Interview with Sandra Matz: "Big Data, Privacy, and Psychological Targeting"

Dr. Sandra Matz is an Assistant Professor of Management at Columbia Business School.

Dr. Matz’s research uses Big Data methods to explore the relationship between psychological variables and people’s digital footprints.

Everything that we know about [ourselves] as people is expressed in behavior to some extent, so it’s very unlikely that there is something we really absolutely cannot predict.”

• Who are you, where do you work, and what is the primary focus of your research?

• What kinds of data are you making use of in your work?

• How do you source this data?

• Are people aware of how their data is being used?

• Is it possible to retain privacy while using popular digital services?

• Could you describe your work on personality-based psychological targeting?

• What are the effects of buying things in line with one’s personality traits?

• Does psychological targeting constitute manipulation?

• Is it a problem that “revealed preferences” sometimes clash with people’s long-term interests?

• Are there limitations in what can be predicted from Big Data?

• Is there anything dehumanizing about a data-driven view of the human experience?

• In the context of, for example, hiring decisions, will there come a time when human judgment becomes obsolete?

• Do you need individualized data for targeting?

• If you don’t, does that threaten the prospect of effective regulation?

• Are there regulatory recommendations you would make?

• Do companies need to get good data (and therefore engender privacy concerns) to build competitive services?

• Would a shift from advertising to subscription-based services help?

• Beyond personality, what other types of inferences can we make from digital footprints?

• Are these techniques methods different in kind or only in degree from earlier advertising methods?

• How should we think of autonomy in light of how manipulable we are?

• Would people actually prefer more autonomy?

• Are our online environments less like echo chambers than our physical ones?

• Will Big Data methods separate prediction from explanation?

• Will we reach a limit to what predictions we’re able to explain?

• Do you have a way of distinguishing ethical from unethical persuasion?

• How should individuals make responsible decisions about privacy?