General Online Research 2019
F 4: Poster Session (Part IV)
Optimizing Response Rates in Web Surveys of Establishments: The Effects of Contact Mode
1Institute for Employment Research; 2University of Mannheim; 3University of Michigan
Identifying strategies that maximize participation rates in population-based web surveys is of critical interest to survey researchers. While much of this interest has focused on surveys of persons and households, there is a growing interest in surveys of establishments. However, there is a lack of experimental evidence on strategies for optimizing participation rates in web surveys of establishments. To address this research gap, we conducted a contact mode experiment in which establishments selected to participate in a web survey were randomized to receive the survey invitation with login details and subsequent reminder using a fully crossed sequence of paper and e-mail contacts. We find that a paper invitation followed by a paper reminder achieves the highest response rate and smallest aggregate nonresponse bias across all-possible paper/e-mail contact sequences, but a close runner-up was the e-mail invitation and paper reminder sequence which achieved a similarly high response rate and low aggregate nonresponse bias at about half the per-respondent cost. Following up undeliverable e-mail invitations with supplementary paper contacts yielded further reductions in nonresponse bias and costs. Finally, for establishments without an available e-mail address, we show that enclosing an e-mail address request form with a prenotification letter is not effective from a response rate, nonresponse bias, and cost perspective.
Brand Relationship Quality on YouTube: The emergence and impact of strong between recipient and creator relationships
Trimexa GmbH, Germany
YouTube, with 1.5 billion users in 2017 (Statista 2017), has become the second largest social media platform. Some YouTube creators have gathered large fan communities. To get a deeper insight into the fandom of YouTube viewers and their bond with the creators, the multidimensional construct of Brand Relationship Quality (BRQ) by Fournier (1994) was transferred to the recipient-creator relationship. Creators are understood as personal brands for this study. The study also reviews the impact of BRQ on the brand loyalty of viewers towards creators, in order to see if a high BRQ is able to keep a creator relevant in the long term.
Another goal of this study is to examine the extent of influence a creator might have concerning their BRQ. For this purpose, the influence of the creator related determinants of authenticity, expertise, self-disclosure and brand personality strength was measured. In order to examine the mentioned relationships, a corresponding model is developed.
An online survey was conducted and a total of 713 subjects took part in it. Respondents were viewers of one out of four different YouTube creators.
To measure the dimensions of the BRQ, existing scales of Leung (2016), Thorbjörnsens (2002) and Henkel/ Huber (2005) were adapted to the recipient-creator relationship.
To determine the relationships between the creator-related determinants, a multiple linear regression analysis was performed.
The impact of BRQ on brand loyalty was determined using a simple linear regression analysis.
It can be verified that the BRQ with a R of, 518 has a highly significant impact on brand loyalty.
The influence of the creator-related determinants is collectively model R-squared, 504 and is also highly significant. However, the influence of the determinants of authenticity and expertise could not be confirmed.
It can be confirmed that the BRQ has a positive influence on the loyalty of recipients. The BRQ thus helps to keep a creator relevant in the future.
In addition, a creator can have a positive impact on the BRQ by increasing its perceived self-disclosure and brand personality with their behavior.
SurveyMaps: A sensor-based supplement to GPS in mobile web surveys
1University of Göttingen, Germany; 2University of Mannheim, Germany; 3University of Frankfurt, Germany
Integrating web tracking and surveys to investigate selective exposure in news consumption
GESIS - Leibniz Institute for the Social Sciences, Germany
Relevance & Research Question: Computer-mediated communication has become deeply ingrained in political life. People use digital technologies to get political information and news or directly follow and interact with political actors. The measurement and analysis of these activities pose considerable challenges for researchers since they are distributed across multiple channels and platforms, intertwined, and ephemeral. With a case study relating a direct measurement of news consumption in Germany to political predispositions and attitudes of people we demonstrate the added benefit of integrating survey data and digital trace data.
The advent of right-wing populist actors and growing mistrust of political elites has coincided with negative public attitudes towards the news media. Given these political predispositions, selective exposure theory predicts that supporters of parties sending anti-media cues and people with populist attitudes would choose their news sources more selectively or even actively avoid entire types of news.
Methods & Data: We use a novel data set that links web browsing histories from 1,261 German internet users to their responses in an online survey. That way, we can objectively measure people's online behavior while surveying them for sociodemographic and political variables.
Results: Our findings in various regression models indicate that party cues are strongly related to news consumption while overshadowing the role of populist attitudes. Supporters of the German party with the most outspoken anti-media stances, the AfD, expose themselves less to news than supporters of other parties. This pattern remains robust for hard news, soft and tabloid news, and is most pronounced in the case of public broadcasting news.
Added Value: Our project makes substantive contributions to the fields of selective exposure and political communication. Methodologically we make a contribution to synthesizing the two paradigms survey research and computational social science as they have the potential to compensate for their respective weaknesses when combined in a systematic way.
Benchmarks for E-Health Evaluations
1University of Münster, Germany; 2Bielefeld University of Applied Sciences, Germany
Relevance & Research Question: Two in three Internet users regularly search for health information online. Content, usability, and aesthetics perceptions as well as trust towards a website provider are central dimensions of web users’ experiences. Each dimension is crucial for users’ acceptance, appreciation, revisit and recommendation of specific websites. For example, if users do not understand the content or distrust the website’s provider, they will seek another one. The aim of the present study is to provide aids in terms of benchmarks, helping to better understand use and adoption of health information presented online.
Methods & Data: We combined data of two studies, with a total of more than n=2.500 participants, evaluating m=33 health-related websites. The website pool was based on evaluations of seven experts (including two of the studies’ authors). Study 1 (n = 355, m = 3) used a within-subject design; study 2 (n > 2.200, m = 30) used a between-subject design. Each website was presented fully-functional and rated by 60 to 355 participants (Mean = 105) using established questionnaires: Web-CLIC (website content), PWU-G and UMUX-Lite (usability), VisAWI-S (aesthetics), and trusting belief scales of McKnight et al. (2002). Study 1 participants were recruited via the panel PsyWeb (https://psyweb.uni-muenster.de/), study 2 participants via a commercial panel. Websites were clustered in different categories such as “charity websites”, “educational and clinician websites”, “government websites” or “media websites”.
Results: Data collection was only recently finished, and thus we just started the data analyses. We aim to test for significant differences between categories on the different web user experience questionnaires. Resulting benchmarks in terms of mean scores for each category and measure will be presented at the GOR.
Added Value: All widely-used evaluation tools yield continuous scores, e.g. leading to a website usability score of 5.5. In themselves these scores are difficult to judge – benchmarks and cut points enable a meaningful interpretation of such individual scores. This is of particular importance for websites presenting health-related information. Thus, we hope that both, scientists evaluating existing e-health services as well as practitioners creating new ones, find these benchmarks useful.
Online recruiting methods from the perspective of job candidates
University of Münster, Germany
Relevance and research question:
Because of the shortage of skilled workers, demographic changes and the strong technological progress, personnel recruitment has become more important over the past years (Chapman et al., 2005; Weitzel et al., 2015). Furthermore, there is an increasing use of Social Media and mobile devices. Thus, online recruiting methods expanded the possibilities to recruit employees and are used even more by companies (Thielsch et al., 2012; Seng, Fiesel, & Krol, 2012, Weitzel et al., 2015). In the present study different online recruiting methods were analysed with regard to use and rating from an applicant’s point of view.
Methods & Data:
Nine online recruiting methods (e.g., corporate websites, online-communities, job portals, …) were examined in a web-based study. The sample (N=1121) was recruited and surveyed by respondi AG. For each method the participants had to indicate if they „know, but did not use“, „used before“, „use now“ or „do not know“ the following method. In addition, they rated these methods on a scale from 1 („does strongly not appeal to me“) to 5 („strongly appeals to me“).
Job advertisements on corporate websites and online job portals are used by most of the participants (55%). These methods have also the highest ratings (M=3.68, SD=1.073; M=3.50, SD=1.084). This is followed by Online-Communities like Xing or LinkedIn with a user rate of 25.5%. Among the examined Social Media platforms, Facebook has the highest user rate (18.3%) in comparison to Instagram (8.6%) and Twitter (9.0%). These platforms are not known by most of the participants (52.7%; 52.4%) and also have the lowest ratings (M=2.15, SD=1.122; M=2.10, SD=1.091).
The study shows that some online recruiting methods, such as online job portals or corporate websites, are often used by people searching for a job. Further, Facebook and online communities such as Xing or LinkedIn are used by many people – yet, some Social Media platforms (e.g., Twitter or Instagram) are unfamiliar to most of the respondents. Thus, many but not all online recruiting methods are used and accepted by most potential candidates.
The Relevance of IT-Competencies in a Digitalized Work Environment
Bielefeld University of Applied Sciences, Germany
Relevance & Research Question: Digitalization and new technologies have entirely changed job design and workforce tasks. Several studies reveal that the competency requirements have risen in general and especially within digital and media skills. Yet, our overall understanding of competencies and specific changes in certain sub-areas of digital skills, both today and in the future, remains unclear. Based on this gap in research, the following questions were proposed:
What is the general understanding of the term “competency”? How relevant is digital and media competency in departments outside of IT? Which technologies and drivers of digitalization are most relevant employee tasks today and in the future (i.e. 10 years)?
Methods & Data: A 12-question web survey of employed persons with various jobs [n=261] was conducted in Nov.-Dec. 2017. Participants were addressed by social media, personal approach and faculty email lists. They were asked to answer technology related questions on a 4-point rating scale [1=unimportant, 4=very important].
Results: Most respondents (23.4%) understood competencies as “The situation-related adapted acting through acquired qualifications.” Other understands followed: “One’s self-organised ability to act in an existing situation” (19.9%) and “Application and implementation of the acquired knowledge into a certain action” (19.2%). The “Skill to deal with difficult tasks” (6.1%), however, ranked among the lowest of responses.
Concerning the relevance of digital competencies in certain departments, respondents believed that marketing & sales (3.45 points on average) and research & development (3.40) are most important. Least critical are the legal department (2.50) and production (2.78).
Respondents considered e-learning (2.99 today/3.39 in future), cloud computing (2.93/3.36), new business models (2.91/3.22) and big data (2.81/3.33) to be the most important technologies and drivers of digitalization. Although 3D printing (2.11/2.67) and VR/AR (2.30/2.93) ranked last, they had the widest distance regarding the future and thus untapped potential.
Added Value: This study reveals which specific media literacy related skills will be relevant in the future. Additionally, the study not only contributes to our understanding of the perception of the term competencies, but also shows which company departments are most affected by digital and media skills.
A selection bias of Facebook respondents, which need to be taken into account
1University of Applied Sciences, Upper Austria; 2Johannes Kepler University, Linz
Relevance & Research Question: Social media has become an integral part of our daily life. Thus, it comes as no surprise that it is playing an increasingly important role in survey research. Despite methodological as well as factual criticism, surveys conducted via social media are growing more popular, especially when it comes to "Facebook". Companies with an Facebook account are asking their „Facebook-user“ to complete surveys.
Since social media usage is part of media usage and thus tied to the perceived uses and gratifications, influencing motivations on why individuals participate and how this affects survey data should not be ignored. To the best of our knowledge, no study exists that focusses on the influence of user-specific characteristics on the Facebook-survey participation and outcome, in order to interpret results in the correct way. Therefore, the following question arises: How can one account for an assumable “Facebook-survey-selection-bias”, triggered by the participants’ characteristics?
Methods & Data: German and Austrian public utilities collaborated to collect data via Facebook. Eight German and six Austrian utilities helped in the empirical study by posting an invitation for the inquiry on their Facebook-accounts. They shared the link to the online survey in 2017. 258 Facebook-users participated in the study. The participants were grouped based on configural frequency analyses. Indicators used for this procedure were involvement (importance of user-comments and of encouragement to ask questions) as well as confidence about the surveys’ impact (probability of change). Finally, linear regressions were performed in order to examine the impact of the group assignment.
Results: Three different groups of Facebook-respondents were computed. Group A consisted of respondents who want active involvement in current issues on the company’s Facebook page and have positive expectations about the impact of the inquiry, B is the opposite and C is in between. The group-allocation is responsible for different outcomes – the more active and positive a respondent seems to be, the more positive their answer. Moreover, the conducted “substantial” regressions gain a higher power to explain, when the grouping variable is considered.
Added Value: The paper highlights the importance of the participants’ characteristics for interpreting Facebook-survey-results.