The BARSIT test is a brief paper-and-pencil test that estimates general learning ability based on 60 questions of information, vocabulary, and reasoning, to be answered in 10 minutes. It was designed for elementary school students and adults with that level of education, and in Mexico it has been used primarily for selecting candidates for operational positions. Its norms are outdated, and its validity in predicting performance is poorly documented, so it works better as a baseline assessment than as a screening tool.
Table of Contents
What is the BARSIT test and where did it come from?
BARSIT is an abbreviation for Barranquilla Rapid Survey Intelligence Test, In Spanish, it is known as the Barranquilla Rapid Test or Rapid Intellectual Ability Measurement. According to the UNAM thesis by Eloísa Franco Vázquez (1969), The story, as summarized in the author's manual, is this:
- In 1940, an Institute of Applied Psychology was created at the Barranquilla School in Colombia to study student performance.
- One of the tests that was tried was the Rapid Survey Intelligence Test by Rudolf Pintner, but its time limits proved too short for those students.
- Francisco del Olmo then developed BARSIT, in 1943 and 1944, to cover the same objective without that problem.
- The English version of the manual was published in 1958 by The Psychological Corporation of New York.
This data is important for HR because it explains the test's original purpose: to quickly rank large groups of schoolchildren according to their learning aptitude. Its use in personnel selection came later.
What does BARSIT measure?
It measures general cognitive ability with a strong verbal component. 1969 thesis Describe 60 items ordered by difficulty and five types of questions mixed together:
| Question type | What does it intend to evaluate? | What to keep in mind |
|---|---|---|
| General information | Knowledge acquired | It depends a lot on schooling and cultural environment |
| Vocabulary (opposites) | Word comprehension | Some words may sound different in Mexico than in Colombia or Venezuela |
| Verbal reasoning | Finding the concept that doesn't fit into a group | It blends reasoning with language mastery |
| Logical reasoning (analogies) | Relate elements according to a known rule | It is the closest thing to abstract reasoning |
| Numerical reasoning (series) | Deduce the rule of a series of numbers | It is the only type that is not answered with multiple choice. |
What it doesn't measure also matters: personality, values, integrity, soft skills, technical knowledge of the position, and experience. Nor is it a nonverbal test like the Raven's test, which works with figures and is closer to what psychometrics calls g factor. The authors of the Mexican theses insist on this difference: the BARSIT mainly collects verbal intelligence, information and numerical reasoning.
One practical consequence: since general knowledge and vocabulary depend on education, a low score may reflect fewer educational opportunities rather than a reduced ability to learn the job. This is why the comparison group is so important.
How is it applied and graded?
It is a group test, although it can also be applied to a single person. Broadly speaking, the procedure it outlines is... 1969 thesis This is it:
- Prepare the room. Good lighting, separate tables to prevent copying, and, for groups of more than 25 people, an assistant.
- Explain the instructions. The administrator reads them aloud, the participants solve some examples aloud, and it is verified that everyone understood how to answer.
- Time. There are 10 minutes from the start order to the stop order.
- Rate with the key. Each correct answer is worth one point. The total is called the raw score.
- Convert to percentile. The raw score is looked up in the norms table to obtain a percentile, which can be translated into a qualitative category, from very low to excellent.
According to that source, the manual doesn't require special training for the administrator, but it does require a thorough understanding of the test and the ability to guide the group. That doesn't resolve the interpretation issue: reading a percentile and making a decision based on it requires psychometric judgment.
We do not reproduce questions, keys, or answers here. The material is copyrighted, and if it circulates among candidates, the results cease to be informative about the individual. The same applies to online versions of unknown origin: if the format or timing changes, the manual's guidelines are no longer valid for comparison.
How are the results interpreted?
A raw score, by itself, means nothing. What gives meaning to the result is the percentile, and the percentile depends on the group you're comparing it to. In the BARSIT, this point is delicate because the documented reference groups are few and outdated.
| Rule | With whom was it prepared? | Font |
|---|---|---|
| Schoolchildren of Venezuela | Students from third to sixth grade, applied by Del Olmo and collaborators | Franco Vázquez, 1969 |
| Adults from Venezuela | Workers aged 20 to 30 with third to sixth grade education | Franco Vázquez, 1969 |
| Adults in Mexico | 679 men applying for railway worker positions, all with a sixth-grade education | López Ortiz, 1967 |
The 1969 thesis indicates that Venezuelan standards were being used in Mexico. López Ortiz He developed Mexican standards and warned that they are only useful for people with a sixth-grade education and socioeconomic conditions similar to those of his sample; with other groups, the comparison would be, in his words, grossly approximate.
The test use guidelines from the International Test Commission They generally ask the same thing: do not draw conclusions by comparing with standards that are irrelevant to the people being evaluated or that are outdated. For a candidate with a high school diploma or bachelor's degree in 2026, none of the tables above represent their group.
When reporting a result, always include three things: the raw score, the percentile, and the norm used (country, year, and population type). If the report only says "average level," you don't know what the person was compared to.
What does the evidence say about its reliability and validity?
Reliability is better documented than validity. Here is the data we found in open sources:
| Study | With who | What he found | What doesn't answer |
|---|---|---|---|
| Author's manual, according to Franco Vázquez (1969) | Schoolchildren, Caracas, 1948 | Test-retest reliability of .91 with weeks' difference; correlation of .71 with school grades | If it predicts job performance |
| López Ortiz (1967) | 679 applicants for worker positions | Half-half reliability of .87 on average and internal consistency of .91; correlation of .55 with Raven's Progressive Matrices in 75 people | The validity, which the author expressly leaves pending |
| Franco Vázquez (1969) | 413 applicants seeking to join or advance in a railway company | Total correlation of .54 with Raven | If it predicts job performance |
Three points to consider before reading this:
- They are old studios and from a single organization. The two Mexican theses were done with applicants for working-class positions, almost all men, more than 55 years ago.
- A moderate correlation with Raven's Progressive Matrices does not constitute validity for the position. Both authors state: the two tests measure different aspects of intelligence, and the Raven is not a useful criterion for job success.
- The reliability of a rapid test can be inflated. In tests with such tight time limits, split-half methods tend to overestimate performance, because missed questions count equally in both halves. Test-retest is more informative in this case.
In our review, we did not find any studies published in journals that linked BARSIT to job performance. The most useful reference for this category is the meta-analysis by Sackett and colleagues (2022), which estimated a validity of .31 for cognitive aptitude tests in general, compared to .42 for the structured interview. This is a category figure, not a specific test figure, and should not be extrapolated to the BARSIT without its own data.
For what positions does it make sense in the selection process?
The prudent approach is to use it only where it fits with your design and as just another font.
- It can contribute in operational positions with a primary or secondary education requirement, in processes with many applicants, as a first sign of ease in learning procedures or written instructions.
- It contributes little In professional, technical, or managerial positions: the test falls short for people with higher levels of education, and there are no regulations for them. A [unclear - possibly "a specific type of professional" or "a specific type of professional"] is a better fit for those roles. logical reasoning test more up-to-date, a work sample, or proof of technical knowledge.
- It doesn't work To evaluate attitudes, values, or team behavior, structured interviews, references, or tests designed for that purpose are needed.
No cognitive test, much less a 10-minute one, should be the sole determining factor. López Ortiz herself wrote in 1967 that the result was a starting point for a broader study of the person, not a definitive finding.
If your company already uses it, what should you check?
Removing it overnight isn't always feasible. These steps help you decide based on data:
- Identify the version and the standard. Which edition is used, who distributes it, and what percentile table is used for grading? If nobody knows, that's already a discovery.
- Compare the standard with your candidates. If your population has more education than the table shows, the percentiles will be inflated and almost everyone will look "superior".
- Remove automatic cuts. No candidate should be excluded solely because of this result, especially if the rule does not apply to them.
- Analyze your own data. With the support of a psychometrics specialist, compare the results with the performance or training of those who have already been hired. If there is no correlation, you have grounds to change the test.
- Protect consent and data. This explains the purpose of the test, who will see the results, and how long the results are stored, in accordance with applicable regulations. This text is not legal advice; please consult your legal department.
- Check the battery level. The guide on how to choose psychometric tests for work It helps decide what to measure based on the position and with what type of test.
If you're redesigning the evaluation of an operational position, combine a cognitive test with technical documentation, some competency tests that they observe what the position requires and a structured interview. At Hirint, the psychometric tests for Mexico They are applied online along with skills tests, and the results are comparable between candidates so that the team can make the decision with all the information available.
Frequently Asked Questions
What does the BARSIT test measure?
It assesses general learning aptitude through 60 questions covering general knowledge, vocabulary, and verbal, logical, and numerical reasoning. It has a strong verbal component and is partly dependent on educational attainment. It does not measure personality, values, integrity, or technical skills, so in selection processes it should be used in conjunction with interviews and other tests, never as the sole criterion.
How long does the BARSIT test take?
The timed test takes 10 minutes, plus the time for instructions and examples. This is one of its practical advantages, but also a limitation: in such a fast-paced test, reading speed significantly impacts the result. Maintain the same time and instructions for all candidates, because any change will skew the comparison with the norms.
Who created BARSIT?
It was developed by Francisco del Olmo in 1943 and 1944 at the Institute of Applied Psychology of the College of Barranquilla, in Colombia, based on a test by Rudolf Pintner whose timing did not work for those students. The English version of the manual was published in 1958. This is summarized in a 1969 thesis from UNAM that includes that manual.
Is BARSIT useful for personnel selection?
It can serve as supplementary data for operational positions requiring basic education, which is the population for which standards exist. For profiles with higher levels of education, the available tables do not represent the candidates, and the results become difficult to interpret. We did not find any open studies that link it to job performance, so it should not be used as an automatic filter.
Where can I find the PDF or the answers for BARSIT?
We don't share them here: the material is copyrighted, and knowing the answers will invalidate your results. If you're in HR, get the test and its manual through a channel that documents the version and rules. If you're a candidate, the best thing to do is get a good night's sleep and carefully read the instructions.
If you want to review how you evaluate your candidates in Mexico, at Hirint you can combine psychometric tests, skills tests and video interviews in the same process.