Measuring the Big Five Personality Factors

I occasionally get questions about how to measure the Big Five personality factors, probably because I co-wrote a Handbook of Personality chapter on the Big Five with Oliver John. I have created this page to address a few of the more common nuts-and-bolts questions I get about measuring the Big Five. I have written this page in a fairly informal style, and I have not attempted to be comprehensive. For a fuller treatment of measurement and theoretical issues, I recommend that you look at the chapter.

(Note: This page was written to help researchers who want to measure the Big Five as part of their research programs. If you came across this page out of curiosity about your own personality, you can fill out a free questionnaire and get instant, personalized feedback at www.outofservice.com. Two longer versions, offering more detailed feedback, are available from John Johnson's webpage.)

Contents:

What are the Big Five?

The Big Five are five broad factors (dimensions) of personality traits. They are:

As should be evident, each of the Big Five factors covers a lot of ground and consists of a variety of more specific traits. The Big Five structure were derived from statistical analyses of which traits tend to co-occur in the population, but the underlying correlations are probabilistic and thus exceptions are possible. For example, talkativeness and assertiveness are both traits associated with Extraversion, but they do not go together by logical necessity: you could imagine somebody that is assertive but not talkative (the "strong, silent type"). However, numerous studies indicate that people who are assertive are typically also talkative (and vice versa), which is why they go together under the broader Extraversion factor.

For this reason, you should be clear about your goals when choosing your measures. If you expect that you might need to make finer distinctions (such as between talkativeness and assertiveness), a broad-level Big Five instrument might not be enough. You could use one of the longer inventories that make facet-level distinctions (like the NEO PI-R or the IPIP scales - see below), or you could supplement a shorter inventory (like the Big Five Inventory) with additional scales that measure the specific dimensions that you are interested in.

It is also worth noting that there are many aspects of personality that are not subsumed within the Big Five. The phrase personality trait has a special meaning in personality psychology that is narrower than the everyday usage of the term. Motivations, emotions, attitudes, abilities, self-concepts, social roles, autobiographical memories, and life stories are just a few of the other "units" that personality psychologists study. Some of these other units may have theoretical or empirical relationships with the Big Five traits, but they are conceptually distinct.

What is the difference between the terms Big Five, Five-Factor Model, and Five-Factor Theory?

The Big Five are, collectively, a taxonomy of personality traits: a framework for understanding which traits go together. The Big Five are an empirically based phenomenon, not a theory of personality. The Big Five factors were discovered through a statistical procedure called factor analysis, which was used to analyze how various personality traits are correlated in humans. The original derivations relied heavily on American and Western European samples, and researchers are still examining the extent to which the Big Five structure generalizes across cultures.

Some researchers use the term Five-Factor Model instead of "Big Five." In scientific usage, the word "model" can refer either to a descriptive framework of what has been observed, or to a theoretical explanation of causes and consequences; the Five-Factor Model (like the Big Five) is a model in the first sense only. The term "Big Five" was coined by Lew Goldberg and was originally associated with studies of personality traits used in natural language. The term "Five-Factor Model" has been more commonly associated with studies of traits using personality questionnaires. The two research traditions yielded largely consonant models (in fact, this is one of the strengths of the Big Five/Five-Factor Model as a common taxonomy of personality traits), and in current practice the terms are often used interchangeably. A subtle but sometimes important area of disagreement between the lexical and questionnaire approaches is over the definition and interpretation of the fifth factor, called Intellect/Imagination by many lexical researchers and Openness to Experience by many questionnaire researchers. This issue is discussed in the aforementioned chapter.

Five-Factor Theory, formulated by Robert (Jeff) McCrae and Paul Costa (see, for example, their 1999 Handbook of Personality chapter), is an explanatory account of the role of the Big Five factors in personality. Five-Factor Theory includes a number of propositions about the nature, origins, and developmental course of personality traits, and about the relation of traits to many of the other personality variables mentioned earlier. Five-Factor Theory presents a biological account of personality traits, in which learning and experience play little if any part in influencing the Big Five.

Five-Factor Theory is not the only theoretical account of the Big Five. Other personality psychologists have proposed that environmental influences, such as social roles, combine and interact with biological influences in shaping personality traits. For example, Brent Roberts has recently advanced this approach under the name Social Investment Theory.

Finally, it is important to note that the Big Five are used in many areas of psychological research in ways that do not depend on the specific propositions of any one theory. For example, in interpersonal perception studies a researcher may find the Big Five to be a useful model for organizing people's perceptions of one another's personalities; such research does not always depend on any particular theoretical account of what traits "really" are. Regardless of whether you endorse any particular theory of personality traits, it is still quite possible that you will benefit from measuring and thinking about the Big Five in your research.

Where can I learn more about the Big Five?

For an introduction to the conceptual and measurement issues surrounding the Big Five personality factors, a good place to start is the Handbook of Personality chapter that I wrote with Oliver John. (Yes, I am self-promoting here. Shoot me.) The chapter covers a number of important issues:

The chapter includes a conceptual and empirical comparison of three measurement instruments: Oliver John's Big Five Inventory (BFI), Paul Costa and Jeff McCrae's NEO Five Factor Inventory (NEO-FFI), and Lew Goldberg's set of 100 trait-descriptive adjectives. There is no one-size-fits-all measure, but the chapter includes our recommendations on which instrument(s) you should use for different applications.

Where do I get the Big Five Inventory (BFI)?

The Big Five Inventory (BFI) is a self-report inventory designed to measure the Big Five dimensions. It is quite brief for a multidimensional personality inventory (44 items total), and consists of short phrases with relatively accessible vocabulary. A copy of the BFI, with scoring instructions, is reprinted in the chapter as an appendix. Oliver John is the copyright holder for the BFI. If after weighing the fair use implications (see below) you decide that you want or need formal permission, you will need to contact Oliver John, not me.

What are other ways of measuring the Big Five?

The BFI is not your only option for measuring the Big Five...

The International Personality Item Pool, developed and maintained by Lew Goldberg, has scales constructed to work as analogs to the commercial NEO PI-R and NEO-FFI scales (see below). IPIP scales are 100% public domain - no permission required, ever.

If you want items that are single adjectives, rather than full sentences (like the NEO) or short phrases (like the BFI and IPIP), you have several options. For starters, there is Lew Goldberg's set of 100 trait-descriptive adjectives (published in Psychological Assessment, 1992). Gerard Saucier reduced this set to 40 Big Five mini-markers that have excellent reliability and validity (Journal of Personality Assessment, 1994). More recently, Saucier has developed new trait marker sets that maximize the orthogonality of the factors (Journal of Research in Personality, 2002). Saucier's mini-markers are in the public domain.

The NEO PI-R is a 240-item inventory developed by Paul Costa and Jeff McCrae. It measures not only the Big Five, but also six "facets" (subordinate dimensions) of each of the Big Five. The NEO PI-R is a commercial product, controlled by a for-profit corporation that expects people to get permission and, in many cases, pay to use it. Costa and McCrae have also created the NEO-FFI, a 60-item truncated version of the NEO PI-R that only measures the five factors. The NEO-FFI is also commercially controlled.

If you need a super-duper-short measure of the Big Five, you can use the Ten Item Personality Inventory, recently developed by Sam Gosling, Jason Rentfrow, and Bill Swann. But read their journal article first (it is on Sam Gosling's web page). There are substantial measurement tradeoffs associated with using such a short instrument, which the article discusses.

Do I need anybody's permission to use these psychological inventories?

Disclaimer: I do not claim to speak for anyone else, I am not a lawyer, don't trust a word I say, do not taunt Happy Fun Ball.

The official answer

Under U.S. copyright law, every written work is automatically copyrighted at the moment of creation. The general rule is that you may not copy and distribute a copyrighted work without permission. However, there are two major exceptions to this rule. The first exception is that if a copyright holder has declared a work to be public domain, then anybody can use it. (This is the case for the International Personality Item Pool as well as for Gerard Saucier's mini-markers). The second exception is the so-called fair use doctrine. If you are using intellectual property in a way that qualifies as fair use, you do not need to get permission to use it. In fact, fair use means you can use something even if the rights-holder loudly and strenuously objects.

Unfortunately, fair use doctrine in U.S. law is pretty fuzzy. Several factors contribute to whether a particular usage is considered fair use. My personal belief is that noncommercial academic research merits protection under fair use doctrine and the First Amendment. However, quoting from Stanford Law's excellent site on fair use: "Unfortunately, the only way to get a definitive answer on whether a particular use is a fair use is to have it resolved in federal court." (chapter 9-b) Caveat scholar.

In addition to general principles of fair use, the American Psychological Association warns that there may be special legal and ethical considerations that apply to psychological tests. A major concern for the APA is that prior exposure to a test may invalidate future responses. For example, this is a significant concern for IQ tests, which are subject to practice effects. Prior exposure is almost certainly less of a problem for self-report personality inventories like Big Five measures. Nevertheless, it is an issue that you should consider when using any psychological test.

The practical answer

On a pragmatic level, I have heard anecdotally that you may be more likely to face objections if you try to use instruments published by for-profit testing corporations than if you use instruments whose rights are held by individual researchers. An objection may not have any merit, but it will be a hassle to deal with anyway, which is probably the point. If you want to use a commercial instrument for academic (noncommercial) purposes, you should either pay for it, get the author's permission, or be prepared to defend your actions as fair use.

By contrast, many individual researchers I have talked to would be delighted to see their measures used (and cited!) by other scientists, even if the researchers have not declared the measures to be in the public domain. If you are conducting academic research, are not making a profit, and you are not conducting a large-scale data collection (such as a national survey or a public Internet study, which might clash with other such efforts), then it is probably safe to go ahead and use a noncommercial measure without formal permission.

How do I cite this page?

Srivastava, S. ([this year]). Measuring the Big Five Personality Factors. Retrieved [today's date] from http://www.uoregon.edu/~sanjay/bigfive.html.

Acknowledgments

Thanks to Lew Goldberg and Jeff McCrae for helpful comments and corrections to this page.