This analysis applies the first-level agenda-setting framework to explore the way APNews framed coverage of the Trump Administration’s tariffs. While tariffs are generally perceived as an economic policy tool, they also carry political and emotional weight because they can be portrayed either as protection for American industries or as a cause of financial strain. The study focuses on two contrasting frames: Economic Gain, referring to coverage that presents tariffs neutrally or without negative implications, and Economic Burden, referring to coverage that associates tariffs with harm or economic setbacks.
Although this project is guided by framing theory, the first-level agenda-setting code was used because it effectively measures the volume of attention devoted to each frame over time. By comparing the weekly counts of Economic Gain versus Economic Burden coverage, the analysis tests whether one perspective dominated APNews reporting. The paired-samples t-test provides a statistical test of that difference, helping determine if the emphasis on negative framing was significant or simply due to random variation.
The weekly volume of APNews coverage framing tariffs as an Economic Burden will be significantly greater than the weekly volume of coverage framing tariffs as an Economic Gain.
Weekly APNews.com coverage volume of the two tariff frames served as the analysis’s dependent variable. It was measured continuously as the number of stories published per week. The independent variable was frame, measured categorically as either Economic Gain (neutral or positive framing) or Economic Burden (negative framing) and operationalized as containing key words or phrases likely to be unique to each framing of the Trump Administration’s tariffs.
Key words and phrases used to identify stories emphasizing the Economic Gain frame (V1) were: “tariff,” “tariffs,” “trade policy,” “trade deal,” “trade agreement,” “U.S. trade,” “American manufacturing,” “economic growth,” “domestic production,” “jobs,” “economy,” “economic boost,” and “recovery.”
Key words and phrases used to identify stories emphasizing the Economic Burden frame (V2) were: “tariff,” “tariffs,” “cost,” “prices,” “inflation,” “trade war,” “retaliation,” “economic slowdown,” “loss,” “deficit,” “layoffs,” “farmers,” “backlash,” “higher prices,” “hurt,” “decline,” “economic pain,” “instability,” “recession,” “criticism,” “tension,” “uncertainty,” and “global market.”
A paired-samples t-test assessed the statistical significance of variation in weekly coverage volume between the two frames, comparing Economic Burden (V2) to Economic Gain (V1).
The figure below summarizes APNews.com’s weekly coverage of the Trump Administration’s tariffs coded by frame. Across the period analyzed, coverage emphasizing the Economic Burden frame consistently appeared more frequently than coverage emphasizing the Economic Gain frame. This trend suggests that APNews tended to focus on the challenges and negative outcomes associated with tariffs rather than highlighting potential benefits or neutral economic perspectives.
The Shapiro-Wilk test indicated that the pair differences were approximately normally distributed, allowing the use of a paired-samples t-test. The results of the t-test revealed a statistically significant difference between the two frames, with Economic Burden coverage appearing more often on average than Economic Gain coverage. A Wilcoxon Signed Rank test produced consistent results, further supporting the reliability of the finding.
Overall, the analysis suggests that APNews framed tariffs in ways that emphasized their economic costs more than their potential gains. These results align with expectations based on framing theory, which proposes that the salience of particular aspects of an issue can influence public understanding. Future analysis will take a closer look at the timing and context of these coverage shifts to see how they align with specific political or economic events during the period.
| Descriptive Statistics: Pair Differences | ||||
| count | mean | sd | min | max |
|---|---|---|---|---|
| 40.000 | 45.500 | 13.834 | 17.000 | 72.000 |
| Normality Test (Shapiro-Wilk) | ||
| statistic | p.value | method |
|---|---|---|
| 0.9773 | 0.5910 | Shapiro-Wilk normality test |
| If the P.VALUE is 0.05 or less, the number of pairs is fewer than 40, and the distribution of pair differences shows obvious non-normality or outliers, consider using the Wilcoxon Signed Rank Test results instead of the Paired-Samples t-Test results. | ||