03 · What You Need to Know
Objective and Subjective Measures Answer Different Questions
What Researchers Usually Mean by an Objective Measure
An objective measure generally relies on observable events, performance, records, physical quantities, or standardized procedures rather than asking a participant to judge or report the target phenomenon directly.
Examples might include task completion time, examination performance, attendance records, physiological measurements, device-recorded activity, or counts of predefined behaviors.
Calling these measures “objective” does not mean that human judgment disappears. Researchers still decide what to measure, how to define events, which instrument to use, how to calibrate it, what data to exclude, and how to interpret the result. Some observational measures also require raters to classify behavior, introducing another layer of judgment.
What Researchers Usually Mean by a Subjective Measure
A subjective measure intentionally obtains information based on a person's perception, judgment, evaluation, recollection, feeling, or reported experience.
If you ask participants how satisfied they are with a course, how much pain they experience, how confident they feel, how difficult they perceived a task to be, or whether they feel socially connected, the participant's perspective is not necessarily unwanted contamination. It may be the phenomenon you actually want to measure.
Objective measure
Typically represents behavior, performance, events, records, physical quantities, or other observations without relying primarily on the participant's personal evaluation of the target phenomenon.
Subjective measure
Intentionally represents a person's perception, experience, judgment, feeling, evaluation, or self-reported state.
The terminology varies across fields, and the boundary can become blurry. A teacher rating a student's engagement is based on observation but also requires human judgment. A standardized performance task produces a numerical score, yet task design and scoring still embody methodological decisions. Objective and subjective are therefore useful distinctions, but not perfectly pure categories.
Ask What the Construct Actually Is
The measurement choice should begin with the construct rather than with a preference for a particular data source.
If your construct is perceived usefulness, participants' perceptions are part of the construct itself. Replacing those perceptions with system usage data would not make the measurement objectively superior. It would change what you are measuring.
Likewise, if your research question concerns actual time spent using a platform, asking participants to estimate their usage may introduce recall and estimation problems when reliable system logs are available.
This is why measurement should follow the process of turning the construct into an appropriate measurable representation, rather than beginning with the assumption that one category of data is inherently preferable.
Subjective Measures Have Characteristic Vulnerabilities
Self-reported and judgment-based measures can be influenced by memory, interpretation of questions, response styles, social desirability, current mood, reference frames, and willingness or ability to report accurately. The importance of these influences depends on the construct, instrument, administration conditions, and intended interpretation.
These limitations should be taken seriously, but they do not make subjective data invalid by definition. If the research question concerns a person's perceived stress, satisfaction, confidence, attitudes, or experienced quality, removing the person's perspective may remove precisely the information the study seeks.
The more specific distinction between self-report and observed or recorded measures is useful here because subjective and self-report are related but not perfectly interchangeable categories.
Objective Measures Also Have Characteristic Vulnerabilities
Device readings, administrative records, behavioral traces, and performance measures can appear authoritative because they produce precise numbers. Precision, however, does not guarantee that the variable represents the intended construct.
A wearable device can misclassify activity. Administrative data may contain missing or incorrectly coded records. Digital traces record what the system was designed to capture, not everything a participant actually did. A performance task can be affected by language, prior experience, accessibility, test conditions, or other factors beyond the intended construct.
More fundamentally, an objective variable can be conceptually wrong for the question. Login frequency may be recorded with near-perfect accuracy and still be an inadequate representation of cognitive engagement.
Watch Out
Do not confuse precision of recording with validity of interpretation. A system can measure the wrong thing with extraordinary precision.
Objective and Subjective Measures May Legitimately Disagree
Researchers sometimes treat disagreement between objective and subjective measurements as evidence that the subjective report is wrong. That conclusion may be premature.
Consider physical activity. Device-recorded movement and self-reported activity may differ because of recall error or reporting bias, but they can also differ because the methods define and capture activity differently. Similarly, measured sleep characteristics and perceived sleep quality need not correspond perfectly because physiological or behavioral characteristics and lived experience are not identical constructs.
Disagreement can therefore reveal measurement error, differences in construct coverage, or genuinely different aspects of the phenomenon. Determining which explanation is plausible requires attention to what each measure actually represents.
Using Both Can Be Useful When the Research Question Requires Both
Some studies benefit from collecting objective and subjective evidence together. An educational technology study might examine both system-recorded usage and students' perceived usefulness. A workplace study might combine recorded absence with perceived job stress. A health study might collect both device-recorded activity and participants' experiences of exertion.
This is useful when the research question genuinely concerns both aspects. It should not become a ritual of adding an “objective” measure merely to legitimize a subjective one.
Different measurements may represent different valid ways of approaching a construct, or they may represent genuinely different constructs. Your conceptual framework should tell you which interpretation makes sense.
Neither Category Escapes Measurement Error
Subjective measures may be affected by response processes; objective measures may be affected by devices, coding systems, observers, environmental conditions, sampling procedures, or data-processing decisions. Both can contain systematic and random error.
The useful question is therefore not “Which one has error?” Both can. Ask instead what sources of measurement error are plausible for the particular method, how consequential they may be, and what can reasonably be done about them.
06 · What This Means for You
Choose the Measure That Matches What You Want to Know
Start by completing this sentence as precisely as possible: “I need to know...” The answer often reveals whether subjective experience, observed behavior, recorded performance, physical measurement, or some combination is appropriate.
A simple decision framework
If your construct is inherently a perception, attitude, feeling, judgment, or experienced state
A well-designed subjective measure may be the most conceptually appropriate evidence.
If your question concerns actual recorded behavior, performance, events, or physical quantities
Prefer an appropriate observational, administrative, performance-based, or device-based measure when feasible.
If subjective recall is likely to be problematic and reliable records exist
Consider whether recorded data provide a closer representation of the variable you actually need.
If behavior and personal experience are both theoretically relevant
Consider measuring both, while treating them as distinct sources of information rather than assuming they should produce identical results.
If the objective measure is only a convenient proxy for your real construct
Do not prefer it merely because it appears more objective; evaluate the conceptual distance between the proxy and the target.
The best measurement strategy may also depend on feasibility, ethics, privacy, participant burden, access, cost, and the quality of available instruments. These constraints can legitimately influence the choice, but they do not change what each measure actually represents.
If feasibility forces you to use a narrower or more indirect measure, adjust the interpretation accordingly rather than quietly expanding the variable's meaning.
07 · A Quick Checklist
Before Choosing Objective or Subjective Measurement, Check the Fit
Before selecting your measurement approach, check:
Have you defined precisely what you want to know rather than simply what data are easiest to collect?
Is the construct itself experiential or perceptual, making subjective information theoretically relevant?
If you need actual behavior, performance, events, or physical quantities, is a reliable recorded or observational source available?
Have you identified likely sources of error or bias in each candidate measurement approach?
Does an apparently objective variable actually represent your construct, or is it merely a convenient proxy?
If objective and subjective measures are both used, have you specified what distinct information each contributes?
Are your conclusions phrased according to what each measurement actually supports?
