When Should a Personality Report Use More Than One Source?

The answer depends on the claim, not the number of tests

Use more than one source when a report’s claim needs corroboration, concerns behavior another person can observe, or will guide a consequential decision. Ask whether each source bears on the same conclusion and adds information the others could not provide. Two similar questionnaires may repeat one person’s perspective; a relevant observation may bring a different vantage point. Neither becomes persuasive simply by being called a second source. For low-stakes self-reflection, one transparent self-report can be a provisional prompt. “I describe myself as preferring time to think before responding” stays close to what the respondent reports. A claim that the score establishes how that person will act in future situations goes further and needs support suited to that interpretation. Readers need to know what conclusion the report draws from a score and what evidence supports it. The 2014 Standards for Educational and Psychological Testing frame validity around evidence for proposed interpretations and uses of scores. That principle concerns the basis for the report’s claim; another person’s opinion about this individual is a separate kind of information. More personal information cannot repair weak evidence that the instrument measures what the report claims. The Standards discuss collateral information in professional assessment; casual reflection does not automatically call for an observer. Stakes change how much support a conclusion should have, but do not create a universal minimum number of sources. Before taking another quiz or asking someone to rate you, identify the exact conclusion you are being asked to accept. Then ask whether the proposed source can test it, or whether the report first needs clearer evidence for what its score means.

Sources: Standards for Educational and Psychological Testing

Two different questions: does the score mean this, and what else do we know?

A personality report raises two different questions. First, does the assessment support the meaning assigned to its score for this purpose and population? Second, what else is known about this person that may clarify or qualify the interpretation? An observer may add context about behavior; that account cannot establish that the questionnaire measures what its report says it measures. The Standards for Educational and Psychological Testing define validity as evidence and theory supporting interpretations of test scores for proposed uses. In plain language, validity is support for a specific conclusion drawn from a score, not a permanent stamp attached to a test. A report that says a score reflects a tendency to plan carefully makes one interpretive claim. A further claim that it establishes how someone will act in every workplace needs evidence for that broader use too. Support for one interpretation does not automatically transfer to another. The Standards describe several ways to examine that support. Test content matters: do the questions represent the named construct, or only a narrow slice? The scoring structure matters: do responses relate in a way consistent with the proposed dimensions? Relationships with other measures or relevant outcomes can also support or challenge the interpretation. The case depends on how evidence fits together and applies to the population and use at issue. These are questions about the instrument and its interpretation, not extra observations about one reader. Reliability, also called precision, asks how consistent scores are and how much measurement error may affect them. A score can be consistent without supporting the meaning a report assigns it. The Standards caution that simply calling a test “reliable” is not enough: documentation should describe the evidence, the group it came from, and the conditions under which it was obtained. Consistency does not establish that a score measures the intended characteristic or justifies a particular conclusion. Person-level information is a separate source. It might include the respondent’s examples, an observer’s account, an interview, or records of behavior in a defined setting. Such material can help assess whether a description fits what has happened or needs narrowing. It does not replace evidence about the questionnaire’s items, scoring, or intended interpretation. More data about a person and better evidence for a test are different kinds of support. “Validity: On the Meaningful Interpretation of Assessment Data,” an article about medical education examinations, makes a related argument: scores are not simply valid or invalid; evidence supports or refutes a particular interpretation at a given time. This is an analogy from another assessment field, not direct validation research on personality reports. Its useful point is to ask what conclusion is being made and what evidence bears on it. Check whether a report identifies the assessment, the population its evidence concerns, and the use its interpretation is meant to serve. If that information is absent, a second opinion may still prompt reflection, but it cannot close the instrument-level gap. Narrow the conclusion to what is supported, or withhold it until relevant evidence is available.

Sources: Standards for Educational and Psychological Testing; Validity: On the Meaningful Interpretation of Assessment Data

What should a second source add?

A second source is useful when it changes what can reasonably be concluded. Before requesting another questionnaire, rating, or record, name the uncertainty: Is the report describing a tendency visible to others? Does it concern a particular setting or period? Is a response hard to interpret without context? If added information cannot answer a defined question or alter the interpretation, it may make a report look fuller without making it more informative. The Standards for Educational and Psychological Testing describe relevant, credible collateral information as potentially helpful and list methods such as structured behavioral observations, checklists, ratings, interviews, and records. This is professional assessment guidance, not a rule that every low-stakes personality questionnaire must gather outside information.

Compare three possible additions. A similar questionnaire may produce a fresh score but preserve the same respondent’s perspective, item interpretations, and response conditions. It may help assess consistency across occasions when timing and purpose make that comparison meaningful. It is not independent confirmation simply because it comes from another provider. A knowledgeable observer changes the vantage point, but only for behavior they have had a fair chance to see. A time-bounded example or record changes the kind of information: it can anchor a question in a defined situation, such as how plans changed during a particular project. That example does not establish a broad trait by itself; it can test whether a description fits that circumstance.

Different methods can reveal different aspects, but difference alone is not a quality mark. An observer may have limited access, remember unusual incidents, or have an interest in the outcome. A record can establish that an event occurred while leaving its meaning uncertain. An interview can clarify an answer, yet depends on the questions asked and evidence available. The Standards for Educational and Psychological Testing advise professionals to check collateral information’s accuracy and reliability, particularly when a third party may have a vested interest. A reader can ask who supplied the information, what they could observe, which period it covers, and whether it addresses the same claim as the assessment. Relevance depends on the match between source and question, not apparent authority.

The American Psychological Association’s guidelines for occupationally mandated psychological evaluations offer a higher-stakes example: use multiple sources of relevant, reliable information, and give greater weight to evidence with stronger known relevance, reliability, and validity. It shows why counting sources is a weak shortcut. Two poorly matched inputs do not necessarily outweigh one relevant observation, and several similar ratings may repeat a shared limitation. An additional source earns its place when its contribution is clear and proportionate to the decision.

A report should explain how distinct information becomes one conclusion. If scores or observations are combined into a composite, the Standards for Educational and Psychological Testing say its basis and rationale should be provided. Readers can ask whether unlike inputs were averaged, considered side by side, or weighted by an explained method. Without that account, a combined label can conceal differences or imply unearned precision. Before adding evidence, complete this sentence: “I need this source because it could help determine whether…” Then identify what result would change, narrow, or leave the conclusion unchanged. If no plausible result would affect the interpretation, the extra source may not be needed.

Sources: Standards for Educational and Psychological Testing; Professional Practice Guidelines for Occupationally Mandated Psychological Evaluations

When can another person add a useful lens?

Another person can add a useful lens when the report makes a claim about behavior that person has had a fair chance to observe. Their account may describe how someone contributes in meetings or follows through on a shared task. It cannot settle what the person privately felt or intended. The two reports answer overlapping but different questions; the claim should remain within what their evidence supports. A 2007 meta-analysis of Big Five research illustrates the overlap. Across research samples, corrected average correlations ranged from .46 to .62 across the five dimensions. Each source also contained substantial unique variance, and acquaintance duration strongly moderated agreement. These group-level summaries are not a standard for an individual pair. They do not show that averaging ratings improves a report or that one source is always more accurate. They suggest that perspectives often converge while retaining distinct information. The 2025 review, “Self- and Observer Reports of Personality,” describes a similar pattern in literature on closely acquainted pairs using full-length Big Five or HEXACO measures. Its abstract reports relatively high self–observer convergence in many cases, with somewhat lower convergence for cooperation-related traits. This supports the possible value of a knowledgeable observer; it does not validate a particular commercial report or observer form. The review also complicates the idea that closeness simply makes a rating better. Closely acquainted people showed some similarity and assumed similarity for Openness to Experience and Honesty-Humility, and to some extent Big Five Agreeableness. Shared dispositions or seeing oneself in someone close may contribute to agreement. Closeness can help, but agreement is not automatically independent confirmation. Consider an illustrative work question: a person says they prepare carefully before a meeting but rarely speak early, while a colleague describes them as a steady contributor once discussion begins. Those accounts need not cancel out. The person can report private preparation; the colleague can describe when contributions become visible. If the report’s claim concerns outward participation in that setting, the colleague’s observation is relevant. If the claim concerns private confidence or motives, the colleague’s account is incomplete. This illustrates different access to behavior and experience; it is not general evidence about behavior. A 2019 systematic review offers a bounded comparison from a different field. Reviewing 46 studies of personality pathology, it reported moderate concordance. Informants added information for externalizing traits and interpersonal functioning; self-reports better captured intrapsychic characteristics. The review noted little empirical research among older adults. This clinical construct and population cannot establish the same pattern for everyday, nonclinical personality reports. It illustrates that source value depends on the question. It does not justify carrying clinical findings into general trait interpretation. When a report proposes another person’s rating, ask whether that person knows the relevant setting, has seen the behavior over a useful period, and is being asked about something observable. A close friend may know broad social patterns but not task behavior at work; a colleague may see task behavior but not private effort or preference. Keep the self-report’s perspective in view, too. The useful second lens is the one that can see the part of the claim the first source could not, with its limits stated plainly.

Sources: Self- and Observer Reports of Personality; The Convergent Validity between Self and Observer Ratings of Personality; Personality Pathology in Older Adults: A Review of Self-Informant Agreement

Which source fits the question being asked?

Match a source to the part of a report’s claim it can inform. A self-report records how someone describes thoughts, preferences, and intentions. An observer can add a view of behavior in settings they know. Repeated observations or records help establish whether a pattern occurred in a setting and period. Relevance and opportunity to observe matter more than a source’s title or number. Personality language can stretch beyond what one perspective sees. “I prefer time to think before responding” reports a preference. “I rarely contribute in meetings” concerns behavior others may observe. “I miss deadlines” concerns events checkable against a defined work period or project record. These claims may overlap, but are not interchangeable. A source useful for one does not settle another. Consider an illustrative report that calls someone “reserved in groups.” The person can describe feeling quiet or tired in a meeting. A colleague who attended may recall how often the person spoke, but not what they thought or whether they contributed in writing. A manager who sees deadlines may discuss timeliness, yet have little basis for judging private effort or preference for solitude. Each account answers less than the broad label suggests. The review *Self- and Observer Reports of Personality* discusses self- and observer questionnaires for full-length Big Five or HEXACO measures, focusing on closely acquainted observers such as spouses, relatives, or close friends. It reports high self–observer agreement, somewhat lower for several cooperation-related traits, and similarity or assumed similarity for some traits among close pairs. This supports treating an observer as potentially informative, not treating closeness as proof of accuracy. The review does not establish who knows a particular work behavior best. Records have a different boundary. A calendar, project log, or written exchange may establish when something happened; it rarely explains motive. A missed deadline may reflect workload, unclear ownership, or planning difficulty. The record anchors discussion in an event, but cannot alone turn it into a stable personality conclusion. *Standards for Educational and Psychological Testing* names interviews, observations, ratings, records, and other collateral material as professional assessment inputs, while advising users to consider accuracy and reliability. This supports careful source choice, not collecting every record for casual reflection. A useful request names the question: “How do I contribute in the weekly design meeting?” is easier to answer with self-reflection plus observations from attendees than “Am I a reserved person?” If the concern is whether behavior repeats, set a time window and look for instances. Ask what an observer saw directly, what they inferred, and which situations they did not see. Keep private experience in the person’s account rather than asking colleagues to guess it. Choose the source with access to relevant information, and state what remains outside its view. Ask what uncertainty an extra questionnaire or informant is meant to reduce, whether it covers the same setting and period, and whether it changes the conclusion. If no source can see the behavior, narrow the claim or choose a better-defined observation. An uninformed opinion adds to the pile, not its fit.

Sources: Standards for Educational and Psychological Testing; Self- and Observer Reports of Personality; Personality Pathology in Older Adults: A Review of Self-Informant Agreement

What does disagreement tell you?

Disagreement is a reason to inspect what each source was asked to describe, when it was observed, and what each person could know. It does not establish that one account is wrong. If a report combines a self-rating of private effort with another person’s rating of visible participation, the difference may reflect distinct vantage points rather than a contradiction. Narrow the conclusion until behavior and context are precise. Start by checking whether the accounts refer to the same behavior and period. “I prepare carefully” and “I speak early in meetings” are not opposite claims; preparation and speaking are separate actions. Even two ratings of speaking may refer to different settings, such as a familiar small group and a large formal meeting. A report that collapses these situations into one trait label can make context look like inconsistency. Ask how often the observer saw the behavior, how familiar they are with the setting, and whether the self-report covers the same interval. If access differs, give each account a limited scope instead of forcing a winner. Then consider other explanations before interpreting the gap. Sources may understand a word differently, recall different examples, use response scales differently, or be affected by ordinary measurement imprecision. An observer may have repeated opportunities to see one kind of conduct and little access to another. The Standards for Educational and Psychological Testing discuss collateral information such as observations, interviews, ratings, and records, and emphasize considering the accuracy and reliability of the information. Check the source and its limits; disagreement does not always require another test or more data. Research on self and observer reports also gives reason not to treat a mismatch as an error to erase. The 2007 meta-analysis, “The Convergent Validity between Self and Observer Ratings of Personality,” reported corrected average correlations from .46 to .62 across Big Five dimensions, alongside substantial unique variance in both self and observer reports. These are group-level results from research samples, not an expected score gap for an individual. They show overlap and perspective-specific information. The 2025 review, “Self- and Observer Reports of Personality,” likewise summarizes convergence among closely acquainted observers and self-raters, while noting similarity and assumed similarity between close pairs for some traits, including Openness and Honesty-Humility and, to some extent, Agreeableness. Closeness can bring useful knowledge, but it does not make agreement an independent check by itself. A practical sequence follows: align the behavior and time window; check each source’s opportunity to know; consider wording, setting, and measurement; then decide whether a narrower description or a targeted observation would resolve what remains. If the accounts still differ, report that uncertainty and lower the scope or confidence of the interpretation. Do not average divergent ratings into a single verdict unless the assessment’s documented method supports that calculation. A discrepancy may prompt clarification; it does not license a label.

Sources: Standards for Educational and Psychological Testing; Self- and Observer Reports of Personality; The Convergent Validity between Self and Observer Ratings of Personality

How should the decision’s stakes change the evidence?

The amount of evidence a report needs should rise with what someone plans to do with it. For private reflection, a clear self-report can be enough to start a question: “Does this description fit how I tend to prepare for uncertain work?” The result remains a prompt, not a settled account of the person. The reader can compare it with specific experiences. No formal observer is required by default. Coaching and development call for a more defined check. Coaching should ask whether a tendency helps explain a particular work pattern and what might be tried next. If a report suggests that someone prefers time to consider a decision, the conversation could identify a recent decision, available time, what the person did, and what followed. Later examples can show whether a small change in preparation helped in that context. These observations help assess whether the report’s language is meaningful for the goal. Another perspective is worthwhile when it can illuminate the agreed question, not as a vote on the person. The threshold changes when an evaluation may materially affect someone. A personality report should not carry a consequential conclusion on its own merely because its score is precise or several people supplied information. The decision-maker needs evidence supporting the exact interpretation and use, a process suited to the question, and a clear account of how the conclusion follows. Where another source is relevant, its role and limitations should be explained. Evidence about one person can add context; it cannot establish that an instrument supports the proposed use. If that support is absent, more questionnaires or ratings cannot manufacture it. The American Psychological Association’s “Professional Practice Guidelines for Occupationally Mandated Psychological Evaluations” offer a bounded example of higher-stakes practice. Approved in 2017, the guidelines are intended for psychologists conducting evaluations tied to employment or credentials, including some fitness-for-duty questions. They recommend choosing tools validated for a population appropriate to the evaluation, using multiple sources of relevant and reliable information, and documenting opinions in clear language. They also connect the evaluation to a defined referral question and relevant job demands. This is guidance for specialized practice, not a rule that every workplace conversation or casual questionnaire must gather collateral information. Ask what question is being answered, whether the information fits it, and how its quality was weighed. The 2014 “Standards for Educational and Psychological Testing” clarify the difference between person-level information and support for a test interpretation. They state that adequate support for proposed interpretations for specific uses will generally draw on multiple sources of evidence. Here, that means evidence supporting the interpretation, such as how test content represents the construct and how scores relate to relevant information. It does not mean that an individual must take several personality tests. The Standards also describe greater consequences as a reason for greater care, and note that a decision may sometimes be reversible or corroborated by other information. These principles explain why support should fit the intended use. They do not validate a particular commercial report or prescribe one universal evidence package for personality assessments. Apply the distinction with two questions. First: what evidence supports this report’s interpretation for the proposed purpose and population? Second: what additional information about this person would clarify the claim in this situation? A “yes” to the second cannot repair a missing answer to the first. For self-reflection, a reader may stop after checking a transparent report against experience. For coaching, agree on an observable question and revisit relevant examples. For a consequential evaluation, request an explanation of the referral question, evidence for the intended use, and how relevant information is weighed. If the answer is only “we gathered more opinions,” the conclusion still needs a stronger basis. As stakes rise, demand better-matched support and a fair, documented process; do not use extra personality scores to create certainty the assessment cannot support.

Sources: Professional Practice Guidelines for Occupationally Mandated Psychological Evaluations; Standards for Educational and Psychological Testing

What should you ask before accepting the report’s conclusion?

Ask the report’s author or provider to state the conclusion in one sentence: what does this score say about which tendency, in what context, and for what use? Then ask what evidence supports that interpretation for this instrument and population. The *Standards for Educational and Psychological Testing* define validity in relation to evidence and theory supporting score interpretations for proposed uses. If the provider cannot identify the intended use or supporting evidence, narrow the conclusion rather than seeking more opinions about you. Next, ask what a proposed second source would add. Is it meant to check behavior another person can observe, clarify a setting, or address a specific uncertainty? Ask how the source’s opportunity to observe and relevance to the question are judged, and how conflicting or missing information changes the conclusion. Finally, ask what the result will be used to decide. For private, low-stakes reflection, a transparent self-report can be a prompt to compare with your experience; it need not become a formal verdict. If consequences are substantial, ask for an appropriate process and evidence matched to that use. The American Psychological Association’s *Professional Practice Guidelines for Occupationally Mandated Psychological Evaluations* recommend multiple relevant, reliable sources in that specific professional setting; they are not a blanket instruction for casual questionnaires. Keep the conclusion proportional to the evidence, including uncertainty or disagreement. A useful question to take into a conversation is: “What would change your interpretation, and what does this report leave unresolved?” If the answers remain unclear, pause before acting on the conclusion.

Sources: Professional Practice Guidelines for Occupationally Mandated Psychological Evaluations; Standards for Educational and Psychological Testing