Community Matters: Content Analysis of Children in Immigration Media Coverage, 1990-2020

ABSTRACT

What role do children play in media coverage of immigration, and what might this tell us about coverage of (and attitudes about) immigration more broadly? This study examines U.S. newspaper coverage of immigration from 1990 to 2020. Using multiple content analytic approaches, I find that newspaper coverage of immigration that includes mentions of children: (a) tends to be more positive in net sentiment, (b) tends not to focus on topics of politics and violence, and (c) tends to correlate with topics about family, education, religion, and community. Threat is found to be a regular feature of this news coverage; however, threat language does not vary systematically with the language of childhood or race. In all, these findings point to the salience of (positive) language about community in coverage about immigrant children. These findings are discussed as they relate to the impact of childhood representation in news coverage about immigration in the U.S., and how cueing community, such as family, when evaluating immigrants has the potential to produce more pro-immigrant attitudes amongst the American population.

KEYWORDS:

• Immigration
• children
• news media
• STM
• dictionary
• content analysis

Disclosure Statement

No potential conflict of interest was reported by the author.

Notes

1. About 32% of the data is made up of articles from The New York Times, Washington Post, and Los Angeles Times.

2. Please note that the term “undocumented” was not used in this data collection. Although this was initially an oversight, the word “undocumented” in news is typically followed by “migration”, “immigration” or “immigrant(s)/migrant(s)”, which are captured by the dictionary.

3. The term “minor*” complicates this Child Dictionary. “Minor” can refer both to (1) a person under 18 years of age or (2) an adjective. For this reason, the word minor can create false positives and noise in this dictionary. I chose to keep the word, however, because the discussions of “unaccompanied minors” to the U.S. in news stories are important in this analysis.

4. KWIC is a common tool used in content analysis dictionary building (e.g., Dun et al., 2021; Soroka et al., 2015) that ensures that every word (or almost every word) is used in the relevant context. In this instance, it is in the context of an article about children.

5. Figure A2 in the appendix is the same analysis in Figure 2, except without the New York Times, Washington Post, and Los Angeles Times (337,687 articles total). This analysis is presented because there may be concerns that the results are driven by these three newspapers. As seen in Figure A2, compared to Figure 2, there are no substantial differences in these results.

6. To confirm this suspicion, I ran a sentence-level analysis of threat words in this year with only articles that do not mention children, and found that to co-occurring words include “officials,” “attacks,” “Bush,” “Taliban,” “security,” “Afghanistan,” “war,” “Sept,” “military,” “Bin Laden,” and “FBI.” It should be noted that the word and variations of the word “terror*” is part of the Threat Dictionary, which can contribute to this uptick. .

7. Using the top words from the articles in this period, I created a 9-word 9/11 Dictionary that includes the words: “attack*,” “terrorist*,” “taliban,” “bin laden,” “osama,” “qaeda,” “al-qaeda,” “twin tower*,” and/or “world trade center” to identify articles that specifically focus on the Sept. 11 attacks. The news stories relating to the Sept. 11 attacks were identified if an article uses one or more of the 9/11 Dictionary words.

8. I created a 34-word Latino Dictionary that includes the words “latin*,” “hispanic*,” “beliz*,” “Bolivia*,” “Brazil*,” “Chile,” “chilean*,” “Colombia*,” “Costa Rica*,” “Cuba*,” “dominican republic,” “dominican*,” “Ecuador*,” “el salvador,” “salvador*,” “guian*,” “guadeloupe*,” “Guatemala*,” “guyan*,” “Haiti*,” “hondura*,” “martinique,” “mexic*,” “nicarag*,” “Panama*,” “Paraguay*,” “Peru*,” “puerto rico,” “rican,” “saint-barth,” “saint-martin,” “Suriname*,” “Uruguay*,” and/or “Venezuela*.”

Additional information

Notes on contributors

Guadalupe Madrigal

Guadalupe Madrigal is a Postdoctoral Fellow at the University of Missouri, Columbia, Missouri, USA.