Ok Google: Using virtual assistants for data collection in psychological and behavioral research

Area of Science:

• Psychological research methods
• Human-computer interaction
• Behavioral science

Background:

• Voice-controlled virtual assistants (e.g., Amazon Alexa, Google Assistant) are increasingly popular.
• Traditional research methods face challenges in participant recruitment and accessibility.

Purpose of the Study:

• To evaluate the reliability and validity of virtual assistants as a data collection medium for psychological research.
• To assess the feasibility of using virtual assistants for replicating established social psychological experiments.

Main Methods:

• Developed Survey Mate, a Google Assistant extension, for data collection.
• Conducted two studies: one comparing virtual assistant and online questionnaire data (procrastination, shyness, social desirability), and another replicating five social psychological experiments.
• Collected data from smartphone and smart speaker users.

Main Results:

• Virtual assistant data showed comparable internal consistency, construct validity, and criterion validity to online questionnaires.
• Replicated social psychological experiments yielded comparable effects across virtual assistant users, regardless of device.
• Identified potential for increased participant recruitment, including hard-to-reach populations (e.g., visually impaired, lower education) and international samples.

Conclusions:

• Virtual assistants represent a viable and reliable new medium for psychological and behavioral research.
• This technology offers advantages in participant recruitment, accessibility, and potential for enhanced theory generalizability.
• Limitations include data privacy concerns and potential for socially desirable responses, necessitating careful consideration.