All time references are in CEST
Applications, Potentials, and Challenges when Using Google Trends in Combination or as Substitute for Surveys 1 |
|
| Session Organisers | Professor Florian Keusch (University of Mannheim) Ms Johanna Mehltretter (University of Mannheim) Dr Christoph Sajons (University of Mannheim) |
| Time | Wednesday 19 July, 09:00 - 10:30 |
| Room | U6-07 |
Aggregated Internet search data from Google Trends are increasingly used as a supplement or alternative to survey data. Proponents of Google Trends argue that anonymous search queries of Internet users are a good reflection of true interest, behaviors, and attitudes, particularly for sensitive topics, where surveys suffer from measurement error due to social desirability. In addition, Google Trends allows researchers to study changes in topic salience, attitudes, and behaviors across time and geographic areas at much finer granularity than possible in surveys. On the downside, using Google Trends data may include multiple problems. First, not everybody uses the Google search function, potentially leading to selection bias. Second, Google Trends only provide search volumes based on a sample of all search queries, thus questions of reliability arise. And third, it is often unclear how validly the selected search terms measure the constructs of interest.
In this session, we aim to bring together empirical evidence on the state-of-the-art use of Google Trends data in combination with or as an alternative to self-reports from surveys. Submissions can be methodological in orientation or can be substantive applications that demonstrate the usefulness and assess the quality of Google Trends data. Potential topics for submissions include, but are not limited to:
- Validation of Google Trends data
- Comparison of different approaches to select appropriate keywords
- Approaches to overcome reliability issues of Google Trends data
- Triangulation through joint use of Google Trends with surveys
- Analysis strategies for Google Trends data
- Best practices for transparent documentation when working with Google Trends
- Social science applications of the use of Google Trends data to measure specific attitudes, behavior, and topic salience
Dr Franziska Pradel (Technical University of Munich) - Presenting Author
Search engines are commonly used for online political information seeking globally, yet it remains unclear how political search predictions that reflect the latent interest of Internet users vary across countries and over time. We provide the first systematic analysis of search predictions for European and national politicians in online search. Search predictions are the first information users see when searching for information in the Google search bar, automatically completed information by the Google algorithm based on previous searches on the same topic. This allows us to estimate the latent interest of online users in European politicians by analyzing collected Google search predictions in ten countries for European party leaders, Spitzenkandidaten in the 2019 EP election, and cabinet members. We find that Google search predictions are less stable over time in the politicians’ country of origin, when the politicians hold a supranational role and being female, indicating more diverse interest for these politicians across countries. Moreover, search predictions are more similar across countries for political leaders and male politicians. We conclude with a discussion of possible future directions for studying the information-seeking about European politicians in online search.
Professor Malina Voicu (Romanian Academy) - Presenting Author
Dr Simona Mihaiu (Romanian Academy)
Gender based violence is a highly sensitive topic, subject to social desirability and societal normativity, being difficult to assess using regular instruments (such as survey research or official statistics). The challenges come from different sources. On one hand, the victims are reluctant to report it due to social desirability, like shame or fear, leading to a bias based on self-reporting. On the other hand, social norms related to domestic violence change over time too, due to formal and informal criminalization of violence against the partners, children, or elderly. While under-reporting makes it difficult to assess based on survey data, changes in social norms impede over time comparisons.
This paper assesses to what extent Google Trends can be employed to measure the dynamic of gender-based violence with special reference to Romania, a country that scores very low on self-reported exposure according to FRA survey 2015, but has high numbers of the cases convicted by law. One challenge faced when using Google Trends to tap gender-based violence in Romania deals with the limited digital literacy among the victims, as well as with its dynamic. We propose a way to account for the change in digital literacy based on survey data, coming from EuroBarometer in 2010 (EB 73.2) and 2016 (EB85.3) and other two national surveys Gender Roles Barometer (2018) and Gender Based Violence Survey (2022). In the first step we employ Changing Parameter Model approach, to assess the over-time change of the association between reporting gender-based violence and digital skills. In the next step we weight the data coming from Google Trends based on the results of Changing Parameter Models and we assess the validity of this measure against the official statistics.
Mr Sihle Khanyile (University of Michigan (ISR)) - Presenting Author
The potential of internet search data as an indicator for different economic activities has been demonstrated at least in the advanced economies. What makes internet search data so attractive is the timeliness advantage it has over traditional government surveys that are released with a time lag. Emerging economies have been slow in taking advantage of this stored internet data. A case in point is South Africa which has not exploited this data to service policy makers with timely data. In this paper I explore how a range of internet search queries have behaved in comparison to the official unemployment figures as published by through the Quarterly Labour Force Survey(QLFS). The econometric relationship between the official unemployment data and google search term data is modeled and compared. It is found that the labour market related search terms are correlated to actual employment trends as measured by the QLFS.