Article Summary
Introduction
The rise of the internet allows citizens to independently select their news sources, enabling citizens to access websites with partisan news and consume their content. This development may result in increased consumption of partisan media and reduced exposure to balanced or out-partisan media. Against this backdrop, the authors pose the following question: Does easier access to the internet increase partisan animosities among citizens? In particular, the authors leverage varying costs between U.S. states for internet providers to offer broadband access to examine whether speedier access to the internet is related to growing affective polarization.
Analytical Approach
Surfing the internet can be related to various citizen characteristics, such as household income or political interest. These characteristics, in turn, may also be related to affective polarization. Hence, to rule out that internet consumption and affective polarization are simultaneously related to other citizen characteristics, the authors propose to leverage a variable that induces varying access to faster broadband internet but is unrelated to affective polarization: Right-of-Way Laws (ROW). The ROW indicator is a composite measure of various regulations that increase the costs and obstacles for providers to offer broadband access between states. Variance in the ROW indicator should arguably be unrelated to partisan affect, and the authors verify that the indicator is not associated with the level of education and other covariates at the state level. The authors then use the variation in broadband access at the county level that is explained by the ROW indicator to predict affective polarization. To measure affective polarization, the authors draw on two surveys conducted in 2002 and 2008 that enable the researchers to geolocate respondents and match them with the geographic broadband access data. The combination of these data allows the authors to estimate the effect of broadband access on affective polarization at the respondent level, induced by varying broadband policies (ROW), and not by other respondent or region-related factors. The figure below illustrates the research design, which is also known as instrumental variables (IV). The authors complement their analysis with comScore data, which provides information about internet users’ search history, and ANES data, which documents respondents’ recent online media behavior. Specifically, the researchers compare users/respondents with broadband access to users/respondents with slower dial-up access. To address concerns over potential systematic differences in sociodemographic characteristics between the groups (broadband vs. dial-up), the authors use several respondent covariates to balance the two groups.
Main Findings
The authors report that broadband access exacerbates affective polarization. The identified effects are substantial: “Had all states adopted the least restrictive right-of-way regulations observed in the data, partisan animus would have been roughly 2 percentage points higher” (p. 6). The authors also find that users/respondents with broadband consume in-partisan-leaning content more frequently than users/respondents with slower dial-up access.
Implications
The authors conclude that the rise of easier access has intensified partisan animus because citizens can self-select their news sources and become more exposed to news content that aligns with their own partisan identification. At the same time, the authors emphasize that access to the internet is not the sole cause of affective polarization. Instead, they argue that partisan animosities had intensified even before the advent of digital media but are aggravated by the selective consumption of partisan-leaning media.
Questions left unanswered
While the evidence presented by the authors provides a compelling answer to the question of whether the emergence of fast internet access intensifies affective polarization, several questions about the supply side, the news outlets, arise. To what extent have online media outlets adjusted to the widespread availability of internet access? Have these online media outlets become more partisan, making it less likely for citizens to encounter other political viewpoints on the internet? Or is this development primarily driven by citizens ignoring alternative media sources and preferring to access news sites that align with their own partisan orientation?
Methods and Analysis
Was the study and its analyses pre-registered?: No
Did the study rely on proxy variables to measure polarization?: Yes
Feelings toward the in-party and out-party presidential candidate
Were standard p-value thresholds used (p<.05 or 95% Confidence Intervals that don’t overlap zero)?: Yes
- Largest p-value presented as significant: 0.05
Were correlational results interpreted with causal language?: No
Limitations / Weaknesses
While the authors report separate results for Republicans and Democrats when evaluating the ANES data, it might have been insightful to investigate potential heterogeneous responses between the partisan groups to the broadband access treatment. Such an analysis would reveal whether both partisan groups are similarly affected by easier access to online news or if the result is mainly driven by one of the two.
Regarding the choice of research design, there are alternative strategies to study the effect of faster internet access on affective polarization that are not discussed in the paper. For instance, one could implement a staggered difference-in-difference design that compares levels of affective polarization before and after broadband access was provided. These design choices might not be feasible due to limited time-series data, but they could have been discussed when justifying the choice for an instrumental variables approach.
Open Data & Analyses
Does the article make the replication data publicly available?: Yes
Does the article make the replication analysis scripts publicly available?: Yes
Article Citation
Lelkes, Y., Sood, G., & Iyengar, S. (2017). The Hostile Audience: The Effect of Access to Broadband Internet on Partisan Affect. American Journal of Political Science, 61(1), 5–20.
Bibtex
1
2
3
4
5
6
7
8
9
@article{Lelkes.2017,
author = {Lelkes, Yphtach and Sood, Gaurav and Iyengar, Shanto},
year = {2017},
title = {The Hostile Audience: The Effect of Access to Broadband Internet on Partisan Affect},
pages = {5--20},
volume = {61},
number = {1},
journal = {American Journal of Political Science}
}