Description
Collect continuous behavioural data from home sensors or phones and relate it to interviews, experience sampling and validated surveys that supply context and ground truth.
Sensors record what people do without effort on their part: location and dwell time, sleep, conversation, movement, nearby devices and app use. On their own such logs lack meaning, so the design pairs them with self-report. In one study, in-home location sensors were modelled into places of activity and residents were interviewed about them; an odd place spanning kitchen table and couch turned out to reflect a birthday party. In StudentLife, phones sensed 48 students over a ten-week term alongside experience-sampled stress and mood, validated mental health surveys before and after, and grades as outcomes; the linked data revealed declines in sleep and activity towards term end. Rigour comes from defining ground truth in advance, anonymising carefully and explaining anomalies with participants rather than guessing.
Attributes
- Discipline
- Field & Ethnographic
- Loss type
- Siloed
- Stage
- Discovery, Post-launch
- Data
- Mixed
- Effort
- Months
Steps
- Define the behaviours to sense and the outcomes or self-reports that will serve as ground truth.
- Deploy passive sensing (home sensors or a phone app) and agree data transfer, storage, anonymisation and consent arrangements before collection.
- Collect for a defined period, typically weeks to a full term, adding brief experience-sampling prompts for mood, stress or context.
- Administer validated surveys before and after, and gather outcome data.
- Model raw events into meaningful units such as places, routines or sessions.
- Interview participants about derived patterns and anomalies to validate and explain them.
- Mine the linked dataset for correlations and trends against events such as deadlines.
Sources
- Research Methods in HCI (Lazar et al., 2017)
- Interaction Design: Beyond Human-Computer Interaction (Preece et al., 2019)
- The Open Handbook of Experience Sampling Methodology (Myin-Germeys & Kuppens, 2022)
- Experience Sampling Method (Rhee et al., 2020)
- A-Z of Digital Research Methods (Dawson, 2020)
Also known as
- Passive mobile sensing
- Mobile sensing study
- Sensor-derived places
- Location-sensor places validated by interviews
- Combining passive smartphone sensing with self-report
Categories
- User Research
- Field & Ethnographic