Short answer: which tool is being used?
The tool most often cited when psychologists talk about "paranormal belief" is the Paranormal Belief Scale (PBS), published by Jerome Tobacyk and Gary Milford in 1983 in the Journal of Personality and Social Psychology1. Tobacyk himself later printed a revision, usually called the RPBS (Revised Paranormal Belief Scale), in 2004 in the International Journal of Transpersonal Studies2. The revision has 26 items across seven subscales, with divination and astrology falling under the precognition subscale2.
What to grasp before reading any figure on "the share of paranormal believers": this scale does not measure one single thing. It measures seven different things, and a respondent can score high on one subscale but low on another. So the sentence "what percentage of people believe in the paranormal" only means something when it states which subscale, which sample, and which year.
Why a scale instead of a single question?
If you only ask "do you believe in the paranormal?", the answer depends on what the respondent takes "paranormal" to mean. A person may believe in souls but not in divination, in fate but not in witchcraft. A combined question lumps those people into one group and loses information.
The original 1983 paper by Tobacyk and Milford handled that problem through factor analysis: it put forward a set of items, then examined which groups the answers clustered into1. This approach lets different belief dimensions be separated instead of treated as one block. The original paper also examined the links between those dimensions and personality traits1.
In other words, the scale was not created to judge whether someone is "superstitious", but to answer a narrower question: which group of phenomena does this person believe in, at what level, and do those beliefs come with particular psychological traits.
What are the seven subscales of the 2004 revision?
Tobacyk's official 2004 print lists seven subscales2. The table below records the seven subscales under the exact names the revision uses, along with the content direction each subscale targets. The "example question direction" column is Nhat Nguyet's rewording, not the verbatim item text in the scale.
| Subscale | Content targeted | Example question direction (reworded) |
|---|---|---|
| Traditional religious belief | Organised, widespread religious beliefs | "Do you believe what your religion teaches about the afterlife?" |
| Psi | Human ability to act through the mind | "Do you believe people can use the mind to affect objects?" |
| Witchcraft | Magic, spells, sorcery | "Do you believe some people can use magic to cause harm?" |
| Superstition | Small everyday taboos | "Do you believe certain numbers or objects bring bad luck?" |
| Spiritualism | Contact with the dead, haunting | "Do you believe people can contact the dead?" |
| Extraordinary life forms | Creatures not recognised by science | "Do you believe there are undiscovered mysterious creatures?" |
| Precognition | Divination, astrology, foretelling the future | "Do you believe divination or astrology can foretell the future?" |
This division has practical consequences. When an article cites "paranormal believers" but only mentions divination and astrology, it is in fact talking about the precognition subscale, not the whole scale2. When another article talks about belief in souls, it is talking about the spiritualism subscale. The two figures cannot substitute for each other.
How did the 1936 study measure belief in divination among American students?
Before the standardised scale existed, there were individual surveys of particular kinds of belief. One of the earliest in this line is the study by C. M. Diserens and T. W. Wood, published in 1936 in the Journal of Applied Psychology, surveying belief in divination among American students3.
That study has value as a historical marker: it shows the question "do students believe in divination" was posed and measured long before the PBS or RPBS existed3. However, it is a single survey of one group of students in one country at one point in time, so its results speak only about that group3. It should not be taken as a standard for Vietnamese people, for students today, or for any other group without research on that group itself.
From single surveys to a standardised scale: the 1983 turning point
The difference between a 1936-style survey and a 1983-style scale lies in the scale being built through factor analysis1. Instead of each study writing its own questions, the PBS offers a fixed set of items already checked for whether they cluster into meaningful groups1. This lets results from different studies be compared, at least within the range of using the same scale.
The 1983 version also raised the question of links between belief dimensions and personality1. This is a different direction from the question "how many people believe". It is not aimed at measuring population rates, but at seeing which group of phenomena beliefs come with which psychological traits. So one scale can serve two very different kinds of research: prevalence studies and studies of correlation with personality.
What did the 2004 revision change?
Tobacyk's official 2004 print in the International Journal of Transpersonal Studies is a revision of the original scale2. It keeps the multi-subscale structure and states the seven subscales as in the table above2. In terms of use, the revision is the version later studies cite when they say they use the RPBS2.
One point on source scope deserves note: both the 1983 and 2004 papers are peer-reviewed articles, level A evidence in Nhat Nguyet's classification, and both belong to the group of documents that build and print the scale12. What they prove is what the scale contains, how it was built, and how it is used. They do not prove that the phenomena the scale asks about are real or unreal. Those are two entirely different questions, and using a scale says nothing about the answer to the second.
When an article cites "paranormal believers", how should the figure be read?
Four questions are worth asking before accepting any figure about paranormal belief:
- Which scale? The original 1983 PBS, the 2004 RPBS, or an entirely different scale?12
- Which subscale? The figure for "precognition" is not the figure for "spiritualism" or "superstition"2.
- Which sample, where, which year? Results on American students in 1936 cannot be extrapolated to Vietnamese people today3.
- How were the original questions translated and reworded? The same item can be understood differently across cultures.
This is why articles on paranormal belief should state the scale and subscale, instead of only saying "paranormal believers" as a single category.
Where evidence is missing: no standardisation for Vietnamese people
Within the sources cited here, no document discusses standardising the PBS or RPBS for Vietnamese people12. That means when reading a figure about "Vietnamese people who believe in the paranormal", one needs to know which scale it comes from, whether it was translated and tested for reliability on Vietnamese people, or whether it is only inferred from results elsewhere.
This is a data gap, not a conclusion that Vietnamese people believe more or less than people elsewhere. To know that, research is needed on Vietnamese people themselves, with a scale checked for fit with the cultural and linguistic context.
A note on real effects produced by belief
One point easily missed when discussing paranormal belief: belief can change behaviour, and behaviour changes outcomes. For example, a person who believes a certain day is auspicious may deliberately schedule important work on that day, prepare more carefully, and so the outcome differs. That differing outcome is real, but it is not evidence that a scale or system "predicted correctly". It is the effect of belief on behaviour.
This distinction matters because it allows respect for both sides: the believer may have a real experience, and the researcher can measure that belief without concluding whether paranormal phenomena exist. The scale does not settle the metaphysical question; it only describes and classifies belief.
Why articles on this topic should cite the scale
When an article says "research shows X percent of people believe in divination", readers have a right to know how that X percent was measured. If the article does not say, the figure cannot be verified, cannot be compared with other research, and is easily misread as a general category of "paranormal belief".
Citing the scale, subscale, sample, and year is how readers judge for themselves whether the figure relates to them. It is also how a methods article is distinguished from an article asserting the nature of the phenomenon. This article is of the first kind: it explains the measuring tool, and does not conclude about the value of belief.
