Immigrants and the media

Challenge

There wasn't a specific user in this project
- The users I had in mind were immigrants
- Research and news showed that many Americans felt negatively toward immigrants
User need:
- Understand why many Americans feel negatively toward immigrants
- Pursue a solution for improving relations between these Americans and immigrants
I noticed that the media plays a large role in establishing different viewpoints and communicating information to the general public

Goal

Understand how the media can influence the way Americans think about immigrants.

Why is this project important?

Immigrants make up over 25% of the U.S. population.

If a significant portion of Americans feel uncomfortable or threatened by immigrants, this leads to strained relations between two very large groups in the U.S.

Intergroup tension can contribute to the country experiencing larger waves of mental and physical health issues, which in turn leads to lower happiness, wellbeing, and productivity.

Results

We found that when the media portrayed immigrants as economically successful individuals who were adopting American culture, White Americans liked them more
This finding contradicted our hypotheses, because we expected that economically successful immigrants would be seen as a threat to Americans
The results suggest that if the economic benefits of immigrants are emphasized in the media, at least certain Americans would find them less threatening.

Team

My team consisted of:

My supervisor, Dr. Diane Mackie - Distinguished Professor in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Dr. Kyle Ratner - Assistant Professor in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Dr. Heejung Kim - Professor in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Dr. Nancy Collins - Professor in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Dr. Daniel Conroy-Beam – Assistant Professor in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Youngki Hong – Graduate student in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Megan Reed – Graduate student in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
Jack Strelich – Graduate student in the Psychological and Brain Sciences (PBS) Department at the University of California, Santa Barbara (UCSB)
And my Research Assistants who are all undergraduates in PBS at UCSB
1. Marissa Mahoney
2. Katie Sabini
3. Randi Emmons
4. Katie Peterson
5. Sarah Freeze
6. Sarah Allen-Sutter

Research process

Stage 1: Identifying the problem

The news stories during the time featured stories about many Americans’ negative views about immigrants.

I wanted to explore the theoretical reasons for this phenomenon and what factors may potentially mitigate these negative perceptions.

A literature review showed that Americans were more likely to think negatively of immigrants when they felt that immigrants were a:

economic threat: doing very well economically and thus a threat to their jobs and other resources
cultural threat: holding on to their native culture and therefore changing American culture

The literature review indicated that this phenomenon had only been studied through self-report measures and needed to be studied through other means to ensure that the results were accurate and not influenced by participants’ wishes to present themselves in a particular manner.

I conducted several in-person presentations to my team where I discussed

the problem of tension between many Americans and immigrants
the importance of addressing this problem and promoting more harmonious relations between Americans and immigrants
the reasons for such tension: economic and cultural threat
the gaps in our understanding of this tension:
- economic and cultural threat have been implicated in self-report measures
- the effects of these two types of threat have not been studied in other ways

Stage 2: Creating a study to address the problem

I designed a study that would address the gaps in our understanding of this problem.

Design: Participants would read a news story that aroused

economic threat but no cultural threat
cultural threat but no economic threat
both economic threat and cultural threat
neither economic threat nor cultural threat

Then they would record their negative feelings toward immigrants.

First, I pilot tested my news stories to ensure that my participants were interpreting them in the intended way.

I had American participants come in and read through the news stories I had created
Participants were asked to rate the extent to which they felt economic and cultural threat after reading the materials

Based on the pilot test, I made changes to my materials.

To measure Americans’ negative attitudes toward immigrants without resorting to self-report measures, I decided to use a reverse correlation face categorization task. I was advised on how to create this task by my teammates, Megan Reed, Youngki Hong, and Dr. Kyle Ratner.

To prepare the materials for this task, I:

used one single base image

created 400 paired noise patterns to be overlaid on the image (in each pair, one is the inverse of the other)

overlaid each pair of noise patterns over the same image to create two different pictures of the same face

created 400 unique pairs of these faces by overlaying the 400 unique pairs of noise patterns over the same face

Once I had my 400 unique pairs of faces my intention was to

program them into a survey using Qualtrics
have each pair of faces presented on its own
ask participants to choose the face that looked most like that of an immigrant
- Since there were 400 pairs of faces, participants would have to repeat this task 400 times

How reverse correlation contributes to filling the gaps in our understanding of Americans’ attitudes toward immigrants:

After each participant makes their choice between 400 pairs of faces, the chosen faces are compiled across condition
An average “immigrant” face is created for those who were exposed to:
- economic threat but no cultural threat
- cultural threat but no economic threat
- both economic threat and cultural threat
- neither economic threat nor cultural threat
If participants in one particular condition feel more positively toward immigrants than other participants, the average face in this condition would be more positive (e.g., more pleasant looking)

After finalizing the news stories and creating the reverse correlation task, I ran a heuristic evaluation where I had my team members who are experts in the field (Dr. Diane Mackie, Dr. Heejung Kim, Dr. Kyle Ratner, Megan Reed, and Youngki Hong) look over my study and identify any issues with its design.

Based on their suggestions, I made changes to my materials and launched the study.

Stage 3: Obtaining face data

Based on standards for reverse correlation face categorization studies, ~180 American participants were recruited for this experiment.

I trained my undergraduate Research Assistants to guide participants through the study.

Participants arrived at the lab and were asked to sit in individual cubicles. They were then asked to follow the instructions on the computer and open the door to the cubicle if they had any questions.

The participants then followed the instructions on the Qualtrics survey and read the news story and ran through the reverse correlation face categorization task.

Following this lab study, participants were debriefed and compensated by the Research Assistants.

Once the data had been collected, I engaged in data cleaning and manipulation:

I programmed a script on R to run through each participant’s time taken for the reverse correlation face categorization task
This script flagged participants who took too long or too fast to make decisions (based on task standards)
I removed the data from participants who seemed to be distracted during the task on the basis of the R script output
I then separated the face choices of participants by the condition they belonged to and compiled the faces

Stage 4: Understanding face data

The next step of reverse correlation face categorization is to get a new set of participants to rate the compiled faces on a set of personality traits.

I created a survey where each participant was asked to rate each composite face on 13 different traits:

Trustworthy
Attractive
Dominant
Caring
Confident
Emotionally stable
Responsible
Intelligent
Aggressive
Mean
Weird
Unhappy
Sociable

Idea behind face ratings: if participants in one particular condition have more positive feelings toward immigrants, their composite immigrant face would look more positive, and therefore be rated more positively by the new set of participants.

After collecting this survey data, I reverse-scored the negative traits and averaged all the traits to create a composite of positivity. I then ran a reliability analysis using a script in R on this composite to ensure that I was justified in grouping all the traits with each other. The Cronbach’s 𝛼 was 0.86, which was high enough and indicated reliability of the grouping of traits.

Next, I ran a within-subjects ANOVA to see if reading a news story that influenced economic and/or cultural threat regarding immigrants predicted the original participants’ experienced positivity toward immigrants (which would be expressed by how positive the composite immigrant face in their condition looked).

Interestingly, the results contradicted our hypotheses. We had expected that Americans would like immigrants more when they were framed as being neither a cultural nor economic threat; in other words, we expected participants to like immigrants who were not a threat to their jobs but were also adopting American culture.

To the contrary, the immigrants that were liked the most were those who were portrayed as doing economically well while adopting American culture.

Given these interesting results, I next engaged in advocating for my research. I first presented my findings to my team, followed by presentations in conferences, where I presented my research to individuals unfamiliar with both the methods I used as well as the subject area. The image below is from my presentation of the results. I used the cartoon characters to indicate whether immigrants were portrayed as economical successful or unsuccessful (economic threat or not) and as adopting American culture or not adopting American culture (cultural threat or not).

Constraints

I had a few constraints in this project.

Given our limited timeframe, we could not focus on Americans of other races/ethnicities
- It would be interesting to see how non-White Americans think about economically and/or culturally threatening immigrants
Looking back at the materials, we recognized that the media articles that showed immigrants as not economically threatening, described them as doing very poorly
- We discussed that in a way this may have been viewed as an economic threat as well, since economically unsuccessful immigrants would need government support.
- Further research is needed to see if Americans truly embrace economically successful immigrants and whether this stems from their belief that such immigrants would benefit the country's economy

Immigrants and the media

CONTACT