What Can an Informant Personality Report Add to Self-Report?

What can an informant report add to your self-report?

An informant report adds another person’s account of behavior they have had a chance to observe. Its value depends on whether that person repeatedly sees the behavior your question concerns; a second rating is not automatically a correction, truth test, or more accurate verdict. If you are deciding whether to ask someone, first name the behavior and context you want to understand, then consider who actually encounters it. What an informant can contribute remains uncertain and depends on the specific instrument and its evidence, so treat the account as a perspective to examine rather than a stand-alone conclusion. That distinction matters when a report is used for reflection: an example from someone who sees the relevant situation can sharpen a question about what happened, while a person with little access may only repeat an impression. The report adds testimony, not independent verification.

What can another person notice that you may miss?

Self-report and informant report can differ because the two people do not have equal access to every kind of information. The SOKA model—short for Self–Other Knowledge Asymmetry—organizes this difference around two dimensions: observability and evaluativeness. Observability asks how readily a characteristic shows itself in behavior that another person can see. Evaluativeness asks how strongly a characteristic carries social approval or disapproval. They describe routes to information, not fixed categories of traits or a ranking of raters.

In the study reported in “Who Knows What About a Person? The Self–Other Knowledge Asymmetry (SOKA) Model,” 165 participants provided self-ratings and were rated by four friends and up to four strangers in a round-robin design. The researchers tested a model in which self and observers could have different advantages depending on these dimensions. The abstract reports the predicted pattern: self-knowledge had an advantage for less observable characteristics, while observers had an advantage for highly evaluative characteristics. This is evidence for a proposed asymmetry under that study’s design, not a universal map assigning each personality trait to one preferred reporter.

The distinction makes sense because a person has direct access to private experience, such as thoughts or intentions that may not be expressed. An observer instead encounters outward conduct and its setting: what someone says or does, when it occurs, and how it appears to another person. Yet visibility alone does not settle what a behavior means. A quiet response, for example, might be observable while its motive is not; an observer can describe the response without thereby knowing the internal reason. Conversely, a person may report an inner experience accurately even when others cannot see it. These are differences in access, not proof that one side is mistaken.

Evaluativeness introduces a separate route. When a characteristic is socially judged, a person may have reason to manage how it is presented or interpreted; other people may notice the conduct that attracts that judgment. But being visible and socially loaded does not guarantee that an observer has a complete or unbiased view. The study’s model helps explain why the information available to self and observer can vary; it does not establish that observers always see through self-presentation, or that self-report is inherently distorted. A socially evaluative quality can also be expressed differently across settings, leaving an observer with only a partial sample.

Use these dimensions to frame the question before choosing a reporter. Ask: am I trying to understand a private experience, an outward behavior, or how a behavior is received by others? For a private experience, self-report may be the more direct source. For conduct that regularly occurs in front of someone, an informant may add a view of what is actually expressed. If the concern is socially evaluative, ask whether the observer sees the relevant situations and can describe behavior rather than simply assign a favorable or unfavorable label. The practical choice follows the target question; it does not follow a general rule that a friend, partner, or colleague is always the better judge.

Keep the model bounded. The SOKA study used a specified sample, ratings, and friend-and-stranger design; its abstract does not give a lookup table for every trait, relationship, or commercial report. In particular, it cannot tell a reader which informant version to trust without evidence about that instrument and intended interpretation. It does offer a useful prompt: locate the information needed for your question. Is it mainly inside the person, visible in repeated conduct, or tied to socially judged behavior? That prompt helps select a relevant perspective while leaving the account open to context and uncertainty.

Sources: Who Knows What About a Person? The Self–Other Knowledge Asymmetry (SOKA) Model

What kind of evidence can an observer contribute?

An observer report can be evaluated against at least three different questions: do observers reach similar judgments, do their judgments converge with the target’s self-description, and do their ratings predict a behavior criterion? These are distinct outcomes. Consensus means that multiple observers give related accounts; it says their judgments are reproducible across raters, not that they are correct. Convergence with self-report means two sources describe the person in related ways, but agreement alone does not establish which source is closer to a relevant fact. Prediction asks whether a rating is associated with an independently defined behavior or outcome. It is a more direct test of criterion relevance, but the answer still depends on which criterion was measured and for what purpose. The labels “accuracy” and “agreement” therefore cannot be treated as interchangeable evidence.

The 2010 meta-analysis “An Other Perspective on Personality: Meta-Analytic Integration of Observers’ Accuracy and Predictive Validity” makes this distinction visible in its design. It synthesized 44,178 target individuals from 263 independent samples and organized analyses around interrater consensus or reliability, self–other relations, and behavioral prediction. Its scale matters: this is a broad synthesis across studies, measures, and criteria, not a validation study of one named commercial report. The three analyses answer related but nonidentical questions about observer ratings. A finding in one column cannot simply be carried into another: dependable agreement among observers does not by itself demonstrate behavioral prediction, and self–other convergence is not a substitute for either test.

The synthesis also treats opportunity to observe as relevant to accuracy. The source reports that interaction frequency improves accuracy, while intimacy is needed for substantial increases in other-rating accuracy. These moderators point to the conditions under which an observer may have useful information: repeated contact can expose patterns, and an intimate relationship can provide access that a brief encounter lacks. They do not establish that intimacy guarantees accuracy, that frequent contact always concerns the behavior a report asks about, or that one observer is best for every trait and setting. An observer may know a person well in one domain yet rarely see the specific conduct at issue in another.

For a reader, the key question is therefore what kind of support the report actually provides. If it presents another person’s rating, that demonstrates that an additional perspective was collected. To claim agreement, the report would need to show how ratings relate across observers or to self-report. To claim prediction, it would need evidence against a defined behavioral criterion, with the relevant population and use made clear. To claim added value beyond self-report, it would need to show that the informant information contributes something useful after the self-report is considered; the mere presence of two scores does not demonstrate that increment. This is an evidential distinction, not a judgment that observer information lacks value.

The meta-analysis finds predictive utility in the literature, so the careful conclusion is not that observers only offer opinions. Rather, observer judgments can support different claims, and the appropriate claim depends on the outcome tested. A report meant to prompt reflection may make modest use of a second account as a question to investigate. A report used to make a consequential decision would need evidence tied to that decision and its criterion. Broad findings about observer ratings cannot validate an unnamed instrument, informant form, population, or use. When documentation says “accurate,” ask: accurate by which outcome, against what reference, and for whom? A criterion must also be distinguished from the rating itself: repeating the same questionnaire in another voice is not an independent behavioral outcome. The stronger claim requires a specified external measure and a clear link between that measure and the intended interpretation.

Sources: An Other Perspective on Personality: Meta-Analytic Integration of Observers’ Accuracy and Predictive Validity

What makes an informant a useful observer?

Choose an informant for their opportunity to observe the behavior in question, not simply for their relationship label. Someone who repeatedly encounters the relevant situations can describe a broader sample than a person who has only seen one striking moment. The practical choice depends on the target behavior, setting, and period: the useful observer is the person who has had access to those particulars and can report what they actually noticed. This access-first rule is an editorial inference from the evidence, not a tested screening instrument or guarantee that the selected person will be accurate. A person’s title or status—manager, partner, or best friend—cannot substitute for that direct access. The prompt should name a bounded behavior rather than ask someone to summarize the whole personality; that keeps the requested judgment closer to what the observer has actually encountered.

The study “Friends and Strangers: Acquaintanceship, Agreement, and the Accuracy of Personality Judgment” offers a bounded comparison. Approximately 150 undergraduates completed self-descriptions using a Q-sort procedure. Each was also judged by two close acquaintances and two strangers who had viewed one spontaneous interaction recorded on video. The study compared judgments from people with continuing acquaintance against judgments formed from a single brief observation, rather than comparing informants chosen for different work or family roles. In the abstract-reported result, close acquaintances agreed more with one another and with targets than the strangers did. The design therefore illustrates why a person’s access history can matter when interpreting an observer account.

That result does not make “close friend” a universal answer. The sample consisted of undergraduates, the self-description and ratings used a particular Q-sort method, and the strangers’ evidence came from one videotaped interaction. The contrast tests that particular gap in acquaintance; it does not rank every friend, partner, family member, or colleague, nor establish that the closest person is accurate about every behavior. Closeness can bring a longer record of observations, but it can also bring assumptions, shared loyalties, or a stake in how the target is seen. Acquaintance increases opportunity to know; it does not certify the interpretation.

Translate the evidence into four practical questions before inviting someone. First, do they regularly encounter the behavior or setting you want to understand? Second, have they seen it across enough occasions to distinguish a recurring pattern from an unusual event? Third, can they answer about the specified period and context rather than filling gaps with a general impression? Fourth, are they willing to give their own account independently, without pressure to match your answer or defend a shared interest? These questions are a reasoned selection aid drawn from the study’s contrast and the broader role of observation opportunity; the study did not test these four criteria as a checklist.

The best informant may differ with the question. For example, if the question concerns how someone handles a recurring team handoff, a person who regularly participates in that handoff may have more relevant observations than a close relative who rarely sees that work. This is an illustrative application, not a claim about an actual participant or a result tested by the study. If no one has suitable access, the honest choice may be not to request an informant rating for that question. A report based on a willing observer’s limited exposure should be read as a narrow account of what that person could see, not as a complete portrait.

Sources: Friends and Strangers: Acquaintanceship, Agreement, and the Accuracy of Personality Judgment

When can two perspectives be complementary?

Two perspectives are complementary when each contributes observations or interpretations that the other source does not contain, and both bear on the same clearly framed question. Complementarity is about the information sources can bring together; it does not mean that their scores should be averaged, that a difference itself proves added value, or that a general personality test can diagnose a condition. Whether combining perspectives improves a particular score interpretation is a separate empirical question requiring evidence for that instrument and use.

The systematic review “Concordance and Added Value of Informant- Versus Self-Report in Personality Assessment: A Systematic Review” offers a limited example from a clinical research area. It searched Scopus, PsycINFO, and PubMed for studies published from 2000 through 2018 and included 46 studies of adults and older adults assessing personality pathology. The abstract reports moderate concordance between self and informant reports, higher concordance in intimate relationships such as partners, and concludes that the perspectives can be complementary. That conclusion belongs to the review’s pathology-assessment literature; it is not a finding about ordinary workplace personality reports or evidence that those reports can establish a diagnosis.

The review is useful here because it makes the distinction between contribution and sameness concrete. Two people can report overlapping impressions while also having access to different episodes, settings, or meanings. A partner may see recurring behavior in shared daily life; the person themselves may be better positioned to describe private intentions or experiences. The review’s reported relationship pattern also suggests that the connection between raters can matter, but it does not show that intimacy always improves every judgment or that a partner is the right reporter for every question. The underlying studies were heterogeneous, and an abstract-level summary cannot supply a transferable effect size or a rule for selecting an informant in general personality assessment. The review title refers to “added value,” but its abstract-level conclusion that reports may be complementary should not be inflated into a claim that informants always improve prediction or add statistically useful information beyond self-report. Those are specific tests: they ask whether one source contributes to a defined outcome after the other source is accounted for. Nonidentical accounts may prompt a richer discussion without demonstrating that either account improves a score’s accuracy. Keeping those meanings separate prevents a useful conversation from being mistaken for psychometric evidence, and preserves the review as a bounded example rather than a general recommendation to collect another rating.

For low-stakes reflection, translate the idea into a conversation about one behavior rather than a contest between two profiles. Someone considering a work-pattern issue might ask a colleague who regularly attends the relevant meetings for one recent example of how they respond when plans change, then compare that observation with their own recollection and ask what context each person saw. This is an editorial application, not an outcome tested by the clinical review. The colleague’s account can add a situated observation to examine together; it should not be converted into a diagnosis, a score average, or a verdict about ability. If the observer did not see the behavior, their answer cannot supply the missing perspective.

Sources: Concordance and Added Value of Informant- Versus Self-Report in Personality Assessment: A Systematic Review

What should you expect from a second perspective?

For a broad expectation about close observers, the 2025 review “Self- and Observer Reports of Personality” synthesizes findings from full-length Big Five and HEXACO measures. Its abstract describes generally comparable means and rather high convergent agreement between self and close-observer reports, while noting qualifications by trait: agreement is lower for some cooperation-related characteristics. The review also discusses similarity and assumed similarity for selected traits. This is a review-level picture of particular measure families and close relationships, not a guarantee about every observer, personality inventory, short form, or commercial report. The qualification matters: “close observer” and “full-length Big Five or HEXACO measure” define the scope of the synthesis, rather than a universal property of informant reports.

The trait qualifications resist an easy rule such as “observers see the person more clearly.” The reports may converge substantially overall while differing in how well that pattern holds for specific characteristics. Similarity between the person and observer, including assumed similarity discussed in the review, is another reason not to read agreement as an independent stamp of truth. These findings describe patterns across research; they do not identify the best person to ask about a particular behavior or establish what happened in an individual interaction. A broad synthesis can guide expectations, but a reader still needs to keep the question and the observer’s actual opportunity to see the behavior in view.

A different large-scale finding pushes back on a common claim about why an informant report is needed. “Self–Other Agreement in Personality Reports: A Meta-Analytic Comparison of Self- and Informant-Report Means” compared self and informant report means for 33,033 people across 152 samples. The abstract reports a small average difference, δ = −.038, generally not different from zero; stranger ratings were a qualification. That group-level result does not support the blanket story that people invariably rate themselves more positively and therefore need an informant to correct them. It is counterevidence to a universal self-enhancement narrative, not proof that self-ratings are unbiased in every case or that informants never identify something a person has missed.

These two sources answer different broad questions. The 2025 review summarizes convergence for close observers and full-length Big Five or HEXACO measures, with trait-specific variation; the 152-sample meta-analysis addresses whether average self and informant means are systematically different. Neither result decides whether a specific report is accurate for one person, whether a particular pair of ratings matches, or whose account better fits a behavior in a given setting. A population-level pattern cannot settle the details of a single report. Use the findings to avoid expecting automatic correction or universal agreement, then treat any particular observation as a claim to explore in context rather than as a winner in a score comparison.

For a practical question, start by asking what the second person could actually contribute: a distinct observation, a different setting, or a report of an experience unavailable to them? If the answer is only that they are expected to provide the objectively correct version, the research summaries above do not justify that expectation. If they can describe a relevant episode, their account may still be worth considering as an additional source of information. The useful outcome is a better specified question—what happened, where, and what each person noticed—not a decision that one reporter must always be preferred. For example, if someone wants to understand whether they leave planning discussions before decisions are clear, a teammate who attends those discussions could describe a recent instance and what remained unresolved. The person can then add what they understood or intended at the time. This illustration does not presume either person is right; it shows how a second account can make the behavior and setting more concrete. If the teammate has not attended those discussions, selecting them for their general closeness would not fill that information gap.

Sources: Self- and Observer Reports of Personality; Self–Other Agreement in Personality Reports: A Meta-Analytic Comparison of Self- and Informant-Report Means

What does the report need to establish before you act on it?

Before treating an informant score as evidence for a decision, ask what interpretation the score is meant to support and whether evidence supports that interpretation for this use. The Standards for Educational and Psychological Testing defines validity in terms of the degree to which evidence and theory support the interpretations of test scores for proposed uses. In this framework, validity does not belong to a test as a blanket property: a score can support one carefully bounded interpretation without supporting a different, more consequential one. The question is not simply whether the instrument has been studied, but whether the relevant evidence bears on the conclusion someone wants to draw from this particular score.

That distinction matters when an assessment has both a self-report and an informant form. Evidence about observers in general, or about another instrument that also asks someone else to rate a person, cannot by itself establish what this report’s informant score means. The evidence needs to match the instrument and version, the wording and format of the informant form, the population in which it is used, the criterion against which the interpretation is checked, and the proposed use. A provider should be able to explain these links rather than offer a general statement that informant reports are useful or that the assessment is validated. This is a claim-specific standard of support, not a demand that every reflective question be settled by a high-stakes validation study.

For a private, low-stakes reflection, an informant’s example can be treated as a prompt to investigate: what behavior did they observe, in what setting, and what might explain the difference between their view and yours? That use makes no claim that the resulting score ranks you or predicts an outcome. When the score is used to affect someone’s opportunities or treatment, the evidentiary burden changes. A general personality report cannot be assumed suitable for hiring, promotion, compensation, performance management, diagnosis, or surveillance merely because it produces a number or collects ratings from more than one person. The Standards supply the validity framework; they do not establish that any unnamed product meets it for those decisions.

A useful question to put to a provider is: “What evidence supports interpreting this exact informant form, for people like me, against which relevant criterion, for the decision I have in mind?” This compact question is an applied heuristic for opening the discussion, not a formal checklist or a substitute for reviewing technical documentation. If the answer identifies only a different population, a different form, general observer research, or a use unlike yours, that is a reason to narrow the interpretation rather than fill the gap with confidence. If evidence does match the proposed interpretation and use, a particular informant assessment may be justified; the conclusion is conditional, not that informant assessment is inherently unsuitable. Without that match, keep the score within the modest role of a reflection prompt and do not turn it into a verdict.

Sources: The Standards for Educational and Psychological Testing

What is the next useful step?

If a work-pattern question still feels vague, write down one behavior, the setting where it occurs, and what you want to understand about it. Make the prompt narrow enough to answer from observation—for example, ask what happens during a recurring handoff, rather than asking whether you are generally organized. You can invite a willing person who regularly sees that situation to describe one specific example: what they observed, when it happened, and what effect it had. Listen for the event before interpreting it, then add what you remember or intended. Keep the exchange voluntary and reflective; if no suitable observer comes to mind, or the question concerns something they cannot see, self-observation is a reasonable next step. You can decline to share or stop if the exchange becomes evaluative. An informant is optional, and the useful result is a clearer question to consider, not a verdict about you.

If the unresolved issue is how your own tendencies combine across planning, collaboration, conflict, adaptation, and learning, the live [Work Pattern Report](/assessment) offers a separate self-reflection route. It is based on your responses, not informant ratings, and does not provide an employment decision score. For more guidance on reading assessments, continue to [personality report topics](/topics).

Questions readers ask