Using Internet Search Data to Produce State-level Measures: The Case of Tea Party Mobilization

Abstract

This study proposes using Internet search data from search engines like Google to produce state-level metrics that are useful in social science research. Generally, state-level research relies on demographic statistics, official statistics produced by government agencies, or aggregated survey data. However, each of these data sources has serious limitations in terms of both the availability of the data and its ability to capture important concepts. This study demonstrates how state-level Google search measures can be produced and then demonstrates the effectiveness of such measures in an empirical case: predicting state-level Tea Party movement mobilization. Drawing on existing studies of the Tea Party movement and theories of right-wing and conservative mobilization, state-level Google search measures for anti-immigrant sentiment and economic distress are developed and compared to traditional metrics that are typically used to measure these concepts, such as the unemployment rate and the international immigration rate in their ability to successfully predict Tea Party event counts. The results show that the Google search measures are effective in predicting Tea Party mobilization in a way that is consistent with existing theory, while the traditional measures are not.

The dynamics of public opinion following terror attacks: Evidence for a decrease in equalitarian values from Internet Search Volume Indices

“…We therefore decided to use only single tokens for equalitarian values, which fluctuations reflect active interest for content related to equalitarian (whether pro or con) topics which we chose to interpret as potentially more favourable collective attitudes towards equality. This interpretation is in line with previous work, which demonstrated for instance that fluctuations in the single token 'TeaParty' predicted Tea Party mobilization across US states (DiGrazia, 2017).…”

Latinos’ deportation fears by citizenship and legal status, 2007 to 2018

“…These data are Changes in Latino U.S. citizens' deportation fears may therefore more closely reflect a growing national salience of-rather than observable changes to-deportation policy and practice (10,35,45). Figure 5 uses Google Trends data to measure the U.S. public's online searches for "deportation" (in English and Spanish) across the study period (46). This measure, normalized to range from 0 (lowest) to 100 (highest), evaluates deportation's relative salience to the U.S. public in the months surrounding data collection for each survey.…”

“…The "perceived" context of deportation threat is measured using Google Trends data, which provides aggregate information on the general public's Google search behavior. Research shows that the Google Trends data reliably capture the U.S. public's awareness of important social problems such as immigration and immigration enforcement (46). I define the search period from January 1, 2007 through December 31, 2018 to capture the long-term relative salience of deportation in the U.S. public's Google search behavior throughout the study period.…”

Indices of Non-Ignorable Selection Bias for Proportions Estimated from Non-Probability Samples

“…() and Nwosu et al . ()), or collected from other sources such as commercial databases, on‐line searches (Shlomo and Goldstein, ; DiGrazia, ) and on‐line surveys (e.g. Evans et al .…”

“…These 'infodemiology' data might be scraped from social media platforms such as Twitter (e.g. Myslín et al (2013), Nascimento et al (2014), Reavley and Pilkington (2014), McCormick et al (2017) and Nwosu et al (2015)), or collected from other sources such as commercial databases, on-line searches (Shlomo and Goldstein, 2015;DiGrazia, 2015) and on-line surveys (e.g. Evans et al (2007), Brooks-Pollock et al (2011) and Heiervang and Goodman (2011)).…”