Norms and Percentiles Depend on a Clearly Defined Comparison Sample
You see “higher,” “lower,” or a range on a personality assessment page and want to ask: “Does this tell me which percentile I am in?” Do not rush to turn a label into a ranking. Norms are references established from a clearly defined comparison sample and its score data; percentiles indicate where a score stands relative to that specified group. Both require a verifiable instrument, scoring method, comparison sample, and conversion rules. When information is missing, personality scores cannot establish an individual ranking, nor
Norms and Percentiles Depend on a Clearly Defined Comparison Sample
You see “higher,” “lower,” or a range on a personality assessment page and want to ask: “Does this tell me which percentile I am in?” Do not rush to turn a label into a ranking. Norms are references established from a clearly defined comparison sample and its score data;percentiles indicate where a score stands relative to that specified group. Both require a verifiable instrument, scoring method, comparison sample, and conversion rules. When information is missing, personality scores cannot establish an individual ranking, norms, validity, or comparative conclusions; the most accurate record is
. (APA, “Test norm”) (ETS, Standards: Quality and Fairness, glossary)
This article explains the boundaries of measurement terminology. It does not calculate additional percentiles for any personality product or judge whether someone's personality is good or bad. Personality traits are not an examination report card. Even when an instrument provides comparison information, it describes a position relative to a specified sample, rather than ability, worth, health, hiring eligibility, or future performance.
First distinguish norms, percentiles, and scores themselves
Imagine that Zhou completes a personality questionnaire and sees “Openness: higher.” He wants to turn “higher” into a specific individual percentile. This step requires more known information, rather than stronger wording.
A norm is a comparison benchmark established through a clearly defined standardization or reference group. APA's terminology explanation places test norms on the basis of a tested standardization group and an analysis of its scores. (APA, “Test norm”) Thus, “compared with whom” is part of the interpretation, rather than a footnote: whom the sample represents, when it was collected, its regions or language settings, the version used, and how participants were recruited and data processed all limit what the comparison can say.
A percentile rank does not mean “how many questions you answered correctly” or “what percentage of a personality trait you have.” ETS's terminology standards define it as the proportion of scores below a particular score in a defined group. (ETS, Standards: Quality and Fairness, glossary) Personality questionnaires generally do not have this sense of “correct answers.” Describing a personality result as a percentile still refers only to its position relative to a particular reference sample; it cannot automatically become a judgment of superiority, diagnosis, or destiny.
Do not read “middle percentile” as “I am ordinary in every situation,” or “higher percentile” as “I am better.” A percentile condenses a score's position in a sample distribution. It does not explain the scope of the questions, life situations, behavioral consequences, or personal worth for you. (ETS, Standards: Quality and Fairness, glossary)
The comparison sample must be explainable
Interpretable norms or percentiles should at least let readers know the answers to the following questions:
Check
What needs to be available
What can be said when it is missing
Conclusions that cannot be supplied
Instrument and scores
Instrument name, version, language, item/scoring explanations, and the type of score displayed on the page
“The current score's calculation and scale are insufficiently explained”
Different instruments can be converted, or this score has been validated
“Compared with people of the same age/the national population”
Comparison rules
Definitions of raw scores, transformed scores, ranges, or percentiles, and how tied scores are handled
“Percentile conversion rules are unknown”
Inferring a particular percentile from “higher”
Applicability boundaries
For whom, which version, and what uses the comparison applies
“Applicability to the current use context is Unknown”
Scores can be used for diagnosis, screening, or prediction
This does not ask readers to establish norms themselves. It simply turns “where do I rank?” back into a request for reviewable information: if the comparison sample has no name or explanation, there is no verifiable group behind the comparative conclusion.
The International Test Commission (ITC) guidelines for translating and adapting tests also place score interpretation within the intended population, evidence, and documentation. When instruments are used across languages or cultures, matching names alone cannot justify assuming equivalent comparisons. (International Test Commission, 2017) This methodological boundary does not mean that a particular language or group “cannot be compared.” It means that every comparison needs to match its specific version, population, and evidence.
The same framework does not mean the same measuring scale
The Big Five is a widely used trait framework, rather than a naturally unified questionnaire, sample, or percentile table. Soto and John's research on the BFI-2 discusses a specific instrument with defined items and structure. It cannot make every assessment labeled “Big Five” automatically identical or provide norms or percentiles for another product. (Soto & John, 2017)
Official material makes this distinction tangible: Berkeley Personality Lab's explanation of the BFI-2 explicitly states that there is no official BFI-2 manual with published norms. (Berkeley Personality Lab, Big Five Inventory) This does not mean that the BFI-2 has no research, or that no instrument can have a comparison sample. It shows that “using a well-known framework” and “having publicly verifiable norms” are separate matters.
Research by Srivastava, John, Gosling, and Potter used a large adult Internet sample to examine mean-level patterns across age groups. (Srivastava et al., 2003) It can help us see that research findings always come with a sample, timing, instrument, and research question. It cannot supply an individual percentile for a reader whose instrument and version have not been specified. A research sample is not automatically a universal set of product norms.
A quick-answer framework: ask four questions before deciding whether to retain Unknown
When you see “high, medium, low,” “above most people,” or any wording that appears to rank people, pause through the following four steps. This framework does not calculate scores, certify instruments, or demonstrate that results are valid; it only prevents missing information from being written as a personal conclusion.
What score is this? Record the instrument, version, language, and the page's exact wording about the score. If the score type is unspecified, do not call it a percentile.
Compared with whom? Check whether the comparison sample's population, timing, region/language, recruitment, and processing are public. Without a sample, there is no verifiable “most people.”
How is it converted? Confirm whether the page explicitly provides raw scores, conversion rules, or a percentile definition; “higher” is not a synonym for a particular percentile.
What can this tell us? Even complete comparison information describes relative position only within its stated scope; it does not independently establish your ability, health, relationships, career fit, or future.
Return to Zhou's situation: the page displays only “higher,” without publishing a sample name, sample date, scoring version, or conversion rules. The fact he can retain is “the page gives the description ‘higher.’” He cannot rewrite this as “the 80th percentile,” or conclude from it that he is more open than others, that the assessment has been shown to be valid, or that it will produce a particular career outcome. For percentiles, norms, and these comparative conclusions, the correct record is Unknown.
Why “looking precise” is still insufficient
Numbers often create a sense of completion, but the precision of comparative language depends on the information supporting it, rather than the number of decimal places. Without a clearly specified reference sample, readers cannot know whom a percentile is relative to; without a clear version and purpose, they cannot know whether a sample is appropriate for the current interpretation. Testing standards connect score interpretations and intended uses with the evidence supporting those interpretations, so displaying a score cannot replace that information. (International Test Commission, 2017)
This also explains a common misunderstanding: “My friend and I both scored high, so which of us is higher?” If the instruments, versions, languages, scoring rules, and reference samples used by both people have not been checked, the question itself may lack an interpretable common scale. Even when the conditions are the same, this article does not provide individual comparisons, rankings, or judgments of superiority. Big Five results are suited to limited clues for self-observation, rather than acting as an adjudicator between people.
FermatMind: which information must remain Unknown
FermatMind's public Big Five assessment page describes 90- and 120-question versions, five domains and 30 facets, and scoring methods including reverse scoring of 1–5 responses as y = 6 − x, means and z/T transformations. With usable reference parameters, it describes an estimated percentile under a normal model; that model-based reference is not an observed population percentile. Without usable norms, the linear compatibility score is not a population percentile either. Read the reference source and its CALIBRATED, PROVISIONAL or MISSING status. These disclosures do not provide the complete item key or establish representative norms or product validation. Product-specific study samples, reliability, validity, measurement error, cross-language equivalence, retest properties, meaningful change, intervention effects, individual prediction and career or relationship matching remain Unknown where supporting evidence has not been made public.
Accordingly, this article does not turn other Big Five research, BFI-2 pages, research samples, or general terminology into product evidence for FermatMind. They also cannot support conclusions about “your ranking,” “which norms you belong to,” “the results have been shown to be valid,” or “comparability with a particular group.” Unknown does not mean a negative conclusion; it means that public, reviewable information is insufficient for this article to take that inferential step.
If you want to use results for self-reflection, begin with the Big Five to understand the language of dimensions, or treat the Big Five personality assessment as an entry point into “measurement—interpretation—action—review.” None of these is a norm certification, ranking service, or tool for high-stakes decisions.
Frequently asked questions
Does a percentile mean what percentage of a personality trait I have?
No. A percentile describes a score's relative position in a defined comparison sample, rather than the percentage of a trait you possess. It is not equivalent to the percentage of correct answers, ability, or a judgment of worth. (ETS, Standards: Quality and Fairness, glossary)
Without a public sample, can I estimate my percentile from “high”?
No. “High” may be a descriptive range on a page, but without a public comparison sample and conversion rules, it cannot be estimated as any specific percentile. Keep the percentile as Unknown.
Big Five research has reported sample averages. Can they directly serve as my norms?
No. Research samples, scales, languages, timing, and purposes all have boundaries. They can explain what a particular study observed, but cannot automatically become comparison standards for another instrument or an individual result. (Soto & John, 2017) (Srivastava et al., 2003)
Do high norms or high percentiles mean a better personality?
No. Norms and percentiles describe, at most, relative position within a specified group. They cannot independently support conclusions about personality superiority, health, work performance, hiring, education admission, relationships, or future outcomes. (ETS, Standards: Quality and Fairness, glossary)
Can personality percentiles be used for hiring or other important decisions?
This article does not support that use. A self-assessment result, unspecified norms, or an individual ranking cannot independently bear any high-impact decision; this article is only for educational self-observation and understanding methods.
Visible sources and boundaries
American Psychological Association. Test norm. This article uses it to define norms as dependent on a standardization group and score analysis; it does not validate any specific product. Accessed: 2026-07-23.
Educational Testing Service. ETS Standards: Quality and Fairness (glossary). This article uses it to define percentiles as score positions relative to a defined group; it does not provide norms for personality products. Accessed: 2026-07-23.
International Test Commission. (2017). The ITC Guidelines for Translating and Adapting Tests (second edition). This article uses it to explain that interpretation across populations, languages, and versions requires matching evidence and documentation; it does not generate rankings for any individual. Accessed: 2026-07-23.
Soto, C. J., & John, O. P. (2017). The Next Big Five Inventory (BFI-2): Developing and Assessing a Hierarchical Model With 15 Facets to Enhance Bandwidth, Fidelity, and Predictive Power. Journal of Personality and Social Psychology, 113(1), 117–143. https://doi.org/10.1037/pspp0000096. This article uses it to explain that a named scale is a specific construction; it does not provide FermatMind norms, percentiles, or validity. Accessed: 2026-07-23.
Berkeley Personality Lab. Big Five Inventory. https://www.ocf.berkeley.edu/~johnlab/bfi.html. This article uses its BFI-2 norms explanation as a public example; it does not establish norms for other instruments. Accessed: 2026-07-23.
Srivastava, S., John, O. P., Gosling, S. D., & Potter, J. (2003). Development of Personality in Early and Middle Adulthood. This article uses it to explain that research conclusions belong to a specific sample and study design; it does not provide individual comparison rules for any product. Accessed: 2026-07-23.
This article is for education, self-observation, career exploration, and capability development. It does not constitute medical or psychological diagnosis and is not a basis for hiring, employment selection, education admission, legal, or financial decisions. For norms, percentiles, or product validity without public supporting evidence, retaining Unknown is more accurate than manufacturing a ranking.
Personality Norms and Percentiles: Check the Sample | FermatMind
