Do (not!) track me: Relationship between willingness to participate and representativeness of online information behavior tracking research

Gil-Lopez, Teresa; Makhortykh, Mykola; Urman, Aleksandra (7 September 2021). Do (not!) track me: Relationship between willingness to participate and representativeness of online information behavior tracking research (Unpublished). In: 8th European Communication Conference (ECREA 2021) - "Communication and trust". Online. 06.09.-09.09.2021.

Full text not available from this repository. (Request a copy)

Traditionally, political communication researchers have relied on self-reported participant information, but a recent trend is to combine it with automated tracing of online information behavior. Compared with surveys (self-reported data), tracking can provide more reliable insights about (online) information behavior that are not affected by social desirability or participants' inability to recall past behavior. However, the use of tracking often raises concerns about privacy or security which can influence participants' willingness to share data. Considering that such willingness determines the composition of research samples, it can affect the representativeness of tracking studies and add biases related to the political attitudes and proclivities which are at the core of the research question. This is particularly worrisome in the case of studies employing online access panels, for which the lack of generalizability is already a significant issue.

This paper explores how willingness towards participating in behavior-tracking research affects participation rates and sample characteristics/representativeness. We use insights from a pilot study on the relationship between political attitudes and citizen’s political information diets. Using an online access panel (N = 1714), a pre-screening for participation was done based on willingness to share tracking data until 200 interviews were successfully completed. With the purpose of ensuring representativeness, sampling quotas were used for gender and age across 2 incentive experimental conditions (10/15€). Demographic characteristics were then compared between participants willing to provide tracking data and those a) refusing participation (N=485); and b) initially willing to share but dropping out at some point in the study (N=400). The two monetary incentive conditions were also compared to see if increments in incentives enhanced willingness to participate.

Our findings align with previous studies reporting low willingness to participate in online behavior-tracking research. Of the 1714 participants, 629 declined immediately after the purpose of the study was communicated; they were significantly older (+5 years on average). In terms of sample characteristics, no significant demographic differences existed between those who agreed and those who disagreed to being tracked, but the latter were overall less prone to provide information (1,1% vs 0,47% missing rate for income). While no significant differences in gender, education, and income were observed between the 200 participants in our final sample of successful interviews and the remaining respondents, the former were younger (M = 48.17, SD = 15.38 versus M = 52.82, SD = 15.01) and had a slightly higher presence of the left-center and the conservative-right extreme political attitudes. Finally, no statistically significant differences were found between incentive conditions, which suggests that greater participation is unlikely to be achieved through (moderate) increases in monetary incentives. Likely, other aspects are weighted more heavily on decisions about sharing one’s own private online usage. Overall, our findings support existing concerns about both high refusal and item non-response rates, which may obstruct reliable comparison between demographic groups and entail other fundamental disparities. Despite this, our results also indicate that quotas, by ensuring a certain demographic structure, can be useful in broadening the composition of samples for tracking studies.

Item Type:	Conference or Workshop Item (Paper)
Division/Institute:	03 Faculty of Business, Economics and Social Sciences > Social Sciences > Institute of Communication and Media Studies (ICMB)
UniBE Contributor:	Makhortykh, Mykola, Urman, Aleksandra
Subjects:	300 Social sciences, sociology & anthropology
Language:	English
Submitter:	Mykola Makhortykh
Date Deposited:	04 Oct 2021 14:14
Last Modified:	05 Dec 2022 15:53
Uncontrolled Keywords:	tracking, representativeness, sampling, information behavior, methodology
URI:	https://boris.unibe.ch/id/eprint/159631