Stated attitudes and observed behavior often diverge
Aliases: attitude behavior gap · self-report behavior discrepancy · intention behavior gap
What it is
An attitude–behavior discrepancy occurs when stated preference, intention, memory, or evaluation differs from action under concrete constraints. The sources are not true versus false. Self-report shows how people understand and account for experience; observation shows what occurred under current opportunity, cost, and context. Their disagreement can be the finding.
Why it happens
Attitudes are often summary judgments, while behavior combines goals, habit, defaults, skill, social pressure, and opportunity. Someone may value privacy yet accept tracking for work, or report never using a feature while forgetting rare use. Questions change the reference frame, while logs capture only defined events. Discrepancy can therefore reflect psychology, measurement error, or two sources addressing different objects.
Studying it
Align evidence to the same person, task, time window, and operational definition. Analyze sources independently, then use a joint display for convergence, complementarity, conflict, and missingness. Investigate conflicting cases through opportunity, defaults, recall interval, and event definitions rather than choosing a supposedly truer side. Record reactivity when observed behavior was produced by the study setting.
Where it stops holding
Attitudes sometimes predict later choice, and behavior may reveal stable preference; divergence is not inevitable. Logged clicks do not equal intent, and interview accounts are not lies. Future products lack natural behavior, requiring proxy choices or adjacent behavior with explicit distance. Direct observation of sensitive action cannot bypass privacy and consent in pursuit of authenticity.
Applying it
- Define one attitudinal and one behavioral indicator for high-stakes claims with matched object and time.
- Preserve conflict and code opportunity, definition, recall, or possible attitude change rather than averaging it.
- When intention is strong but action weak, inspect cost, defaults, and access before adding persuasion.
- Compare stated intent with post-launch adoption pathways and retain unexplained counterexamples.
Related
- Same group: Q1.10.1 Methods vary across three dimensions · Q1.10.3 One question may require all three dimensions · Q1.10.4 A qualitative–quantitative binary hides context
- Adjacent: Q2.01 User interviews · Q3.11 Telemetry and instrumentation
- Search terms:
attitude–behavior gap·self-report bias·behavioral trace