Mobile Diary Method — Capturing Lived Experience via Smartphone
- Introduced by
- Csikszentmihalyi & Larson (ESM foundation); mobile adaptation through 2000s smartphone proliferation
- Year
- 1987 (ESM roots); mobile form ~2007–2010
- Type
- Longitudinal self-report data collection technique
- Family
- Data collection
What does it do?
The Mobile Diary Method is a longitudinal self-report technique in which participants record their thoughts, feelings, behaviors, or events using a smartphone app or mobile platform over a defined study period — ranging from days to months.
Rooted in the classic diary method and the Experience Sampling Method, its mobile form enables real-time, in-context capture of experience, dramatically reducing retrospective recall bias compared to one-shot surveys or end-of-day questionnaires.
The intuition behind the method
Think of the mobile diary as giving every participant a pocket research assistant. Instead of asking people to recall their week at a single sitting — where memory fades and filters — the app prompts them to log what is happening right now: their mood, the activity they are doing, who they are with, or what just occurred. Because entries happen in the moment and in the natural environment, the data reflect actual lived experience rather than reconstructed summaries. Over many such entries, researchers can trace patterns, trajectories, and day-to-day variability that no cross-sectional measure can reveal.
History
The Mobile Diary Method evolved directly from the paper diary tradition in psychology and from the Experience Sampling Method (ESM) developed by Csikszentmihalyi and Larson in the 1980s. ESM used pagers to signal participants at random intervals; participants then completed brief paper forms describing their current activity and mental state. The proliferation of smartphones from approximately 2007 onward enabled researchers to move signalling, form delivery, and data capture onto a single device, creating what is now variously called the mobile diary method, mobile ESM, or ecological momentary assessment (EMA). Bolger, Davis, and Rafaeli's 2003 Annual Review paper systematised diary methods as a research family and remains the most widely cited methodological overview.
When to use it
Use the Mobile Diary Method when you need to capture within-person variation over time and across natural contexts — for example, tracking daily mood fluctuations, stress appraisals, health behaviors, or social interactions as they unfold. It is especially valuable when retrospective recall is likely to be distorted (memory bias is a documented threat in single-session surveys) or when the phenomenon is inherently dynamic. The method requires participants who own or can borrow a compatible smartphone and who are willing to respond to multiple daily prompts across the study window; low compliance — typically defined as fewer than 60–70% of prompts completed — can seriously bias the sample of observations. Avoid it when the research question concerns stable traits or single-occasion behaviors that do not vary within persons, when participants lack reliable smartphone access, or when the prompt burden would be ethically inappropriate (e.g., acutely ill or highly vulnerable populations without careful safeguards).
Limitations and considerations
- Participant burden is substantial; repeated prompting across days or weeks leads to fatigue and declining compliance.
- Requires participants to have reliable smartphone access and basic digital literacy — excluding segments of some populations.
- Data are a multilevel nested structure requiring multilevel statistical models; standard regression is inappropriate.
- Privacy and data-security obligations are heightened because entries may contain sensitive, time-stamped, location-linked information.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Captures experience in real time in natural settings, minimising retrospective recall and telescoping bias.
- Enables analysis of within-person variability and change — questions no cross-sectional design can answer.
- Mobile delivery provides automatic time-stamping, compliance tracking, and large volumes of observations per participant.
- Flexible entry formats (scales, text, photos, audio) suit a wide range of research questions across health, social, and behavioral sciences.
- Ecological validity is high: data reflect everyday life rather than laboratory or one-off survey conditions.
Applications
- Tracking daily emotional dynamics, stress appraisals, and coping strategies in clinical and health psychology.
- Monitoring dietary intake, physical activity, and symptom fluctuations in chronic-disease management research.
- Studying social interaction quality, loneliness, and belonging in real time across adolescent or adult populations.
- Capturing work-related experiences — flow states, interruptions, or burnout precursors — as they occur during the workday.
- Investigating substance-use triggers, craving intensity, and situational context in addiction research.
- Examining learning episodes, motivation, and self-regulation in naturalistic educational settings.
How it works
How it works
The researcher specifies the diary protocol: the study duration, entry frequency (e.g., 3–8 prompted entries per day, or event-contingent entries), the question format (Likert scales, open text, photos, or combinations), and the mobile platform to be used. Timing can be signal-contingent (random or fixed prompts from the app) or event-contingent (participants initiate an entry when a target event occurs).
Participants are recruited and screened for willingness to engage with a mobile device across the study period. Informed consent covers the nature and frequency of prompts, data storage, and privacy. A training session — in person or via video — familiarises participants with the app, explains what constitutes a valid entry, and addresses technical issues before the study begins.
The diary app is installed and configured on participants' smartphones (iOS or Android). Researchers test notifications, confirm data sync to a secure server, and run a short pilot period (typically 1–2 days) to identify compliance problems or technical failures before the full data-collection window opens.
During the data-collection window, the app delivers prompts according to the pre-set schedule. Random-interval prompts are common in Experience Sampling designs to sample experience representatively throughout the day. Participants have a defined response window (e.g., 20–30 minutes) after each prompt; missed prompts are logged automatically, allowing compliance monitoring.
Participants complete each prompted entry directly on their device. Responses are time-stamped and geo-tagged if the protocol requires location data. Entries are transmitted to the server in real time or batched when connectivity is restored. Researchers monitor compliance dashboards and may send reminders or brief check-in messages to sustain engagement.
The resulting dataset is a multilevel, repeated-measures structure: many observations nested within individuals. Typical analytic approaches include multilevel modeling (hierarchical linear models) to separate within-person from between-person effects, intraclass correlation coefficients to quantify variability, and time-series or trajectory analyses to describe change over time. Qualitative content from open-text entries can be thematically coded alongside the quantitative measures.
Common pitfalls
- Ignoring compliance rates — analyzing data without accounting for missing entries (especially non-random missingness) produces biased within-person estimates; always report and examine compliance patterns.
- Over-prompting — too many daily prompts rapidly erode engagement; pilot studies should test the maximum tolerable frequency before the main study begins.
- Using standard regression on nested data — each participant contributes many observations, violating the independence assumption; multilevel models are required.
- Neglecting the training phase — participants who misunderstand what constitutes a valid entry produce inconsistent data; thorough onboarding and a short practice period are essential.
- Treating all missed entries as equivalent — missed prompts can reflect systematic patterns (participants skip prompts when stressed, for example) that introduce selection bias into experience samples.
Frequently asked questions
How many diary entries per participant are needed for reliable within-person estimates?
Simulation studies suggest a minimum of around 30–50 observations per person for stable multilevel estimates of within-person effects, though 20 can be workable for simple two-level models. The required number depends on the intraclass correlation, the expected effect size, and the complexity of the model. Pilot the protocol to estimate compliance rates, then power the study accordingly.
How is the Mobile Diary Method different from ecological momentary assessment (EMA)?
The terms overlap substantially. EMA, coined in the clinical literature, emphasises random-interval sampling to capture momentary states and is particularly associated with symptom monitoring. The Mobile Diary Method is a broader label that includes event-contingent and end-of-day protocols in addition to random-interval designs. In practice, researchers use the terms interchangeably; what matters is describing the specific protocol clearly.
What app or platform should I use?
Purpose-built platforms include m-Path, ExperienceSampler, ilumivu, and Metricwire, all of which handle scheduling, push notifications, and encrypted data transfer. Survey platforms such as REDCap can be configured for EMA with additional scripting. The choice depends on your institution's data-security requirements, your budget, and whether you need iOS and Android compatibility.
Can I combine open-ended qualitative questions with Likert-scale items in the same diary?
Yes. Mixed-format diaries are common; brief Likert items capture quantifiable states efficiently, while periodic open-text prompts (e.g., once per day) add interpretive depth without overwhelming participants. Keep the total time per entry short — under two minutes is strongly recommended to sustain compliance across a multi-week study.
How do I handle data from participants with very low compliance?
A common threshold is 60–70% of prompted entries completed. Participants falling below this level may be excluded from analysis or treated in sensitivity analyses, as their sparse observations are unlikely to represent a stable within-person sample. Report your compliance criterion and the number of participants excluded so readers can assess potential bias.
Related methods
Sources
Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the Experience-Sampling Method. Journal of Nervous and Mental Disease, 175(9), 526–536.
Open DOI referenceCopied.Bolger, N., Davis, A., & Rafaeli, E. (2003). Diary methods: Capturing life as it is lived. Annual Review of Psychology, 54(1), 579–616.
Open DOI referenceCopied.