Passive Mobile Data Collection |
|
| Chair | Professor Florian Keusch (University of Mannheim ) |
| Coordinator 1 | Professor Frauke Kreuter (University of Maryland, University of Mannheim, IAB) |
Research focused on analyzing impacts of media exposure to the audience faces many methodological challenges when applying standard methods of interviewing respondents. Respondents usually do not remember clearly what media contents (e.g. exact commercials, reportages about refugees, ...) they have seen. Moreover recalling the content is dependent on their attitudes to the topic and perceived importance of the topic. These and other challenges of standard research methods make it hard to analyze impact of media exposure on audience and verify 1st level or 2nd level agenda settting hypotheses. In our presentation we want to show how some of these obstacles can be overcome by using electronic passive media measurement in combination with standard (declarative) interviewing of the panelists whose media exposure was measured.
Using adMeter audio-matching based research technology developed by MEDIAN s.r.o. we identified how many TV reportages on immigration / refugee topic adMeter each respondent/panelist saw during one month of the refugee crisis (august 2015). Then we interviewed the panelists/respondents to gather information: a) perceived importance of social and political topics and b) attitudes and accepted framing of immigration topic (2nd level of agenda setting).
Using regression models controlling for sociodemographic and general TV consumption patterns we show there is a connection between exposure to the refugee/immigration topic in Czech TV and perceived importance of the topic on individual level (2nd level of agenda setting). We then further analyze relation between media exposure and attitudes towards immigration and refugees (2nd level of agenda setting). In discussion we focus on question whether controlling for sociodemographic and general TV consumption helps to eliminate competing causal explanation of content–attitudes relation (impact of audience to the media, selection of news sources).
adMeter is cross-media passive electronic media measurement tool running on smart-phones using audiomatching for measuring TV and Radio. It also gather information on web usage and geolocation. adMeter was developed by Czech research agency MEDIAN s.r.o. and is used by leading Czech media and advertising planners and sponsors for cross-media ad planning and analyzing effectiveness of commercials (i.e. impact on brand awareness, consumption preferences, etc.). adMeter runs on N=1000 respondents representative panel and it´s results are regularly presented on top media and marketing conferences (see presentation from ASI Budapest 2016 http://www.median.eu/cs/wp-content/uploads/2016/12/ASI_2016_ADMETER-1000_YT_v08.pdf)
Details of adMeter technology, its benefits, challenges it faces, its usage in commercial media research and its methodological implications will be described during the presentation.
Daniel Prokop
Head of Social and Political Research, MEDIAN, s.r.o.
(8 years of experience in commercial research)
PhD. student and lecturer on Faculty of Social Sciences, Charles University in Prague
(4th year of PhD. Programme)
Lea Michalová:
Social and Political Research Analyst, MEDIAN, s.r.o.
published master thesis focused on using adMeter data to analyse agenda setting (Faculty of Social Sciences, Charles University in Prague, 2016)
The paper analyzes mobility data collected using a smartphone-based GPS-Tracking and a web-based trip diary. In the survey respondents had the choice between the two modes of participation and the paper examines the question whether there is a mode effect in the collected data or a mode preference of participants with a specific mobility pattern or with specific characteristics. It also focuses on the challenges and limitations of the passive mobile data collection and the traditional trip diary.
For some years now mobility patterns seem to change. New mobility services like carsharing are implemented worldwide. At the same time, people also seem to change their mobility behavior. While there is much debate on these trends, there is only limited empirical data on their extent and stability. Currently, data on individual travel behavior is still mainly conducted by using traditional survey techniques (like PAPI, CATI). However, high survey costs and high respondent burden are associated with these survey methods. GPS data collection ideally provides more detailed information on route choice and time use than traditional travel survey methods while at the same time decreasing respondent burden and costs. The multimo project was set up to develop and test two approaches to collect reliable and detailed mobility data to explore new trends.
The multimo project was conducted in German metropolitan areas. The sample consisted of self-selected participants approached via Facebook and posters in public transportation. For registration they had to complete a general questionnaire with some personal information as well as information about their everyday mobility. At the end of the questionnaire two modes where offered to collect the travel data within a 14 day period:
-The modalyzer app, which is a fully automated smartphone-based research tool that passively collects location and acceleration data. A post processing algorithm identifies trips and transport modes and calculates distances travelled and trip durations.
-The web-based trip diary, which participants could use to report their trips on a daily basis during the 14 day period.
Participants who reported at least ten days were used for the evaluation sample. The dataset consists of data from 677 participants using the trip diary and 475 participants using the app.
The observed mobility patterns in both groups are similar with regard to used modes of transportation, mean speed and mean distance covered. But there are a lower number of daily trips and a lower daily travel distance among the GPS-Tracking group indicating that the Tracking is not collecting data during all travel time. The analysis shows also a higher involvement of participants using the trip diary: 56%of these participants reported ten days or more, whereas only 47% of the modalyzer group tracked their trips in the same time period. The analyses of routes are the most vital advantage of the recorded GPS-Tracks. Although the data is not sufficient to estimate key mobility parameters it can be used to supplement traditional survey data especially to learn more about transport routes.
Smartphones are becoming increasingly important and widely-used for survey completion. Smartphones offer many new possibilities for survey research: We can, for example, send pop-up questions in real-time, for instance to measure participants' feelings, and record sensor data. However, as nice as these new opportunities are, the questions we ask can get increasingly intrusive, and we risk over-asking participants.
Current study studies the nonresponse bias in each of the stages of asking intrusive questions and recording sensor data through mobile phones. More specifically, the four stages of this innovative Time Use Survey were as following: 1) accept invitation to participate in the study, 2) participate in the Time Use Survey, 3) answer pop-up questions and 4) give permission to record sensor data (GPS locations and call data). We used participants from the LISS-panel, a representative panel of the entire Dutch population.
Fundamental, methodological knowledge about nonresponse in smartphone-only studies and passive mobile data collection is lacking, but very important to understand selection bias. Therefore, our main research question is: How can we predict nonresponse at different stages in a smartphone-only survey? In the presentation, we would mainly focus on the predictors of not giving permission to record sensor data, and whether these are comparable to the predictors of the other stages.
This study provides us with knowledge about bias in passive data mobile collection, a field which remains relatively unexplored. This knowledge about who is and who is not willing to share sensor data can be very valuable for all survey researchers who consider such a study.
In online surveys lots of paradata can be captured as a byproduct of data collection, for example time stamps, information about devices and IP addresses. Even though much of this information is automatically send by the browser, its storage and use by researchers is not always compatible with data protection guidelines and informed consent by respondents is required.
In this paper we study consent to the request to automatically collect geographical information in a large German online panel. Respondents of wave 4 of the German Internet Panel, a probability-based online panel of the general population, were asked for consent to automatically track their location using JavaScript. If consent was provided, the IP-address that is transmitted by the browser was stored and longitudes and latitudes were derived from it. In addition, the same respondents were asked to report their location (city name and postal code).
Geographical information on the location where the respondents fill in the survey is valuable for both substantial and methodological research. Spatial identifiers can be used to link outside information to the survey to enrich the data set with additional explanatory variables, for example weather or climate data or distances to public place like supermarkets, green spaces, or schools.
Automatically collected geographical information can be assumed to be of higher quality than reported locations, especially for respondents who fill in the survey in unfamiliar places or on the road. While response burden is lower for the automated collection, tracking of IP-addresses can also be perceived as intrusive and raise data protection concerns with respondents.
We address the following research questions:
1. What is the acceptance among the general population in Germany towards digitally collecting information on their geographical location?
2. Are there differences between people who consent to the digital collection of their geographical location and people who fill in information about their location manually in terms of socio-demographic and personality characteristics?
We find the acceptance for the digital collection to be about 60 % while about 95 % of the respondents report a city name or postal code. Both reporting a location and consent to digital collection are influenced by personal characteristics and different characteristics determine the willingness to provide the two types of information.