research paper on five factors

An official website of the United States government

Official websites use .gov A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

• Publications
• Account settings
• Advanced Search
• Journal List

The Five Factor Model of personality structure: an update

Thomas a widiger, cristina crego.

• Author information
• Article notes
• Copyright and License information

Issue date 2019 Oct.

The Five Factor Model (FFM) of general personality structure consists of the five broad domains of neuroticism (or emotional instability vs. stability), extraversion (vs. introversion), openness (or unconventionality), agreeableness (vs. antagonism), and conscientiousness (or constraint vs. disinhibition). Each of these domains includes more specific facets (e.g., gullible vs. cynical, meek vs. aggressive, soft‐hearted vs. callous, and selfless vs. exploitative are within the domain of agreeableness vs. antagonism).

The FFM traces its roots to the lexical paradigm, which rests on the compelling premise that what is of most importance, interest or meaning to persons when describing themselves and others will be encoded within the language. Fundamental domains of personality emerge as persons develop more and more words to describe the gradations, variations and nuances of a respective domain. The natural, inherent structure of personality is provided by the empirical relationship among the trait terms, and the structure of the English language has converged well onto the “Big Five”. The Big Five have also been replicated within the German, Czech, Dutch, Filipino, Hebrew, Hungarian, Italian, Korean, Polish, Russian, Spanish and Turkish languages, albeit the replication of neuroticism and openness is not as strong as the replication of the domains of agreeableness, extraversion and conscientiousness 1 .

Empirical support for the FFM as a structural model of personality is substantial, including multivariate behavior genetics, childhood antecedents, temporal stability across the lifespan, cognitive neuroscience coordination, and cross‐cultural replication 1 . The FFM has also been shown across a vast empirical literature to be useful in predicting a substantial number of important life outcomes, both positive and negative 2 . Cuijpers et al 3 compared the economic costs of FFM neuroticism (health service uptake in primary and secondary mental health care, out‐of‐pocket costs, and production losses) with the costs associated with common mental disorders (e.g., mood, anxiety, substance use, and somatic disorders). The economic costs of neuroticism were approximately 2.5 times higher than those of the common mental disorders.

Given that the Big Five account for virtually every trait term within the language, it is not surprising that the FFM accounts for every maladaptive personality trait, including those that define the personality disorder syndromes of the ICD and the DSM 1 . The dimensional trait models included within the DSM‐5 Section III and the ICD‐11 are aligned explicitly with the FFM. The FFM also provides the temperament base and personality foundation for the widely cited Hierarchical Taxonomy of Psychopathology 4 , a dimensional structural model that covers much of all forms of psychopathology.

The ICD and DSM personality disorders are readily understood as maladaptive variants of the FFM, but this does not suggest that any measure of the FFM will fully account for every personality disorder. Most existing measures of the FFM do not assess for all of its maladaptive variants and therefore will not be able to account for all of the components and correlates of a respective personality disorder. For example, there are maladaptive variants for all ten poles of all five FFM domains, but existing measures typically fail to assess for the maladaptive variants of conscientiousness (e.g., compulsivity), openness (e.g., magical thinking), agreeableness (e.g., subservience), low neuroticism (e.g., fearlessness), and extraversion (e.g., dominance), thereby limiting the ability to cover traits central to the obsessive‐compulsive, schizotypal, dependent, and psychopathic personality disorders, respectively. The obsessive‐compulsive personality disorder is defined largely by maladaptive conscientiousness (e.g., perfectionism, compulsivity, workaholism, and ruminative deliberation), but most measures of FFM conscientiousness do not assess for these maladaptive variants. Measures to assess maladaptive FFM traits, though, have been developed, including the Five Factor Model Personality Disorder scales 5 , the Personality Inventory for DSM‐5 6 , and the Personality Inventory for ICD‐11 7 .

There are a number of advantages in conceptualizing the ICD and DSM personality disorders from the perspective of the FFM. Many of the ICD and DSM personality disorder syndromes have limited research interest and inadequate empirical support. The FFM brings to the personality disorders a substantial body of construct validation, including a resolution of such notable controversies as gender bias, excessive diagnostic overlap, and temporal instability. An understanding of the etiology, pathology and treatment of the personality disorders has been hindered substantially by the heterogeneity within and the overlap across the diagnostic categories. The American Psychiatric Association has been publishing treatment guidelines for every disorder within the DSM, but guidelines have been provided for only one of the ten personality disorders (i.e., borderline). The complex heterogeneity of the categorical syndromes complicates considerably the ability to develop an explicit, uniform treatment protocol. The domains of the FFM are considerably more homogeneous and distinct, lending themselves well for more distinct models of etiology, pathology and treatment 8 . Empirically validated treatment protocols have already been developed for FFM neuroticism 9 .

A common concern regarding the FFM and any other dimensional trait model is that clinicians will be unfamiliar with this approach and will find it difficult to apply. However, the FFM organization is consistent with the manner in which persons naturally think of personality trait description. Persons who apply the FFM typically find it quite easy to use. There have in fact been a number of studies concerning the clinical utility of the FFM in comparison to the DSM syndromes. A few of these studies have favored the DSM syndromes but, when the methodological limitations of these particular studies were addressed in subsequent studies, the results consistently favored the FFM 8 . Experienced clinicians prefer the FFM and dimensional trait models for the conceptualization of personality disorders 8 .

In sum, the FFM is the predominant model of general personality structure and offers the opportunity for a truly integrative understanding of personality structure across the fields of clinical psychiatry and basic personality science. The ICD and DSM models for the classification and diagnosis of personality disorder are shifting toward the FFM because of its empirical validation and clinical utility.

• 1. Widiger TA. (ed). The Oxford handbook of the five‐factor model. New York: Oxford University Press, 2017. [ Google Scholar ]
• 2. Ozer DJ, Benet‐Martinez V. Annu Rev Psychol 2006;57:401‐21. [ DOI ] [ PubMed ] [ Google Scholar ]
• 3. Cuijpers P, Smit F, Penninx BW et al. Arch Gen Psychiatry 2010;67:1086‐93. [ DOI ] [ PubMed ] [ Google Scholar ]
• 4. Krueger RF, Kotov R, Watson D et al. World Psychiatry 2018;17:282‐93. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• 5. Crego C, Oltmanns JR, Widiger TA. Psychol Assess 2018;30:62‐73. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• 6. Krueger RF, Derringer J, Markon KE et al. Psychol Med 2012;42:1879‐90. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• 7. Oltmanns JR, Widiger TA. Psychol Assess 2018;30:154‐69. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ]
• 8. Mullins‐Sweatt SN, Lengel GJ. J Pers 2012;80:1615‐39. [ DOI ] [ PubMed ] [ Google Scholar ]
• 9. Sauer‐Zavala S, Wilner JG, Barlow DH. Personal Disord 2017;8:191‐8. [ DOI ] [ PubMed ] [ Google Scholar ]
• View on publisher site
• PDF (74.2 KB)
• Collections

Similar articles

Cited by other articles, links to ncbi databases.

• Download .nbib .nbib
• Format: AMA APA MLA NLM

Add to Collections

The Five-Factor Model: Emergence of a Taxonomic Model for Personality Psychology Nathan C. Popkins Northwestern University

This paper examines the five-factor model, a tool used for dimensionally studying personality. Aspects of the model given attention include the specific variables in the model, other related models, and clinical applications of the model. The quality of the model is then evaluated based on five criteria: compatibility, originality, application, taxonomy, and universality. Ultimately, it is evident that the model fits most of the criteria for greatness, but falls somewhat short of being a great theory in personality because it is not truly a theory.

The Model One of the long held goals of psychology has been to establish a model that can conveniently describe human personality, and disorders therein, with the intent to use this model in the remedying of personality disorders and improving general understanding of personality. Currently, a handful of models have risen to prominence, and have thus far stood the test of time. Some models are more generally accepted than others. Support for some models seems to come and go in cycles. One of the more prominent models in contemporary psychology is what is known as the five-factor model of personality (Digman, 1990). This theory incorporates five different variables into a conceptual model for describing personality. These five different factors are often referred to as the "Big 5" (Ewen, 1998, p. 140). The five-factor theory is among the newest models developed for the description of personality, and this model shows promise to be among the most practical and applicable models available in the field of personality psychology (Digman, 1990). Thorough critical attention is given to the proposal that the five-factor model is in fact a great theory. The Five Factors As it became evident to many psychologists that, mathematically, combinations of five factors were useful in describing personality, there was a need to clearly define what these factors were. Indeed, this process led to some dissent in the ranks. One dissenter from the five-factor theorists was renowned psychologist H. J. Eysenck. Eysenck felt that, due to overlaps in the five factors and their correlates, in fact a three-factor model was more appropriate and accurate. His theory is called the PEN model (which stand for psychoticism, extroversion, neuroticism) (Eysenck, 1991), or sometimes is even shortened to the two factor E-IN model (extroversion-introversion, neuroticism) (Eysenck, 1991). According to Eysenck, "Factor analysis has improved the situation...but the problem of naming factors is of course still with us" (Eysenck, 1991, p. 775). Many psychologists support Eysenck's PEN model. However, of the major "factor-analytic models...the Big Five dominates the landscape of current psychological research" (Ewen, 1998, p. 141). Through extensive debating and experimenting, there is currently a general consensus in the realms of scholarly psychology as to the identity of the five factors, and their basic interpretations and values to analysis of personality. The five factors are extroversion-introversion, neuroticism, agreeableness, conscientiousness, and openness (Ewen, 1998). Extroversion has long been one of the traits that has appeared in factor-analytic models, and is one of the two traits to appear in both the five-factor model and Eysenck's PEN and E-IN models. Extroversion also is sometimes referred to as social adaptability, though the popularity of this term seems to be waning (Zuckerman, 1991). Extroversion is defined as "a trait characterized by a keen interest in other people and external events, and venturing forth with confidence into the unknown" (Ewen, 1998, p. 289). Neuroticism is the other trait to play a role in most of the contemporary factor models for personality. In some studies, adjustment is examined as a factor, instead of neuroticism. In this case, higher scores will indicate a positive result, consistent with the other four factors. This is because the term neuroticism has an inherent negative denotation (Bradshaw, 1997). The bases of neuroticism are levels of anxiety and volatility. Within these bounds, neuroticism is "a dimension of personality defined by stability and low anxiety at one end as opposed to instability and high anxiety at the other end" (Pervin, 1989, p. G-7). Openness, agreeableness, and conscientiousness are all terms with which most people outside the realm of psychology are familiar. In general, openness refers to how willing people are to make adjustments in notions and activities in accordance with new ideas or situations. Agreeableness measures how compatible people are with other people, or basically how able they are to get along with others. Conscientiousness refers to how much a person considers others when making decisions. As with the two factors in the big five from Eysenck's E-IN, these three are also placed on sliding scales. These three scales, like neuroticism and extroversion, slide between their limits to give a clear picture of personality. The limits of these scales give a clear idea of their applications and are defined as "trusting and helpful versus suspicious and uncooperative (agreeableness), hard working and reliable versus lazy and careless (conscientiousness), and nonconformist and creative versus conventional and down-to-earth (openness)" (Ewen, 1998, p. 140). Clinical Use of the Five Factors With the five factors themselves more or less firmly established, the model is of little importance to psychology if it can not somehow be applied. Clinically, the five-factor model works much like Freud's psychoanalysis. People who have a certain characteristic that falls at an extreme on the chart of one or more of the five variables are likely to have some sort of psychological abnormality associated with that trait. People are likely to select their environment in such a way that this trait is perpetuated. To keep this cycle from iterating, psychologists make their patients come to terms with the flawed trait, allowing the patient to break the cycle (Pervin, 1989, p. 318). The five-factor model is perhaps even more useful in research and learning than it currently is for psychological patients. To this end, supporters of the five-factor model point to the fact that factor analysis "serves as a useful bridge between the more clinical theories...and the learning, behavioral theories" (Pervin, 1989, p. 326). Determining the Greatness of the Theory To determine the greatness of a psychological theory, it is necessary to examine several key factors. First, to truly be labeled a great theory, it should be established that the theory does not contradict currently held or accepted theories. Also, once this is established, a great theory would include some kind of taxonomy. In addition, a truly great theory would be applicable. In other words, the theory would have some sort of clinical uses and would be able to predict experimental results without contradiction and with a relatively high level of consistency. Also, a great theory requires a ring of originality. It is not a great deed by anyone's standards to simply restate a long existing belief, or to remold it slightly and call it new. Finally, a great personality theory also needs to be universal. Cultural or situationally dependent models, although certainly of some relevance, are not worthy of being called great. Compatibility To establish the five-factor model's compatibility with other models, it is necessary to examine other popular or long held psychological theories. First and foremost, it is easy to see that the five-factor model is consistent with other factor-analytical models. As mentioned before, Eysenck developed his PEN model from Cattell's sixteen-factor model. In this same vein of factor-analytic models, it is now proposed that "the structure of personality is best conceptualized as consisting of five major traits, rather than Eysenck's three types" (McMartin, 1995, p. 138). Still, it seems natural that any factor analytic model should not directly contradict other factor analytic models. It has been established in recent years that the five-factor model is quite compatible with other popular psychological theories as well, not just those concerned with factor analysis. It has been a recent endeavor of some psychologists to build bridges between the five-factor model and Freud's theories of psychoanalysis. For example, recent experiments have shown correlates between Freud's concept of the ego (and resiliency or control thereof) and the five variables in the five-factor model. As stated in the report of one such experiment conducted by Huey and Weisz (1997), focusing on ego resiliency and ego control in adolescent boys, "Ego resiliency seems to reflect, in part, the well adjusted pole of each FFM (five-factor model) dimension, whereas Ego undercontrol...reflects high extroversion, low agreeableness, and low conscientiousness" (p. 412). The correlation between elements of Freud's theory of psychoanalysis (known for its application, but known as much for its lack of taxonomy) and elements of the five-factor model (among the more quantifiable theories in psychology) helps demonstrate the compatibility of the five-factor model. Taxonomy With the five factors placed on sliding scales, it becomes only an exercise in persistence (through experimentation, survey, and interview) to associate various human characteristics with one or more of the five factors. Strong correlates quickly become evident, and as trends develop it becomes possible to assign very accurate descriptions about people via their placement on the sliding scales associated with the five factors. Certainly, it seems plausible that these results can be used to extrapolate many behavioral characteristics, including such dependent traits as financial success and depth of religious faith. Psychologists are currently working towards this end, attempting to empirically establish what Langston and Sykes (1997) called "a taxonomy of personality" (p. 142). Naturally, this taxonomy has long been the goal of personality psychologists, though it seems that, of late, somewhat of a snag has developed in trying to extrapolate past the very broadest of personality traits (the big five, essentially) (Langston & Sykes, 1997). This problem with the five-factor model is that, although often very categorical and taxonomic, it does not delve deeply into the causation of certain correlates. In fact, some relationships are even somewhat counterintuitive, making extrapolation by common sense difficult. For example, one recent study found that type A behavior, which is characterized by general optimism and ambitiousness (Ewen, 1998), had a low correlation to conscientiousness (Morrison, 1997). Application In accordance with the aforesaid flaw in the second criterion for greatness, the five-factor model runs into the most trouble when approached through the standards set forth by the third criterion. This criterion calls for a great theory to be applicable. McAdams (1992) states, "Personality theories do more than specify traits" (p. 336). In essence, what he is getting at is that the five-factor model (although it provides an excellent basis for the description of much of what, in psychology, falls into the realm of personality study) falls short of attaining or ever having a chance to attain the title of the unified psychological theory (McAdams, 1992). As stated, the five-factor model is used effectively in application. Because of this, it can be said that it is applicable, even though this does not address McAdams' claim that the five-factor theory should not really be considered a theory, but more appropriately just a list of five variables that are useful when attempting to identify and classify personality traits. In general, though, most of the applications of the five-factor model as it now exists seem to come in the academic and experimental forum. In reference to the model's usefulness in academics, Digman (1990) says, "At a minimum, research on the five-factor model has given us a useful set of very broad dimensions that characterize individual differences. These dimensions can be measured with high reliability and impressive validity" (p. 436). In summary, Digman states that the five variables that compose the five-factor model "provide a good answer to the question of personality structure" (Digman, 1990, p. 436). Originality It would be possible to argue that the five-factor model does not meet the criterion of originality. Indeed, when Fiske first derived a model for predicting behavior that consisted of five factors, Cattell's sixteen-factor system was already nearly five years old (Digman, 1990). Despite this, the five-factor model could arguably be considered distant enough from Cattell's model to warrant the label of originality. Perhaps instead, though, it is more useful to think of the factor models together, as a single set of theories that, although very different individually, all fit into one family. This family of models, starting with Cattell's sixteen-factor system, and including Eysenck's model and the five-factor model, grew from a study of language initiated by German psychologists Allport and Odbert (Digman, 1990), who were inspired for their studies by two other German psychologists, Baumgarten and Klages (Digman, 1990). Through this family history of the factor models, it is easy to see that they came to exist more or less independently of any other major psychological theories. Universality The fulfillment of the fourth criterion listed for a great theory in personality is that the theory should be nearly universally true. In a non-physical science, like psychology, it is naturally too much to ask that a theory always hold true. The basic premise, though, is that the theory should transcend culture and situation. As the premise is in two parts, so too shall be the response. It appears that the five-factor model holds very well across cultural and linguistic lines. Digman gives three examples of other cultures (and languages) in which the five-factor theory has held up nicely. His three studies took place in Japan, the Philippines, and Germany, and in all three cases, a five-factor solution was clearly evident at the end of testing (Digman, 1990). On the subject of the ability of the five-factor model to cross cultural and linguistic barriers, Digman (1990) writes, "something quite fundamental is involved here. Is this the way people everywhere construe personality, regardless of language or culture?" (p. 434). A current fault of the five-factor model seems to be that it fails to anticipate behavior in many situations. The same virtue that allows the five-factor model to hold true across cultural boundaries is its fault in specific situations: The five variables are too broad (McAdams, 1992). McAdams accurately sums up this problem by saying, "Because the Big Five operate at such a general level of analysis, trait scores...may not be especially useful in the prediction of specific behavior in particular situations" (McAdams, 1992, p. 338). Conclusion Five criteria were established to test the hypothesis that the five-factor model was a great psychological theory. The first criterion for greatness was that the theory should be compatible with other major psychological theories. On this point, the five-factor model was shown to be not only compatible, but even complementary to other theories, including a theory that would seemingly be at the other end of the spectrum from the five-factor model, Freud's highly subjective theory of psychoanalysis. Second, it was suggested that a great theory should be empirical. Although the five-factor model leaves much to be desired as far as the explanation of the numbers, it was shown that with the sliding scales associated with each of the five variables, the five-factor model was easily quantifiable. The five-factor model does have some real problems when scrutinized for its theoretical qualities. It has been suggested that the five-factor model was not so much a theory, but rather, just an idea or a means of classification. This is certainly a stumbling block for the five-factor model's chances at greatness. Also, whereas the five-factor model passed the test of originality with flying colors, when examined to see if it held true universally (or as close as a psychological theory can get) another flaw appeared. The broad taxonomy that is so elemental to the model makes it difficult to specifically anticipate behavior in many situations. "From the standpoint of a multifaceted personology, the five-factor model is one important model in personality studies, not the integrative model of personality," says McAdams (1992, p. 355) in the conclusion to his critique of the five-factor model. This model does not meet all the criteria for greatness, and though it seems to have great potential, it is, as McAdams suggests, not the final all-encompassing theory of psychology. This is a high mark to hit, but one to which, when models like the five-factor are accompanied with theories like Freud's psychoanalysis, contemporary psychology is getting closer. As this model does not quite live up to the standards for a great theory in personality (it would be tough to find a theory that does) perhaps a more appropriate name for it would be a great taxonomy in personality.

Peer Commentary Issues Descriptive and Evaluative G. Scott Acton Northwestern University In "The Five-Factor Model: Emergence of a Taxonomic Model for Personality Psychology," Popkins summarizes and evaluates a factor analytic description of personality that has risen to prominence in the last decade. However, there are problems with both Popkins' summary and his evaluation that this commentary will bring to light. Perhaps Popkins is well within the scope of mainstream psychology when he discusses the five-factor model as a single, unitary development. However, proponents of the five-factor model prefer to make more fine-grained distinctions. Two related, yet distinct, five-factor models can be distinguished: the OCEAN model of Costa and McCrae, and the big five of Goldberg and colleagues (Saucier & Goldberg, 1996). Costa and McCrae's OCEAN model is known by its acronym for the five factors: openness, conscientiousness, extraversion, agreeableness, and neuroticism. This naming of factors has the advantage of making them memorable to newcomers (through the acronym OCEAN), but the disadvantage that the order is not meaningful, and the further disadvantage that the factors are not all named according to the (evaluatively) positive pole of each axis. In the big five variant, Goldberg prefers to name the factors according to their positive poles, and in the order derived from the amount of variance each explains (determined through factor analysis), as follows: surgency, agreeablenss, conscientiousness, emotional stability, and intellect. Currently, debates rage over whether the fifth factor is a '60's style openness to experience construct, or whether it represents a (perhaps broader) general intellect factor--and if the latter, then how intellect (self- or peer-rated) differs from general intelligence (based on I.Q. tests). A further difference between the OCEAN and big five models is that the former is hierarchical whereas the latter is circular. The big five are circular because many variables (usually adjectives) have large loadings on two factors (i.e., large correlations with two factors), as opposed to only one, or three or more. Because the big five are circular, they have swallowed up both the interpersonal circumplex and the affective circumplex, incorporating these into a broader taxonomy (Hofstee, de Raad, & Goldberg, 1992; Saucier, 1992). (This is not to suggest that the big five taxonomy has swallowed up the whole of interpersonal theory, however, which is more than a taxonomy of variables, but rather includes ideas about how people interact.) Another important difference between the OCEAN model and the big five is the theoretical basis that exists for the latter, but not for the former--namely, the lexical hypothesis. This hypothesis contends that those individual differences that are most salient and socially relevant will come to be encoded as terms in the natural language. This conjecture makes into a meaningful activity the culling of all personality-relevant terms from an unabridged dictionary, and the subjecting of self- and peer-ratings on these terms to factor analysis. Having now discussed the descriptive errors of omission in Popkins' paper, I will go on to point out some evaluative errors of commission. Popkins concludes that the five-factor model is not a great theory because it is not a theory, but only a taxonomy. In corroboration of Popkins' thesis, two leading big five researchers go so far as to state, "we would ban the use of the term 'theoretical'...in favor of more exact terms such as 'premises,' 'assumptions,' and 'hypotheses,' as well as more meaningful distinctions such as 'broad versus narrow' and 'phenotypic versus genotypic' attributes" (Saucier & Goldberg, 1996, p. 22). I challenge Popkins to specify what is the difference between a theory and a taxonomy. The difference is not so obvious as it might appear at first blush. I therefore prefer the more neutral term "research program" (Lakatos, 1970), because research can be done to challenge and improve both theories and taxonomies. Popkins' metatheory dictates that theories should be compatible. This idea, however, is incompatible with the history of science. Instead, scientific progress tends to come about when rival theories clash, and when there are many critics on every side. Take the example of the Copernican revolution. Galileo insisted that Copernicus' heliocentric theory of the solar system be taken as literally true, rather than as merely a possible alternative to Ptolemy's geocentric theory. His insistence won him the retribution of the Church of that time, but the veneration of generations of scientists ever since. Moreover, this change from one contradictory theory to another is generally thought of as a paradigmatic case of scientific progress. Back when Copernicus was insisting that his theory and Ptolemy's were mutually consistent, progress languished. Similarly, I believe that the science of personality will progress most when various theories are found to be mutually contradictory, and the results of scientific tests are used to determine which is superior. This idea of progress through contradiction can be applied to models of personality. Perhaps the OCEAN and big five models are not all that compatible after all. Certainly one substantive area on which they disagree is the interpretation of the fifth factor--is it intellect or openness? If these models contradict each other, then they almost surely contradict such alternative models as Eysenck and company's psychoticism-extraversion-neuroticism (PEN) model (Eysenck, 1991) and Ainsworth and company's attachment styles (Shaver & Brennan, 1992). Theoretical elaboration on the reasons for these divergences would likely engender progress in the science of personality, just as in other sciences.

Peer Commentary Five Factors, But Factors of What? Ethan R. Plaut Northwestern University Popkins' essay on five-factor theories of personality is a relatively good introduction to the basic idea, but falls short of being a good survey and analysis of the related literature. He describes one popular set of five factors, the "OCEAN" model, and also discusses some common criticisms of this and factor theories in general. He then "evaluates the theory's greatness" including addressing the issue of its status as a theory or as only a taxonomy. First, I have a few very specific comments. It is stated in the essay that, "Clinically, the five-factor model works much like Freud's psychoanalysis...." This, I think, is a huge error. To make this claim simply on the grounds that both find a problem and then help the patient to deal with it is a logical leap that I cannot make. Psychoanalysis involves many things that are wholly neglected by any factor model, and recognizes the infinite complexity of human character rather than trying to analyze it along a (small) finite number of dimensions. Psychoanalysis looks for the root and exposes it, whereas five-factor treatment could only involve the five surface traits described in the model. In any case, the two are by no means clinically similar. As for the question of theory versus taxonomy, there is no battle to be fought. A taxonomy is simply a specific genre of theory. Psychoanalysis even involves an element of taxonomy in its unconscious-preconscious-conscious model, among other places, and I think that there really is no disputing whether or not psychoanalysis is a "theory." As to the question of whether or not a taxonomy in and of itself could ever be considered a great theory of personality, that is a separate issue. Also, the PEN model, considered to be the biological model, is not properly addressed in this paper, nor is the issue of the numerous different five-factor models. Beyond the simple factual errors and elements of related literature that are neglected, I find something wrong with the general approach of this paper. Primarily, I feel that the aspect of the underlying research is important in evaluating this theory, and this has been almost completely neglected. Only at the end of the paper is the issue of the research mentioned, and then only tangentially and with no real explanation in the section devoted to the universality of the theory. This is a huge error, to say the least, for a paper intended as an introduction to this theory. The theory was linguistically derived by factor analysis of personality-descriptive terms in natural language, using unabridged dictionaries. The study was done in English, then replicated in other languages, sometimes successfully and sometimes not. This is probably the single most interesting aspect of the five-factor model, and is very helpful in putting this particular theory in the context of other theories. First of all, the method of research on the five-factor model is subjective. This in itself makes the theory scientifically at least a bit questionable. Who is to say what words mean? Dictionary definitions say very little of what a word means in context, and nobody, including psychology researchers, or even linguists for that matter, can provide a complete, objective definition of each of these words that takes connotation into account. Even if such a monstrous task were to be mistakenly undertaken, the data set would become impossibly complicated. Describing words with words is hard enough; quantifying them with numbers is kidding oneself. Beyond the questionable basis of the research, what do the findings really describe? The fact that the study has not been perfectly reproducible in foreign tongues proves what I think should have been a predictable outcome from the start: the theory is a cultural one, not a natural one. It is very interesting in itself to study the nature of how we speak about personality in this culture, and I think it could reveal some things directly and indirectly about our personalities (what kind of people talk about neuroticism?). It would be even more interesting to study the similarities and differences that the research would show in doing many cross-cultural cross-linguistic studies. So what does the theory say as it is? It says what we as a culture think it is important and appropriate to talk about in regards to personality. What does this give us? It gives us a way to see how we analyze and discuss each other, and how we might be discussed. It points out characteristics that we see as important. It is, however, very possible that we are not good judges of what is important. It is possible that we know what is important, but that (for reasons of politeness or otherwise) we do not speak of it. For these reasons, I think that the theory fails to say definitively what characteristics are important.

Peer Commentary Contradictions Lead to Disbelief Timothy Tasker Northwestern University Popkins has put forth a compelling argument for the idea of the five-factor model. The only problem becomes what is the five-factor model? This representation of the five-factor model is so riddled with contradiction and confusion that one feels more befuddled after reading this article than before one started. Specifically, Popkins begins this paper with an abstract that states "[The theory] falls somewhat short of being a great theory in personality because it is not truly a theory." Then, under two paragraphs later, he states, "Thorough critical attention is given to the proposal that the five-factor model is in fact a great theory." Which statement are we to believe? In one Popkins says it is not a theory, but in another Popkins will prove that it is not only a theory, but a great one at that. Further, I found that there was a reference to Cattell more than half way through the article under the heading of Compatibility. "As mentioned before, Eysenck developed his PEN model from Cattell's sixteen-factor model." However, upon closer inspection I discovered that this is the first occurrence of Cattell's name in the whole article. How can we rely on information that is either missing or does not exist? An issue with which I have a more substantial argument is the choice of wording. At times, Popkins argues that it is not a theory, yet refers to the idea as the five-factor theory. Essentially, the use of model, theory, and idea are used interchangeably. I believe that there needs to be an important distinction made among these words because they are entirely separate ideas. These issues need to be addressed before one can even think about rating or commenting upon the value of this "theory." Accordingly, there is a great problem with evaluating a theory as to its greatness. I respect the creativity that was used in outlining a set of criteria to define greatness, but greatness is a highly subjective word. No attention is paid to this subjectivity. The reader is told that this is the definition of greatness and this is what the author is going to use for his analysis of the theory. There are no citations as to whose idea of greatness this is. Faults notwithstanding, I believe that this article is of high quality in that it breaks apart a relatively modern "theory of personality" into smaller pieces and evaluates those. Popkins makes an effective argument for the clinical use of the five-factor model and covers the controversy of its clinical usage in an efficient manner. The concept of this article that I find most effective is its strategic attempt at proving the five-factor model a great "theory" (or whatever it is...).

Peer Commentary The Big Five: Taxonomy of Trait Theory Barbara M. Trzop Northwestern University Advocates of the "big five" often field criticisms that accuse the taxonomy as not living up to the standards of a "true" theory. Popkins points this out in his paper; still, he continues to assess its validity as a "great theory" in personality, and repeatedly uses the words "theory" and "model" as if they were interchangeable. However, the purpose of a model is to illustrate clearly the concepts that underlie a theory. Although I acknowledge the differing opinions in this matter, I support the idea that the big five taxonomy is indeed a model of a much broader theory that continues to be an important influence in personality research: trait theory. Personality theorists can be classified within two different schools of thought: situational theorists and trait theorists (or some combination of the two). There are those who, along with social psychologists and behaviorists, believe that environmental factors are key determinants of personality. These "situationists" declare that specific environments reinforce and influence behavior, ultimately shaping one's personality. Individual differences are explained away by attributing them to "error variance" (Burger, 1997, pp. 194-206), and responses to a given situation are sometimes explained as species-typical reflexes. However, it can be seen both in and out of the laboratory that not everyone reacts the same in similar situations. Trait theorists sharply contradict situational theorists by offering an explanation for the "error variance" found between subjects experiencing the same situation. In fact, these individual differences are supported by the existence of traits, which remain stable across numerous situations, and which influence individuals to act differently when experiencing similar situations. However, individual differences can be quite messy to deal with during an experimental situation. Questions of parsimony arise: How many traits are there? How big are the differences between people? How can one ever classify them all? Answers to these questions are addressed nicely by Popkins, in a brief history of the development of the big five. Apparently, these questions were quite relevant, and numerous theorists set off to find their answers. In order to get around the problem of semantics and duplicate meanings, factor analyses and meta-analyses were employed to shave down and coagulate the mess. After much consideration, the big five model was pieced together from existing theories, and offered a concise description of five traits that attempted to encompass the vast realms of the human personality. Indeed, it illustrated trait theory quite nicely, as it suggested very simply that different combinations and gradations of these traits among individuals could influence the way individuals act in any given situation. One problem with the big five lies in the fact that it is merely a descriptive taxonomy, lumping together similar behaviors under a common label (Revelle, 1995). Other trait theorists take another route, and attempt to link genetic or anatomical structures to traits, offering putative explanations for why people have differing traits, how they develop, and how they actually influence behavior. Given that the big five lacks this, its usefulness in a clinical situation is not always trusted. Perhaps linking the big five traits (such as neuroticism) to biological substrates may lend more insight into the treatment of disorders (such as depression). However, assigning traits and behaviors to specific biological causes is more difficult than it appears, and much of the data resulting from these attempts appear to be more correlative than causal. Despite problems, descriptive trait taxonomies can be useful in other situations, such as assessing potential employees. The descriptive nature of the big five offers exactly what employers need to choose between similar applications and resumes (Burger, 1997, pp. 194-206). Given that the purpose of the big five is to provide correlational evidence between traits and specific behaviors, it may aid an employer to predict which of her applicants is most likely to behave favorably in a given situation. In this way, the big five model serves trait theory quite well, and it is my opinion that this method is much better than guessing. The big five model is an attempt to organize and to quantify traits, which make up the foundation of trait theory. As a descriptive taxonomy, it helps to support trait theory as a "great idea in personality." However, on its own it lacks the causal explanations that it needs in order for it to be considered a theory in its own right. Nevertheless, it provides a foundation for present and future trait theorists to build on, so that these "great ideas" can be formulated into one great theory.

Author Response Semantic and Theoretical Clarifications Nathan C. Popkins Northwestern University Most of the responders to this paper had similar questions or qualms. I feel I can clear up most of these with a few corrections or restatements, whereas others require more specific attention. Generally, many of the complaints arose from the use of the terms "theory," "taxonomy," and "model" somewhat interchangeably, as pointed out by both Tasker and Trzop . First of all, the terms "taxonomy" and "model" are both applicable to the big five, and are therefore correctly used as synonyms throughout this paper. The distinction between these two terms and the word "theory," within the context of this paper, is the following: The term "theory" is used at times referring to the evaluation of the big five as a great theory in psychology, whereas the other two terms can be used to describe the big five in any context. This brings me to another very important semantic point. In his response, Tasker questions two specific instances of where my paper appears to harbor interior contradictions. In the first of these, he questions the two phrases, "[The theory] falls somewhat short of being a great theory in personality because it is not truly a theory," and, "Thorough critical attention is given to the proposal that the five-factor model is in fact a great theory." The first statement is an evaluative conclusion. The second is a proposal, much like a hypothesis, that is eventually debunked (essentially, because of what is said in the first phrase). The second semantic error Tasker points out concerns the statement, "As mentioned before, Eysenck developed his PEN model from Cattell's sixteen-factor model." I erroneously inserted the, "As mentioned before...," in reference to Eysenck's PEN model, not to the entire statement, though upon rereading this, the statement clearly makes no sense as it stands. The best (and simplest) alteration that could be made to fix the error is the simple omission of the, "As mentioned before...." Two of the other major problems responders seem to have with the paper revolved around the discussion of compatibility. First of all, Acton disagrees with me on the point that a good theory (model, whatever) should be compatible with existing theories. He uses the classic example of the two theories of Ptolemy and Copernicus to illustrate his point. Unfortunately, I think his argument is based around a dated example that is too archaic to still be valid. The story of Copernicus, Ptolemy, and Galileo unfolded at a time when science was young, and there was much dissent (as his example shows) as to the nature of even the most basic facts of science. Although I will not dispute the fact that psychology is a relatively young science, we are no longer at the point at which we believe the heart controls emotions or any such things. It is my belief, anyhow, that no longer will there be the grand, romantic, Newtonesque epiphanies of Galileo's time; science has come too far for that (even psychology). Plaut also questions my claim that the five-factor model is used similarly to psychoanalysis. Although I will stick to this statement, it requires more explanation. In psychoanalysis, a psychologist uses psychoanalytic theory to find and remedy the problem. The five-factor model is good at locating the problem, but still leaves much of the cure up to the psychologist. Lastly, I would like to thank the four published commentators and all the peer reviewers (whose comments on this paper are not published here on-line). The system of constructive examination in psychology is the factor most responsible for the rapid improvement in the field over the last century.

References Bradshaw, S. D. (1997). Impression management and the NEO five factor inventory: Cause for concern? Psychological Reports, 80, 832-834. Burger, J. M. (1997). Personality (4th ed.). Pacific Grove, CA: Brooks/Cole. Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology, 41, 417-440. Ewen, R. B. (1998). Personality: A topical approach. Mahweh, NJ: Erlbaum. Eysenck, H. J. (1991). Dimensions of personality: 16, 5, or 3?--Criteria for a taxonomic paradigm. Personality and Individual Differences, 12, 773-790. Hofstee, W. K. B., de Raad, B., & Goldberg, L. R. (1992). Integration of the big five and circumplex approaches to trait structure. Journal of Personality and Social Psychology, 63, 146-163. Huey, S. J., & Weisz, J. R. (1997). Ego control, ego resiliency, and the five-factor model as predictors of behavioral and emotional problems in clinic-referred children and adolescents. Journal of Abnormal Psychology, 106, 404-415. Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge (pp. 91-196). New York: Cambridge University Press. Langston, C. A., & Sykes, W. E. (1997). Beliefs and the big five. Journal of Research in Personality, 31, 142-165. McAdams, D. P. (1992). The five factor model in personality: A critical appraisal. Journal of Personality, 60, 329-361. McMartin, J. (1995). Personality psychology: A student centered approach. Thousand Oaks, CA: Sage. Morrison, K. (1997). Personality correlates of the five-factor model for a sample of business owners/managers: Associations with scores on self-monitoring, type A behavior, locus of control, and subjective well being. Psychological Reports, 80, 255-272. Pervin, L. A. (1989). Personality: Theory and research (5th ed.). New York: Wiley. Revelle, W. (1995). Personality processes. Annual Review of Psychology, 46, 295-328. Saucier, G. (1992). Benchmarks: Integrating affective and interpersonal circles with the big-five personality factors. Journal of Personality and Social Psychology, 62, 1025-1035. Saucier, G., & Goldberg, L. R. (1996). The language of personality: Lexical perspectives on the five-factor model. In J. S. Wiggins (Ed.), The five-factor model of personality: Theoretical perspectives (pp. 21-50). New York: Guilford. Shaver, P. R., & Brennan, K. A. (1992). Attachment styles and the "big five" personality traits: Their connections with each other and with romantic relationship outcomes. Personality and Social Psychology Bulletin, 18, 536-545. Zuckerman, M. (1991). Psychobiology of personality. New York: Cambridge University Press.