What Does a HEXACO-PI-R Percentile Mean Without a Named Reference Group?

What does the percentile say, and what is missing?

If the number on a HEXACO-PI-R report is a conventional percentile rank, it describes where a score stands within a comparison group. Without the group and method, you cannot tell what it means for you. It is not a percentage of a trait, a behavior probability, or a grade of whether the trait is good. For now, set aside labels such as “high,” “low,” or “typical.” This does not show the inventory itself is invalid; this comparison is simply unexplained.

A score and its percentile rank answer different questions. The score is the value produced from responses under a scoring procedure. A percentile rank starts with that score and locates it against scores in a reference group. The *Percentile and Percentile Rank* entry in the *Encyclopedia of Measurement and Statistics* defines the rank as the percentage of group scores below the obtained score. It stresses that this is relative standing: the same score can occupy different positions when the comparison distribution changes. The *Standards for Educational and Psychological Testing* also treats norms as a defined basis for interpreting scores, not an inherent property of a number.

For a hypothetical illustration, imagine one scale score compared with two groups with different score distributions. It could rank above a larger share of one group and a smaller share of the other, although the score stays the same. These answer different comparison questions. Methods may also handle tied scores or round ranks differently. Without the provider’s method, these are possibilities to clarify, not known flaws in this report. Ask what score was converted, by which rule, and against which distribution before treating the percentile as personal standing.

Sources: Standards for Educational and Psychological Testing; Percentile and Percentile Rank

Which HEXACO form and score did the report actually rank?

Ask first what score the report ranked. Was it a broad HEXACO domain or a narrower facet? Which item form and language did the person complete? Were the answers self-reports or observer reports? What score came immediately before the percentile: a raw scale total, an average, or a transformed value? The name HEXACO-PI-R alone does not answer these questions. Until the report or provider identifies them, readers cannot tell what the rank is attached to. The official HEXACO Personality Inventory-Revised materials distinguish 60- and 100-item forms, each available in self- and observer-report versions, and describe a 200-item form for longer facet-level scales. The official guide for non-academic use says HEXACO-60 is intended mainly to assess the six factors, with very brief scales for 24 of 25 facets; HEXACO-100 offers brief scales for all 25 facets; and HEXACO-200 offers full-length facet scales. Form therefore matters especially when a report presents a facet. Factor and facet results are not interchangeable. A raw score is derived from item responses before comparison with a reference distribution. A provider may transform it, for example onto a standardized scale, before deriving a percentile. To trace what the rank summarizes, the reader needs the scoring key and conversion steps. HEXACO’s official materials publish separate scoring keys for the 60- and 100-item versions and list descriptive statistics for English versions with Canadian undergraduate students identified as the population. That documents particular materials and data; it does not show that an outside report used those statistics as its percentile norms. Language and response source also belong in the check. The official inventory page lists multiple language versions and both self- and observer-report forms. The report should identify the version administered and scoring material applied. Self-report and observer-report begin from different sources of judgment, so the distinction matters when comparing scores. The 2018 study “Psychometric Properties of the HEXACO-100” examined English HEXACO-100 forms in online self-report and undergraduate self- and observer-report samples. It does not establish that these participants supplied norms for a particular report. Evidence about an instrument and evidence about a displayed percentile’s provenance answer different questions. A similarly named method should not be confused with a norm-table conversion. “Investigating Relative and Absolute Methods of Measuring HEXACO Personality Using Self- and Observer Reports” examined relative-percentile ratings: people estimated what percentage of a comparison group they believed was lower than a target. Its 142 Australian well-acquainted dyads completed traditional HEXACO-100 and relative-percentile facet measures; 78 participants repeated the latter two weeks later. The abstract reports mostly moderate reliability, generally lower than corresponding traditional scales, and that ratings often formed distributions unlike the expected uniform distribution. This study concerns respondents estimating relative standing, not the algorithm behind an unexplained report percentile. The shared word “percentile” does not make the methods identical. A precise request to the provider can follow the scoring path: identify the form and language; state whether responses were self- or observer-report; name the domain or facet; show how responses became a scale score and any transformation; then explain the percentile calculation and comparison data. Ask whether that group matches the form, language, and response source. These questions do not presume a flaw. They establish what the number represents. Until answered, do not treat the rank as standing against a known population.

Sources: HEXACO Personality Inventory-Revised; HEXACO-PI-R for Non-Academic Use; Psychometric Properties of the HEXACO-100; Investigating Relative and Absolute Methods of Measuring HEXACO Personality Using Self- and Observer Reports

What does a documented HEXACO comparison let you conclude?

An accessible copy of a HEXACO Personality Inventory-Revised report says its percentile indicates the percentage of respondents scoring below the reported value. It names its comparison group as Canadian university students who supplied self-reports in academic studies, and cautions that the percentiles might not apply to other populations. This supports a bounded reading: the rank is relative to that stated comparison, and the report limits transfer beyond it. It says nothing about the sample behind a different report that leaves its reference group unnamed. The example separates two numbers that can look interchangeable. Its scale score is transformed so that the average is 5.00; the percentile then expresses a position relative to the named group. The report notes that its online scoring changed before October 2022. Provenance therefore requires knowing which score was ranked, how responses became that score, and which distribution supplied the comparison. Because this copy is hosted on StudyLib rather than the official instrument site, it is an example of disclosed details, not evidence of current universal HEXACO policy. Its description cannot be assigned to another provider by inference. The official HEXACO Personality Inventory-Revised page offers a different kind of evidence. It provides scoring keys and descriptive statistics for English versions, labeling the statistics as based on Canadian undergraduate students. It lists 60- and 100-item materials, self- and observer-report forms, and describes a 200-item version for longer facet-level measures. These resources identify particular forms and the group behind those descriptive statistics. They do not show that every provider used those statistics as its percentile norm, or that a listed scoring key reproduces a provider’s conversion. Descriptive statistics can summarize a dataset without being the comparison distribution used in a report. “Psychometric Properties of the HEXACO-100” illustrates why instrument research answers a different question from report provenance. Its PubMed abstract says researchers examined the English 100-item form using online self-reports from 100,318 respondents and self- and observer-reports from 2,868 undergraduate students. It reports support for the hierarchical structure in two principal-components analyses, fairly low factor-scale intercorrelations, and strong self/observer convergence relative to discriminant correlations. These findings concern the specified form and samples. The abstract does not say either sample generated a vendor’s percentile table or validate an unreported transformation. A study evaluates defined scores; a scoring key explains how responses are combined; descriptive statistics summarize a dataset; a norm conversion locates an individual score in a reference distribution. These resources are related but not interchangeable. A reader cannot conclude, “HEXACO has Canadian student statistics, therefore my percentile compares me with Canadian students.” The official page establishes that descriptive statistics exist for specified English forms and a stated group. The report example shows what a named comparison and a limit on transfer look like. The study reports evidence about a defined form in specified samples. None reveals the method used by a separate report whose documentation is silent. Thus, a documented HEXACO comparison demonstrates useful disclosure, while an unnamed percentile remains uninterpretable as population standing until its provider identifies the conversion and reference group.

Sources: HEXACO Personality Inventory-Revised Report; HEXACO Personality Inventory-Revised; Psychometric Properties of the HEXACO-100

When is a named group still the wrong comparison?

Naming a reference group is necessary, but it does not by itself show that the comparison fits you or your question. A percentile is interpretable only in relation to a defined score distribution and justification for the interpretation. “General population” may sound broad and a large sample may sound stable; neither tells you how participants were recruited, whom they represent, or whether their scores can be compared fairly with yours. The group name is a starting point for scrutiny, not a guarantee of fit. The first issue is who entered the sample and under what conditions. Volunteers, online respondents, and university participants may differ from people who did not take part. That does not prove a sample is biased; it makes the sampling route relevant when extending results to a wider population. The Standards for Educational and Psychological Testing discuss describing the population represented by norms and evaluating its appropriateness for the proposed interpretation. Ask: who was eligible, how were participants recruited, and whom did the provider intend the scores to represent? Sample size, representativeness, and applicability are distinct questions. Time and language add boundaries. A provider should identify when comparison data were collected and which language version and form were used. A collection date alone does not make a norm outdated, and a translation alone does not make scores incomparable. These details show what evidence is needed to carry the comparison across settings. Official HEXACO inventory materials list different forms and language materials, and label English descriptive statistics as coming from Canadian undergraduate students. This gives the statistics a stated context. It does not establish that they supplied a particular third-party report's norm table, or that a different form has equivalent comparison data. The accessible HEXACO Personality Inventory-Revised Report hosted by StudyLib offers a contrast. It names Canadian university students who provided self-reports in academic studies as its comparison group and cautions that its percentiles may not apply to other populations. This makes the comparison inspectable; for readers outside that description, it leaves a question about generalization. The example does not show that the sample represents all Canadian university students or fits every purpose. It shows what disclosure can do: identify a boundary so readers and providers can examine whether crossing it is justified. A named group enables evaluation; it does not complete it. Administration and intended use matter too. If a report's form, language, response source, or scoring differs from the comparison data, the provider needs a basis for treating the scores as comparable. The Standards for Educational and Psychological Testing connect norm interpretation to proposed use. A sample suitable for describing a research group may not automatically support an individual decision in another setting. This is a question to investigate, not evidence that an undisclosed provider erred. Ask whether administration and scoring match those used to build the comparison, and which interpretation the documentation supports. The conclusion could change if a technical note identifies the sample, recruitment and collection period, form and scoring conversion, and why the comparison fits the report's stated purpose. Then you can judge fit rather than infer it from a label. Until then, ask more than “How many people were compared?” Ask: “Who were they, how were they sampled, and why does this distribution fit the interpretation I am being asked to make?”

Sources: Standards for Educational and Psychological Testing; HEXACO Personality Inventory-Revised; HEXACO Personality Inventory-Revised Report

Does reliability make an unnamed percentile interpretable?

No. Reliability asks how consistently scores are produced under stated conditions; a reference group identifies whose scores define a comparison. A repeatable score can still have an uninterpretable percentile if the report does not name the distribution behind it. These questions are separate, and neither alone establishes that a result supports a consequential decision. In the *Standards for Educational and Psychological Testing*, reliability and precision concern the consistency of scores and the uncertainty around them. Validity concerns the evidence and reasoning that support a particular interpretation for a particular use. They connect, but are not interchangeable: repeatability does not identify the percentile population, and naming that population does not show how much a score may vary. A consistent score also does not automatically justify every conclusion drawn from it. The study *Test-retest reliability of the HEXACO-100—And the value of multiple measurements for assessing reliability* illustrates why a reliability result must travel with its design details. The researchers administered the English HEXACO-100 twice, about 13 days apart, to an online Prolific sample; 416 participants remained in the final analyses after exclusions. They reported a median test-retest correlation of .88 for domains, .81 for facets, and .65 for items. The figures summarize consistency in that sample, interval, and score levels. They are not precision estimates for an unidentified reader or evidence about another provider’s norm group. The study also explains why the interval matters. A longer gap can mix measurement inconsistency with genuine change in the trait being measured; a very short gap can make it easier for someone to remember and repeat earlier answers. The authors chose roughly two weeks as a balance between those concerns. This balances concerns when estimating repeatability; it does not convert scores into population ranks. Interpret reliability in light of form, interval, sample, score level, and method. A domain result is not a facet or item result, and English-form findings do not establish what another language version or scoring implementation did. There is a fair reason to take instrument research seriously: well-designed studies can provide useful evidence about the consistency and properties of specified HEXACO scores. The 2018 *Psychometric Properties of the HEXACO-100* study examined an identified English form in online self-report and undergraduate self- and observer-report samples. Such evidence informs claims about the studied instrument and samples; it cannot supply another provider’s missing conversion record or identify the distribution behind an individual percentile. These are different questions. A percentile also inherits uncertainty from more than one step. First, the obtained scale score may be an imperfect estimate of a person’s tendency under the assessment conditions. Then, if the provider converts it to a rank, the estimated comparison distribution and the conversion method matter too. The displayed rank may be rounded, and ties may be handled in different ways; without the scoring documentation, a reader cannot know whether either issue affected this particular number. That does not show the rank is wrong; it cautions against reading more precision into a bare number than its documentation warrants. So ask for two records separately: the provider’s evidence about score precision, including which form, scale, sample, and interval its reliability evidence covers; and the percentile’s provenance, including the score transformation and comparison distribution. If the provider supplies only one, the other question remains open. Until both are clear, the percentile cannot support a confident claim about relative standing. A scale description may still serve as a tentative prompt for reflection, but repeatability cannot make an unnamed comparison group visible.

Sources: Test-retest reliability of the HEXACO-100—And the value of multiple measurements for assessing reliability; Standards for Educational and Psychological Testing; Psychometric Properties of the HEXACO-100

What evidence would justify using the result beyond private reflection?

Before treating a HEXACO-PI-R percentile as more than a reflection prompt, ask what evidence supports the interpretation and the use you have in mind. A rank can summarize relative standing in a defined comparison group, but it cannot by itself establish a cutoff, diagnose a condition, predict a specific act, or identify a suitable job candidate. For private reflection, a scale description may still suggest a question worth checking against experience. If a report describes a tendency toward careful planning, for example, the useful next step is to recall more than one situation: when did planning help, when did circumstances call for improvisation, and what counterexample complicates the description? They do not validate the percentile, and agreement with a description does not show that the score is accurate. Coaching can make the distinction more practical. A coach and client might use a report as one prompt among the person's own examples, goals, and context. They can discuss what the person notices without treating an unexplained rank as a factual summary of how they compare with others. If the conversation turns on a label such as “high” or “low,” the next step is to request the scoring and norm documentation rather than infer what the label means. HEXACO's materials for non-academic use discuss use of the inventory outside academic research, but that general guidance does not show how an unidentified provider administered, scored, interpreted, or explained this particular report. Nor does it establish that a given practitioner has qualifications or that the report's percentile was produced according to those materials. The standard for a consequential decision is different because the cost of a mistaken interpretation is greater. For selection or another employment decision, a provider or user would need evidence supporting the specific score interpretation and intended use, alongside procedures appropriate to that decision. That means asking what outcome the score is meant to inform, how the evidence connects this scale to that outcome, and how the result is combined with other information. An anonymous percentile supplies none of those links by itself. A population rank alone supplies neither a job-related criterion nor a defensible cutoff. The Standards for Educational and Psychological Testing treat validity as evidence for interpretations and uses, and discuss intended use, norms, reporting, and supporting documentation. Their guidance is general; it does not establish whether this provider followed it, create a legal conclusion, or validate this report for personnel decisions. There is a fair alternative explanation. The report may be brief while a manual or technical note elsewhere gives the form, scoring method, comparison sample, precision information, and intended use. A missing disclosure on the page is not proof that the provider has no documentation, and it does not show that HEXACO research is unsound. Ask the provider for those materials before drawing either conclusion. Until the answer arrives, match the inference to what is visible: use descriptive text, if useful, as a tentative prompt for examples; do not treat the percentile as a ranking, diagnosis, prediction, or employment score.

Sources: Standards for Educational and Psychological Testing; HEXACO-PI-R for Non-Academic Use

What should you ask the provider, and what can you do meanwhile?

Ask for the score path, the comparison, and the intended use as three separate pieces of information. A concise message could read: “Please identify the HEXACO-PI-R form and language used, whether my answers were self-report or observer-report, and whether this result is a domain or facet score. How were responses converted into the score shown, and how was that score converted into a percentile? Please describe the percentile formula, including how ties are handled.” This asks what the displayed number represents before asking what it means. The official HEXACO inventory materials distinguish forms, languages, response materials, and scoring resources; those resources help make the questions concrete, but their availability does not establish which one a particular provider used. Then ask: “What reference sample supplied the percentile distribution? Please describe who was included, how and when the data were collected, and which population the comparison is intended to represent. Is this a comparison for my form and language? What interpretation is this score intended to support, and is there information about its precision or uncertainty?” The 2014 Standards for Educational and Psychological Testing discuss describing norm populations and matching score interpretations to intended uses. They support asking about these features; they do not reveal this provider’s process. A technical manual or scoring note may answer even when a short results page does not. Keep precision separate from provenance. A reliability estimate may describe score consistency under specified conditions. It cannot tell you whose distribution generated a percentile. Conversely, naming a comparison sample does not show how much uncertainty surrounds an individual score or the estimated rank. Ask whether the provider can explain both for this form and use. General HEXACO research can inform interpretation, but cannot identify an undisclosed provider procedure. Until you have an answer, leave the comparative adjective blank. Do not call the result high, low, typical, or rare. If the report includes a plain-language scale description, treat it as a tentative prompt rather than a conclusion established by the percentile. Choose one observable behavior and consider when it appears, when it does not, and what the situation required. For instance, for a description about planning, note one time you prepared early and one counterexample when circumstances led you to act quickly. These reflections are not new test results or proof that the report is accurate. Avoid turning the exercise into rehearsed answers for a future assessment. If the question behind the report is repeated work friction, make that uncertainty more specific: what happens during planning, feedback, disagreement, or a change in priorities? A private, low-stakes Work Pattern Report at /assessment can offer prompts across ten decision-and-collaboration continuums. It has no norm, cutoff, type, or hiring score, and it is not validated for hiring, promotion, compensation, performance management, diagnosis, or surveillance. It cannot repair an unexplained HEXACO percentile. Use it only if describing your own work patterns would help you reflect; otherwise, continue learning how reports work at /topics. The immediate decision remains simple: request the score and comparison documentation, and withhold the rank label while you wait.

Sources: HEXACO Personality Inventory-Revised; Standards for Educational and Psychological Testing

What does the unnamed percentile mean for your decision today?

For now, the percentile is unexplained: the report has not established what score was ranked or which distribution supplied the comparison. That is narrower than saying the HEXACO-PI-R is unsound. The Standards for Educational and Psychological Testing discuss norm-referenced interpretations and supporting documentation; official HEXACO inventory materials describe particular forms and datasets. Neither authenticates an unidentified provider’s conversion. Instrument research can inform questions without filling in details the report omits.

The decision can change if a technical record identifies the form and language, score transformation, comparison sample, and intended use. Those details may support a bounded relative comparison if the sample and method fit the question. Even then, a percentile describes position in that distribution; it does not establish a verdict about the person or predict what they will do. Until the record arrives, set aside labels such as high, low, typical, or rare. Treat accompanying descriptions as tentative prompts, not confirmed conclusions.

If the underlying question concerns a recurring collaboration difficulty, choose one observable exchange and note what happened, the surrounding conditions, and what might support another explanation. This can organize reflection without repairing missing norm information. The practical sequence is to request the technical record, suspend the comparative label, and decide whether the remaining description helps you ask a more specific question.

Sources: Standards for Educational and Psychological Testing; HEXACO Personality Inventory-Revised