Skip to content
Back to search results

Longitudinal Diary Method — Repeated Diary Data Collection Over Time

Introduced by
Allport (1942); systematic longitudinal extension developed by Bolger, Davis & Rafaeli (2003)
Year
1942 (diary method); longitudinal variant formalised 1980s–2000s
Type
Longitudinal qualitative/quantitative data collection
Family
Data collection
Font size

What does it do?

The Longitudinal Diary Method is a data collection technique in which participants record experiences, thoughts, feelings, or behaviors in structured diary entries repeatedly over an extended period — from days to months or even years.

Unlike a one-shot survey, it tracks within-person change, daily fluctuation, and temporal processes in natural settings, making it especially powerful for studying how phenomena evolve over time.

The intuition behind the method

Imagine you want to understand how academic stress develops and fluctuates across an entire semester, not just at one point in time. Instead of asking students to recall the whole semester at the end, you give them a structured diary and ask them to record their stress, coping strategies, and mood each evening. After 14 weeks you have a dense, real-time record of how each person changed — and you can see when stress peaked, how quickly people recovered, and what daily events triggered the sharpest changes. This longitudinal window is impossible to obtain from a single survey.

History

Diary methods in social science trace to Gordon Allport's 1942 use of personal documents, but systematic longitudinal diary research became methodologically rigorous in the 1980s with the development of experience-sampling methods by Csikszentmihalyi and Larson. The landmark review by Bolger, Davis, and Rafaeli (2003) in the Annual Review of Psychology established the diary method as a mainstream tool for studying within-person processes and outlined best practices for design and analysis. Advances in mobile technology since the 2010s have made event-contingent and signal-contingent longitudinal diaries far easier to deploy at scale.

When to use it

Use the Longitudinal Diary Method when the research question concerns change, fluctuation, or development over time within individuals — for example, how stress, mood, symptoms, or behavior patterns shift across days, weeks, or months. It is ideal when retrospective recall would introduce bias, when moment-to-moment or day-to-day processes matter, or when ecological validity in natural settings is a priority. Do not use it when a single cross-sectional measurement suffices, when participants cannot sustain the diary burden over the observation window (elderly, very ill, or cognitively impaired populations require adapted formats), or when the budget cannot support the infrastructure for monitoring compliance and managing attrition across a long fieldwork period.

Limitations and considerations

  • Participant burden is high; compliance and attrition are persistent challenges over long observation windows.
  • Multilevel analysis requires relatively large samples (typically 30+ participants, 15+ measurement occasions) to achieve adequate statistical power.
  • Reactivity effects: the act of keeping a diary may itself alter the behavior or thoughts being recorded.
  • Data management is complex — large, unbalanced panel datasets with missing entries require careful preprocessing.
  • Qualitative longitudinal entries are time-consuming to analyze systematically.

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Captures within-person change and temporal dynamics that cross-sectional designs cannot detect.
  • Reduces retrospective recall bias by recording experiences close in time to when they occur.
  • High ecological validity — data are collected in participants' natural daily environments.
  • Enables both idiographic (person-level) and nomothetic (group-level) analysis within a single dataset.
  • Flexible format — can combine quantitative ratings with qualitative narrative entries.
  • Well-suited to studying fluctuating constructs such as mood, pain, stress, coping, and behavior.

Applications

  • Tracking daily stress, coping strategies, and well-being trajectories in occupational health research.
  • Monitoring symptom fluctuation and medication adherence in clinical psychology and chronic illness studies.
  • Studying how academic motivation and engagement develop across an academic semester.
  • Investigating daily social interactions, relationship quality, and conflict in couples or families over weeks.
  • Capturing behavioral change during interventions (e.g., diet, exercise programs) in real time.
  • Examining how creativity, inspiration, or flow states occur naturally in everyday working life.

How it works

How it works

1

The researcher specifies the diary schedule (e.g., daily, twice-weekly, event-contingent), the observation window length (days to months), and the entry format — open-ended text, Likert-scale ratings, checklists, or a combination. Diary prompts are crafted to capture the target constructs at a level of granularity appropriate to the research question, and the overall burden per entry is kept manageable to sustain compliance over the long observation window.

2

Participants are selected who have directly experienced or are currently living through the phenomenon under study. Because the longitudinal diary method depends on sustained engagement, the sample is typically smaller than a cross-sectional survey — 20–200 participants depending on the analytic goals — and inclusion criteria often specify a minimum commitment period. Participants should be able and willing to complete diary entries at the required frequency for the full study duration.

3

Before the study begins, participants receive detailed training on how to complete entries, what counts as an event worth recording, and how to use the diary platform (paper booklet, app, or secure online form). Clear examples and practice entries help standardize recording behavior and reduce idiosyncratic interpretation of prompts. Written instructions and a researcher contact for queries are provided.

4

Participants submit diary entries according to the agreed schedule throughout the observation window. Signal-contingent designs send reminders at fixed or random times; event-contingent designs ask participants to record immediately after a specified event occurs; interval-contingent designs schedule entries at regular intervals regardless of events. Reminders via SMS or app notifications are standard in modern implementations.

5

The research team monitors response rates and data quality throughout the fieldwork period. Missing entries are followed up promptly; compliance dashboards flag participants at risk of dropout. Interim quality checks catch systematic errors (e.g., invariant responses, entries completed out of schedule) early enough to correct them without compromising the dataset. Attrition is documented for later reporting.

6

Longitudinal diary data are typically analyzed with multilevel modeling (hierarchical linear modeling) to separate within-person variation across time from between-person differences. Lagged analyses test whether events on one day predict outcomes the next day. For qualitative diary entries, thematic or content analysis is applied across the temporal sequence to identify how themes emerge, shift, or resolve over the observation window.

Common pitfalls

  • Entries too long or too frequent — if each diary entry takes more than 10–15 minutes or is required multiple times a day over months, dropout rates rise sharply, producing biased samples of highly motivated completers.
  • Neglecting attrition analysis — participants who drop out are rarely missing at random; failing to report who dropped out and why can substantially distort findings.
  • Ignoring the multilevel structure — analyzing repeated diary entries as if they were independent observations inflates Type I error; multilevel models are required.
  • Standardization drift — without clear instructions and mid-study checks, participants gradually reinterpret prompts differently, introducing measurement error across waves.
  • Insufficient observation window — a window too short to capture the phenomenon's natural cycle (e.g., a two-day diary to study weekly work rhythms) yields data that cannot answer the research question.

Frequently asked questions

How long should the observation window be?

The window should span at least one full cycle of the phenomenon you are studying. For daily mood fluctuations two to four weeks is often sufficient; for developmental changes across an academic year or a clinical treatment period, months may be required. Pilot studies are valuable for calibrating the minimum window needed before committing participants to a long study.

How many diary entries do I need per person for multilevel analysis?

As a rough guide, multilevel models for within-person effects typically require at least 10–15 measurement occasions per person, though more is better for detecting smaller effects. Simulations suggest that increasing occasions per person often improves power more than increasing sample size, up to a point. Consult a power analysis tool for multilevel designs (e.g., the Optimal Design software) at the planning stage.

Can I use paper diaries or must I go digital?

Paper diaries are still used, especially with older populations or in settings with limited technology access, but digital platforms (secure web forms, dedicated apps, SMS-triggered surveys) make timestamping, compliance monitoring, and data management far easier. Digital platforms also reduce transcription error and enable timely reminder notifications that improve compliance.

How do I handle missing diary entries in the analysis?

Report the percentage of missing entries per participant and overall. Examine whether missing data are related to the constructs under study (missing not at random). Modern multilevel software (e.g., R packages lme4 or brms, Mplus) can handle unbalanced data directly; maximum likelihood estimation uses all available observations without listwise deletion. Multiple imputation is an option for data missing at random.

Is the longitudinal diary method qualitative or quantitative?

It can be either or both. Quantitative diary studies ask participants to rate experiences on scales, enabling multilevel statistical analysis. Qualitative diary studies invite open-ended narrative entries analyzed thematically or through content analysis. Mixed-format diaries combining closed ratings and open narrative questions are common and allow triangulation within the same data collection episode.

Related methods

Sources

  1. 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 reference
    Copied.
  2. Scollon, C. N., Kim-Prieto, C., & Diener, E. (2009). Experience sampling: Promises and pitfalls, strengths and weaknesses. Journal of Happiness Studies, 4(1), 5–34.

    Open DOI reference
    Copied.