Evidence review
Graphology: what the evidence says
Graphology claims handwriting reveals personality. Here is the history, the strongest studies, why the claim persists, and what it means for hiring.
Handwriting Personality Test editorial teamPublished 2026-10-21Updated 2026-10-219 min read
Graphology is the practice of inferring personality, character, or ability from handwriting. It is not the same as handwriting measurement, and it is not the same as forensic document examination. It is a specific interpretive claim: that the way a person forms letters carries information about the kind of person they are.
The claim has been tested for more than a century. The results are not a blank slate. They point in a consistent direction: graphologists perform poorly when they are asked to predict outcomes, and the features they cite are not stable enough to support individual decisions. This article explains the evidence and why the practice survives anyway.
What graphology claims
Graphology is not one doctrine. Different schools use different sign lists. Common claims include: a right slant indicates emotional expressiveness; a large signature indicates confidence; a closed loop indicates secrecy; a heavy pressure indicates strong drives; a rising baseline indicates optimism. The signs are often combined into a narrative report.
The reports tend to be broad. A graphologist may describe a client as 'reserved but warm with close friends, practical yet imaginative, and more sensitive than they appear'. Such statements are hard to falsify because almost everyone recognizes part of them. That is the Barnum effect, and it is one reason graphology feels accurate to the people who receive a reading.
A scientific claim needs a defined feature, a defined outcome, and a test that could fail. 'Right slant means outgoing' can be tested: measure the slant, measure extraversion, and see whether the relationship holds across people and occasions. The claim fails when the feature is unstable or the correlation is near zero.
How graphology became established
The modern practice traces to nineteenth-century France. Michon named it and promoted the idea that strokes correspond to character. Crépieux-Jamin developed a more systematic method, and the practice spread through Europe and Latin America. It entered personnel selection, education, and even some medical settings.
The cultural authority of graphology varied by country. In France, it became a common feature of hiring for certain roles. In the United States, it was more often a popular or corporate curiosity. The variation matters because it shows that graphology's acceptance was social and historical, not the result of a decisive scientific trial.
Early tests were not kind. Binet's studies in the 1890s found little support for the claimed correspondences. Later researchers introduced better controls: neutral text, blind scoring, multiple raters, and real outcomes. The better the design, the smaller the effect tended to become.
The meta-analytic evidence
The most cited review is Neter and Ben-Shakhar's 1989 meta-analysis. It pooled studies in which graphologists made predictions about job performance or similar outcomes. When the content of the writing was neutral, the average validity was close to zero. When the content was not neutral, graphologists did better, but so did non-graphologists, which suggests that the text itself was carrying the information.
Ben-Shakhar and colleagues' 1986 study is often cited as a direct test in a personnel setting. Graphologists' predictions did not outperform simple alternatives. Klimoski and Rafaeli's 1983 review reached a similar conclusion: the evidence did not support handwriting analysis as a predictor of personal qualities.
A later study by Dazzi and Pedrabissi (2009) reported a modest association between some graphological signs and self-reported traits. Even if that association is real, it does not establish that a graphologist can make a useful prediction about an individual. A small group-level correlation is not a hiring test, a diagnosis, or a reliable portrait.
Why the strong claim fails
Handwriting is variable. The same person produces different slants, sizes, and letterforms depending on speed, pen, paper, mood, and context. A feature that changes from page to page cannot reliably index a stable trait.
The inference is underdetermined. A closed 4 may reflect a school model, a copied document, or a habit. A right slant may reflect page position or a fast hand. The graphologist selects one explanation from many possible ones, and the selection is not tested.
Feedback is poor. In everyday life, people rarely learn whether a graphological reading was correct. The reading is plausible, the client agrees with parts of it, and no one runs a controlled comparison. Without feedback, a confident practice can persist for decades.
Graphology and hiring
Employment decisions require evidence that a method is job-relevant, reliable, and fair. Graphology has not met that standard. A meta-analysis with near-zero validity is not a basis for screening candidates, and the practice raises fairness questions because the features are unrelated to the work.
Some employers use graphology as one input among many, often with the idea that it adds a little information. The evidence does not support that use. A weak, unstable signal can add noise rather than information, and it can distract from structured interviews and work samples that have better support.
If you are a candidate and encounter a handwriting test, you can ask what the test measures, how it is scored, and what evidence supports its use. Those questions are reasonable for any selection instrument. The answers may tell you more about the employer than the test tells them about you.
Graphology is not forensic document examination
Forensic document examiners compare a questioned writing with known writing to judge whether the same hand produced it. They study line quality, pen lifts, construction, and other features. Their task is identification, not personality. The two fields are often confused because both involve handwriting and both can appear in legal settings.
Forensic examination also has error rates and limitations. The point here is not that it is infallible. The point is that the question is different. A document examiner does not tell you whether the writer is honest; a graphologist does, without the same comparative method or evidence base.
This site's tools are neither. The slant and signature tools measure geometry. The quiz labels a set of shape choices. None of them identifies a writer or diagnoses a personality.
What to do with graphology claims
Enjoy the history and the vocabulary, but ask for evidence when a claim is used to judge a person. A single letterform is not a character reference. A signature is not a personality test. A confident report is not a validation study.
If you want to test a claim yourself, choose a specific feature and a specific outcome. Measure the feature on several pages, measure the outcome with a validated instrument, and see whether the relationship holds. That is a small, honest experiment. It is more informative than another reading.
The most defensible position is modest: handwriting is a rich record of learning, habit, and movement. It can be described, compared, and studied. The claim that it reveals a stable personality is not supported well enough to use on people.
Sources and further reading
- Neter, E., & Ben-Shakhar, G. (1989). The predictive validity of graphological inferences: A meta-analytic approach. Personality and Individual Differences, 10(7), 737-745.
- Ben-Shakhar, G., Bar-Hillel, M., Bilu, N., Ben-Abba, E., & Flug, A. (1986). Can graphology predict occupational success? Journal of Applied Psychology, 71(4), 645-653.
- Klimoski, R. J., & Rafaeli, A. (1983). Inferring personal qualities through handwriting analysis. Journal of Occupational Psychology, 56(3), 191-202.
- Dazzi, C., & Pedrabissi, L. (2009). Graphology and personality: An empirical study on validity of handwriting analysis. Psychological Reports, 105(3), 1255-1268.
Questions
- Is graphology scientifically valid?
- The strong claim that handwriting reveals personality is not supported by the evidence. Meta-analyses of graphologists' predictions have found near-zero validity in personnel settings.
- Who invented graphology?
- Jean-Hippolyte Michon coined the term in the 1870s. Earlier writers had made similar claims, and later systematizers developed elaborate sign lists.
- Can graphology predict job performance?
- The evidence does not support it. Studies and meta-analyses have not found useful predictive validity for personnel decisions.
- Is graphology used in France?
- It has a history of use in some French hiring contexts, but history is not evidence of validity. The practice has been criticized by researchers for decades.
- What is the difference between graphology and forensic handwriting analysis?
- Graphology infers personality. Forensic document examination compares writing samples to judge whether the same hand produced them. They have different goals, methods, and evidence.
- Why do graphology readings feel accurate?
- They often use broad, flattering statements that fit most people. That is the Barnum effect. The feeling of accuracy is not a validation study.
- Does this site use graphology?
- No. The quiz labels shape choices and the tools measure geometry. The site does not claim to read personality from handwriting.
- Where can I read more?
- The sources below include the Neter and Ben-Shakhar meta-analysis, the Ben-Shakhar et al. study, and the Beyerstein and Beyerstein collection.
Keep reading
The result is a filing label for shapes, not a graphology reading. The methodology page explains the scoring rule.
Handwriting Personality Test editorial teamResearch and editorialLast updated 2026-10-21