Handwriting personality tests: what the evidence says

What does graphology claim?

Graphology is the practice of inferring character from handwriting. Its practitioners work from features that can genuinely be measured on a page: the slant of the letters, the size and regularity of the script, the pressure of the pen, the width of margins, the spacing between words, the shape of loops, the way a signature differs from ordinary writing. Each is read as a sign of something about the writer, so a rightward slant might be taken to indicate sociability, heavy pressure intensity, cramped spacing reserve.

The underlying idea is not unreasonable on its face. Handwriting is a motor habit produced by a particular nervous system, it feels personal, and it is stable enough that we recognise a friend's writing on an envelope. Graphology has been taught in structured courses for more than a century, mostly in Europe, and has been used at times in personnel selection, particularly in France. Its practitioners are generally sincere and often very experienced. The question here is narrower: whether the inferences come out right when checked.

What did the controlled studies find?

The best-known test of the occupational claim came from a group that ran two empirical studies and reviewed the methods used in earlier work (Ben-Shakhar et al., 1986). Professional graphologists judged handwritten scripts and predicted how well the writers were doing in their jobs. The crucial design point was the content of the scripts. People asked for a handwriting sample often write an autobiographical passage, and that passage carries information anyone could use. The graphologists' judgements correlated about .2 to .3 with supervisors' ratings, which is roughly what a psychologist with no knowledge of graphology achieved from the same scripts, and a simple linear model built from the information in the text, such as education, outperformed both. In a second study, graphologists asked which of eight professions each of 40 successful professionals belonged to did no better than a chance model.

A later meta-analysis pooled 17 studies covering 63 graphologists, 51 non-graphologists and 1,223 scripts and reached the same place: graphologists were no more successful than non-graphologists at predicting future performance from handwriting (Neter & Ben-Shakhar, 1989). Where accuracy appeared, it traced back to what the writers had said rather than to how they formed their letters.

That is a fair summary of the evidence as it stands. It is not a claim that handwriting carries no information, nor a judgement about the people who practise graphology. It is a finding about one inference: from letter forms to traits and job performance.

Why does a handwriting reading feel accurate?

Several ordinary things combine. The strongest is the one Forer demonstrated with a class of students. He handed every student the same personality sketch, drawn from an astrology book, and told each of them it came from their own test results. Asked how well it described them, they rated it close to the top of the scale, about 4.26 out of 5 as the demonstration is commonly reported (Forer, 1949). Statements that are broadly true of most people, and slightly flattering, read as personal when you believe they were written for you.

A second is the interview effect. Readings are often delivered in conversation, and a skilled reader picks up a great deal from your reactions, your manner and, as the graphology studies showed, from what the sample itself says. A third is selective memory: the hits stay with you and the misses do not.

The same pattern shows up wherever a system is tested rather than experienced. In a double-blind trial, expert astrologers were asked to match natal charts to the case files of volunteers, which included life-history information and personality profiles. They performed no better than chance, in a design they had agreed to in advance (Carlson, 1985). A method can feel reliable to its practitioners and its clients and still not survive that kind of test.

What can handwriting actually tell you?

Some things, and not the ones people usually want. Handwriting reliably identifies the writer, which is why forensic document examination is a real speciality with an evidentiary standard behind it. Comparing two samples to decide whether one hand produced both is a different task from reading a character out of one sample.

Handwriting also carries a few unsurprising facts about the writing itself: whether the writer was hurried, what surface and pen they used, which national script they were taught. Anything about health belongs to a qualified professional and a proper examination, not to an online tool.

What should you use instead?

The problem graphology set out to solve has a solved version. Asking people to rate how accurately a set of statements describes them, across enough statements per trait, produces scores that are stable over time and that relate to behaviour and to how others describe the same person. The method is unglamorous, and it works.

The trait structure those questionnaires recover came out of decades of lexical and questionnaire research, and it survived repeated attempts to replace it (Goldberg, 1993). Reliability rises with the number of items (Nunnally & Bernstein, 1994), and very abbreviated measures understate how much traits matter, so a six-question test is not a shortcut to the same answer (Credé et al., 2012).

• Check what model the test is built on, and whether that model has been replicated.
• Check how many questions stand behind each score.
• Check what your score is compared with, and how that sample was collected.
• Check whether the result could have come out differently, or whether it would fit anyone.

How does Wellington measure this?

Wellington is a science-based personality assessment from Therabot Labs LLC. It measures personality as continuous traits, built on the Big Five and HEXACO models, and writes the results back as a personal report rather than a type.

Wellington asks you about yourself, which is the method with the evidence behind it. The free Snapshot is 60 questions, about 7 minutes, and measures the six broad dimensions: Extraversion, Agreeableness, Conscientiousness, Emotional Stability, Openness to Experience, Honesty-Humility. Each score is the average of several statements rated on a five-point accuracy scale, set against a reference sample as a percentile and reported in one of three bands, Quiet below the 30th percentile, Balanced to the 70th, High above. Our norms are provisional until the norming sample is complete. The bands and the Portrait names exist to make the result easier to talk about; the percentile is the measurement, and it is always shown alongside. The first ten questions are answered before anything is asked; after question ten you sign in with an email so your answers are saved and your results can be shown. There is no card.

The Wellington Membership, $4.99 a month, adds the Portrait, 336 questions in 4 chapters, to reach 90 traits, including every Big Five facet and all 24 VIA character strengths (Peterson & Seligman, 2004), with domain overviews, a growth plan and a downloadable PDF, then adds a short weekly module toward all 255 traits across eight domains. In the written report, every trait has three pages, one per band, written in advance by the writing and psychology team, and your answers choose which you read. Wellington is a wellness tool for reflection, not a medical or psychological diagnosis, and it does not replace care from a qualified professional. It is not used for hiring, and you can export or delete your data at any time. The methods page describes how the questions were built.

Questions people ask

Is handwriting analysis scientifically valid for personality?

The published tests say no. Two controlled occupational studies found graphologists doing no better than a non-graphologist reading the same scripts, and no better than chance at identifying a writer's profession (Ben-Shakhar et al., 1986), and a meta-analysis of 17 studies found graphologists no more successful than untrained judges (Neter & Ben-Shakhar, 1989). Handwriting identification, a different task, is much better supported.

Why do handwriting readings often seem right?

Mostly because the statements fit almost anyone. Forer showed that students given an identical generic sketch rated it as a highly accurate description of themselves (Forer, 1949). Readings delivered in person add live feedback from your reactions, and the sample itself usually carries autobiographical content a reader can use.

Can handwriting show anything real?

Yes, a few things. It identifies the writer, which forensic document examiners do to an evidentiary standard, and it can show haste, an unfamiliar pen or surface, and the national script someone was taught. Those are facts about the writing rather than about character.

What should I take instead of a handwriting test?

A self-report inventory built on the Big Five or HEXACO, with several questions per trait and a stated comparison sample. The free Wellington Snapshot is one, at 60 questions and about 7 minutes, and the BFI-2, the IPIP-NEO and the HEXACO-PI-R are free research instruments that report the same kind of scores.