone had already collected. More importantly, it helped me realize that I have a true passion for research and experimental economics. I have realized that although this may be the end of my thesis project, it is certainly not the end of my involvement in this kind of research. Sometimes, your research can create more questions than answers. I think this is a topic that deserves more exploration, and I am grateful to the Honors College for opening up this new door.