New Research Puts Your Online
Privacy Preferences to the Test
When it comes to privacy, you’ve probably heard this before, “You don’t care about privacy if you have nothing to hide.” When someone says this, they are not taking into account that you may have strong beliefs on the very premise of whether someone should have access to your private information. This belief from you is called the “intrinsic” point of view.
Instead, the speaker is operating from the “instrumental” point of view, which is not centered on personal philosophy or position, but rather practical considerations based on possible rewards or consequences for sharing specific information. When you take the instrumental approach, you may decide your Internet activity is more of an open book to websites because you perceive less downside to sharing. On the other hand, if giving an insurance company access to your real-time driving data could lead to higher insurance rates, you may change your position on Internet privacy in that context. Instrumental decision-making is more likely to change based on the situation.
These issues are at the center of a new study that points to the need for marketers and websites to separate consumer privacy preferences into two components when those consumers are given data-sharing choices and options. These components are intrinsic and instrumental.
The study, published in the current edition of the INFORMS journal Marketing Science is called “Valuing Intrinsic and Instrumental Preferences for Privacy.” It is authored by a researcher from Boston University.
“Increasingly, consent becomes a prerequisite for personal data processing,” said the study author. “Consumer preferences determine what information can be collected and shared. For many consumers, privacy preferences are more situational because the outcomes of sharing data are situational. My research sought to determine why and how this should inform future efforts to collect and share consumer data.”
To conduct the research, Lin conducted a market research survey through the University of Chicago that posed questions to participants to measure their valuation for smartwatch attributes, and their intent to buy a digital device. Participants were asked to share certain identifiable information, from gender and age to relationship status and whether they have children.
With each personal question, participants had the option to choose “prefer not to say.”
Later in the process, participants were given the option to share their completed survey responses with a third party, a smartwatch manufacturer, that wanted to use the data to improve its product design. Participants could choose whether to share each personal variable separately. A gift card was used to compensate people for sharing more information.
To measure instrumental decision-making, the experiment changes the economic payoff to participants at random when the shared data reveals private information about them. In this treatment group, this payoff is higher when the shared data reveals that they are high-income consumers interested in digital products. In the control group, the payoff does not depend on the private information revealed from data. In this control group, instrumental concerns are absent, and participants’ sharing decisions are only driven by intrinsic motives.
“This research did not seek to identify larger consumer preferences in themselves, but rather to determine if indeed we should be taking into account both the intrinsic and instrumental forms of decision-making when it comes to online privacy,” the author added. “Previously, marketers and websites did not divide into these two parts the possible motivations for data sharing. Rather, they made assumptions on the motives on why some people seem more guarded of their privacy, while others are more open. This led to possible bias on the part of marketers in their own analysis of consumer behavior.
“This research clearly illustrates that these two components do allow us to capture the distinct impacts of both,” the author said. “It reduces the likelihood of bias on the part of marketers, and it opens up an entirely new avenue for future consumer research and analysis.”
