Assignment: Clinical Personality Assessments

Chapter 10

REVISED NEO PERSONALITY INVENTORY

The NEO Personality Inventory (NEO PI; Costa & McCrae, 1985) and the Revised NEO Personality Inventory (NEO PI-R; Costa & McCrae, 1992) measure five broad domains or dimensions of personality in normal adults. Three of these domain scales, measur­ ing Neuroticism (N), Extraversion (E), and Openness to Experience (0), have been re­ searched for years and serve as the basis of the name for the original Inventory (NEO). The NEO PI also includes two additional domains, Agreeableness (A) and Conscientiousness ( C). These five domains allow for a comprehensive description of personality in normal adults. The NEO PI-R consists of five global domains and six facets for each domain (see Table 10.1).

Table 10.2 provides the general information on the NEO PI-R.

HISTORY

A long line of research on five-factor models of personality serve as the basis for the NEO PI-R, most of which is beyond the scope of this Handbook (cf. Wiggins, 1996). The rather common finding in the 1980s of five factors in personality research, served as the major impetus for a multitude of studies based on a lexical analysis of words, personality traits, interpersonal theory, or ratings of schoolchildren's behavior. Despite critiques that five-factor models were atheoretical, they have persisted and gained widespread acceptance in the field of personality research. A significant impetus for this widespread acceptance of five-factor models is the prolific work of Costa and McCrae and their publication of the NEO PI (Costa & McCrae, 1985) and NEO PI-R (Costa & McCrae, 1992). A bibliography (Costa & McCrae, 2003) available on the website for Psychological Assessment Resources (www.parinc.com), the publisher of the NEO PI-R, is nearly 60 pages.

Both the NEO PI (Costa & McCrae, 1985) and the NEO PI-R (Costa & McCrae, 1992) have two forms: Form R (Rater) and Form S (Self). Form R is to be completed by a knowledgeable other who is well acquainted with the person and Form S is to be completed by the person being evaluated. Virtually all the research on the NEO PI and NEO PI-R has been conducted with Form S and it is the main form that will be discussed here. More frequent use of Form R in conjunction with Form S seems well warranted because of the important perspective it can provide on the person being evaluated. At a minimum, the reader needs to be aware of the existence of Form R so as to consider the possibility of its use.

315

316 Self-Report Inventories

Table 10.1 Revised NEO Personality Inventory (NEO PI-R) domain and facet scales

Domain Facets

N (Neuroticism) NJ Anxiety N2 Angry Hostility N3 Depression N4 Self-Consciousness NS Impulsiveness N6 Vulnerability

E (Extraversion) El Warmth E2 Gregariousness E3 Assertiveness E4 Activity ES Excitement-Seeking E6 Positive Emotions

0 (Openness) OJ Fantasy 02 Aesthetics 03 Feelings 04 Actions 05 Ideas 06 Values

A (Agreeableness) Al Trust A2 Straightforwardness A3 Altruism A4 Compliance AS Modesty A6 Tender-Mindedness

C (Conscientiousness) Cl Competence C2 Order CJ Dutifulness C4 Achievement Striving cs Self -Discipline C6 Deliberation

NEO PI (First Edition)

The NEO PI (Costa & McCrae, 1985) consisted of five domains: Neuroticism (N); Ex­ traversion (E); Openness (0); Agreeableness (A); and Conscientiousness (C). The name of the inventory-NEO-was formed from the initial letter of the first three names in a concession to an early version of the inventory that contained only those three domains. These five domains measure the broad dimensions of personality in normal adults. The first three domains (Neuroticism [NJ; Extraversion [E], Openness [0]) also had six facets or subscales for each domain.

Revised NEO Personality Inventory 317

Table 10.2 Revised NEO Personality Inventory (NEO PI-R)

Authors: Published: Edition: Publisher: Website: Age range: Reading level: Administration formats: Languages: Number of items: Response format: Administration time: Primary scales: Additional scales: Hand scoring: General texts: Computer interpretation:

Costa & McRae 1992 Revised Psychological Assessment Resources www.parinc.com 18+ 6th grade Paper/pencil, computer, CD, cassette 9 published and 25 validated translations 240 5-point Likert scale 20-30 minutes 5 Domains and 30 Facets None 2-part carbonless Answer Sheet (self-scoring) None Psychological Assessment Resources (Costa & McRae)

NEO PI-R (Revised Edition)

The NEO PI-R (Costa & McCrae, 1992) consists of the same five domains as in the NEO PI. There are only two minor differences between the NEO PI-Rand the NEO PI. First, the facet scales for Agreeableness (A) and Conscientiousness (C) were added; they had not been available on the NEO PI. Second, 10 (4.2%) items were replaced to allow for more accurate measurement of several facets.

Although the NEO PI-R is the focus of this chapter, two other forms of the NEO need to be mentioned: NEO Five-Factor Inventory (NEO-FFI; Costa & McCrae, 1992); and NEO PI-3 (McCrae, Costa, & Martin, 2005). Each of these other forms of the NEO PI-R is described in turn. This description can be very brief for both of them because they retain the essential features of the NEO PI-R.

NEO Five-Factor Inventory

The NEO-FFI (Costa & McCrae, 1992) is essentially an authorized short form of the NEO PI-R. It consists of 60 items from the NEO PI-R that are used only to score the five domains: Neuroticism (N); Extraversion (£); Openness (0); Agreeableness (A); and Conscientiousness (C). It does not contain the items for assessing the facets within each domain. The NEO-FFI is designed for use in circumstances in which time is too limited to present the entire NEO PI-R or only scores on the five domains are required. All the information provided on the domains for the NEO PI-R will apply to the NEO-FFI so it does not need to be discussed explicitly.

NEOP/-3

McCrae et al. (2002) identified 30 items on the NEO PI-R (Costa & McCrae, 1992) that were not endorsed by at least 2% of nearly 2,000 adolescents. A number of these

318 Self-Report Inventories

30 items contained words that adolescents, and even some adults, might not understand.

An additional 18 items were identified that had item-total scores on the facet scales less

than .30. Alternative items were developed for these 48 items and McCrae et al. (2005) found acceptable replacements for 37 of them. The original version of the other 11 items was retained on the NEO PI-3. The items on the NEO Pl-3 are easier to read than those on the NEO PI-R and the NEO PI-3 can be used for adolescents 12 years of age and older. Further research currently is being conducted to determine whether the NEO Pl-3 can be considered as a replacement for the NEO PI-Rat all ages.

The entire December 2000 issue of the journal Assessment was devoted to the NEO PI-R. Anyone who is using the NEO PI-R should review this issue to get a better idea of the broad extent and wide nature of its usage.

ADMINISTRATION

The first issue in the administration of the NEO PI-R is ensuring that the individual is invested in the process. Taking a few extra minutes to answer any questions the individual has about why the NEO PI-R is being administered and how the results will be used will pay excellent dividends. The examiner should work diligently to make the assessment process a collaborative activity with the individual to obtain the desired information. This issue of therapeutic assessment (Finn, 1996; Fischer, 1994) was covered in more depth in Chapter 2 (pp. 43--44 ). The transparent nature of the items on the NEO PI-R and the lack of extensive

means for assessing the validity of item endorsement ( see later section in this chapter) make the task of getting the individual appropriately engaged in completing the NEO PI-Rall the more important.

Reading level is not a crucial factor in determining whether a person can complete the NEO PI-R. First, the reading level of the NEO PI-R is the sixth grade. Second, the exam­ iner may read the items to individuals whose reading abilities are limited and record the responses (Costa & McCrae, 1992, p. 5). The NEO PI-R is the only self-report inventory

discussed in this Handbook that allows the examiner to read the items to the individu­ als. All other self-report inventories explicitly discourage or forbid this procedure (see Chapter 5).

SCORING

Scoring the NEO PI-R is relatively straightforward either by hand or computer. If the NEO PI-R is administered by computer, the computer automatically scores it. If the individual's responses to the items have been placed on an answer sheet, these responses can be entered into the computer by the clinician for scoring or they can be hand scored. If the clinician enters the item responses into the computer for scoring, they should be double entered so that any data entry errors can be identified.

One of the advantages of computer scoring is that the factor score for each domain is

computed directly. The factor scores can be calculated for the domains using the formulas presented in the Manual (Costa & McCrae, 1992, p. 8), and it is recommended that researchers use the factor scores. "In most cases, the domain scale scores are a good

Revised NEO Personality Inventory 319

approximation to factor scores, and it is probably not worth the effort to apply these formulas by hand to individual cases" (Costa & McCrae, 1992, p. 7).

The NEO PI-R (Costa & McCrae, 1992) and the Personality Assessment Inventory (PAI: Morey, 1991) are the only self-report inventories reviewed in this Handbook that do not use "true/false" items. Both of these inventories have the same publisher (Psychological Assessment Resources), and that may account for not using "true/false" items. The NEO PI-R uses a five-point Likert scale ranging from SD (Strongly Disagree), D (Disagree), N (Neutral), A (Agree), to SA (Strongly Agree). These potential response options always are presented in this same order on the answer sheet. When SD (Strongly Disagree) is the scored direction for a specific item, the response options are scored as 4, 3, 2, 1, or 0. When SA (Strongly Agree) is the scored direction, the preceding five response options are scored as 0, 1, 2, 3, or 4. Thus, the total raw score on each eight-item facet scale can range from 0 to 32. The total score on a domain, each of which consists of six facet scales, can range from Oto 192, but the norm tables for adults are truncated at 25 and 172 (Costa & McCrae, 1992, Appendix C, p. 79).

The first step in hand scoring is to examine the answer sheet carefully and indicate omitted items and double-marked items by drawing a line through all five responses to these items with brightly colored ink. Also, cleaning up the answer sheet is helpful and facilitates scoring. Responses that were changed need to be erased completely if possible, or clearly marked with an "X" so that the clinician is aware that this response has not been endorsed by the client.

The answer sheet for the NEO PI-R is self-scoring, that is, no templates or overlays are required for scoring. Instead the top page of the answer sheet is removed and each row of items corresponds to one of the facets for each of the domains. The facets are in numerical order within each domain and the domains are in the order: Neuroticism (N); Extraversion {£); Openness (0); Agreeableness (A); and Conscientiousness (C). The raw score for each facet is the sum of the circled numbers on its row. The sum of the marked scores for the first row is facet N 1, the sum of the second row is facet El, and so on. Once the six facet scores have been calculated for each domain, they are summed to create the raw score for each domain. Thus, the sum of facets NJ, N2, N3, N4, N5, andN6 becomes the raw score for domain N. These raw scores for each domain are entered into the corresponding box at the bottom of the answer sheet.

Plotting the profile is the next step in the scoring process. There are two profile forms that can be used with Form S: adults (21 years of age and older) and college (17 to 20). Profiles are plotted separately for men and women with each of these forms and are on opposite sides of the same page. The college-age profile form is used for all individuals aged 17 to 20 no matter whether they are in college. To remove the ambiguity, it would be more accurate to say that the "young adult" form should be used for all individuals between the ages of 17 and 20 and not call it a "college" profile form.

Once the correct profile form has been selected for the person's age and gender, all the raw scores from the answer sheet are transferred to the appropriate column of the profile sheet (see Figure 10.1). The first five columns on the profile sheet are the five domains (N, E, 0, A, and C) and then the six facets for each domain are presented in order. The raw score on each domain and facet is indicated by either circling the number or marking it with an "x." Once the individual's scores on the five domains have been plotted, a solid line is drawn to connect them. A similar procedure is followed for each of the six facets.

80

II)

! so~-----.,,,::...-------\----~~-----+-'l------+-l~---\-----~---+----1->.-----____:,,,,.~---4-----+-+---­ o u

(/)

1- 50

N E O C A N1 N2 N3 N4 N5 N6 E1 E2 E3 E4 E5 E6 01 02 03 04 05 06 C1 C2 ca C4 C5 C6 A1 A2 A3 A4 A5 A6 NEO Domain and Facet Scales

Figure 10.1 NEO PI-R profile form for Domain and Facet scales.

Revised NEO Personality Inventory 321

The scores for the domains are not connected to the facet scores, and the sets of facets are plotted separately; that is, there will be seven separate lines or profiles on the form.

ASSESSING VALIDITY

One of the few areas of contention with the NEO PI-R is whether validity scales are necessary at all. The focus of this contention revolves around three issues: (1) whether

responses to the NEO PI-R can be distorted and thus should be assessed; (2) the prevalence of such distortions within various groups of individuals; and (3) whether the use of validity

scales to remove questionable profiles actually improves correlates with external criteria. Each of these issues is examined in tum.

A variety of studies have demonstrated that the NEO PI-R, like all self-report instruments, can be distorted by students in simulation designs either in a positive (Ballenger, Caldwell­ Andrews, & Baer, 2001; Griffin, Hesketh, & Grayson, 2004) or negative direction (Berry et al., 2001).

It seems natural enough that distortions of responses occur less frequently in normal adults, where the NEO PI-R is used most often, because there is little motivation for doing so. The frequency of such distortions of responses also should decrease when the NEO PI-R is filled out anonymously, which typically happens in research. Again, finding that validity scales are not useful in normal adults and research settings would seem to reflect the nature of the participants and settings rather than the usefulness of the validity

scales. However, in clinical and personnel screening settings, it seems probable that individuals

may distort their responses in some manner and the preceding research demonstrates that

scores on the NEO PI-R can be distorted. In both clinical and personnel selection settings, the examiner is concerned with assessing potential distortions to the domain and facet scales in this specific individual, because it will affect the interpretation of the scores. Thus, the finding that validity scales may be more useful in clinical and personnel selection settings would seem to reflect the nature of the setting.

Several studies found that using the validity scales to remove NEO PI-R profiles with excessive distorted responses did not increase the relationship with external correlates (Piedmont, McCrae, Riemann, & Angleitner, 2000; Yang, Bagby, & Ryder, 2000). These

findings typically occur when large groups of participants are assessed and the relative

prevalence of such invalid profiles is relatively low. Several studies also have found that using the validity scales to remove NEO PI-R pro­

files with excessive distorted responses increased the relationship with external correlates (Caldwell-Andrews, Baer, & Berry, 2000; Young & Schinka, 2001). These findings typi­ cally occurred in clinical samples that would be more prone to distort their responses and

in most cases were instructed to do so. Another way of framing this contention is whether response distortion is substance, a

characteristic of the individual such as some form of psychopathology, or personality trait or style, an effortful alteration of responses that may be conscious or reflect lack of insight. In true diplomatic fashion, Morey et al. (2002) concluded that both substantive and stylistic variance may be involved in determining responses to the NEO PI-R in clinical patients.

336 Self-Report Inventories

Third, high scores on the Openness (0) domain are not equivalent to intelligence, but rather to divergent thinking and creativity. They also do not imply that persons are unprincipled or without values. They are willing to entertain new ideas and can apply these ideas conscientiously. In a similar manner, low scores on the O domain do not mean that persons are closed, defensive, or authoritarian, but rather that they have a narrower scope and intensity of interest. "Openness may sound healthier or more mature to many psychologists, but the value of openness or closedness depends on the requirement of the situation, and both open and closed individuals perform useful functions in society" (Costa & McCrae, 1992, p. 15).

Fourth, high scores on the Agreeableness (A) domain may seem to be more socially preferable and psychologically healthier, and such persons are generally easier with whom to interact. However, some situations require that the person be independent and skeptical of what is happening and being too agreeable can actually be a detriment. Dependent Personality Disorder would be characterized by a high score on the Agreeableness domain to illustrate that it is not necessarily psychologically healthy.

Finally, high scores on the Conscientiousness (C) domain reflect that the person is more active in planning and organized in carrying out their activities. These qualities may be expressed in academic and occupational achievement or in annoying, fastidious behaviors. Low scores on the Conscientious domain do not reflect that individuals are without principles to govern their behavior, but rather they are more lackadaisical in working toward their goals.

The six facet scores for each domain are intended to flesh out the general qualities that have been described by the parent domain scale. Important differences can be identified between individuals who have similar scores on the parent domain and a different pattern of scores on the facet scales for that domain. Two individuals with similar scores on the Extraversion (£) domain, one of whom has primary elevations on Activity (£4) and Excitement Seeking (£5), while the other has primary elevations on Assertiveness (£2) and Positive Emotions (£6), are very different persons.

The interpretation of the facet scales, in addition to the domain scales on the NEO PI-R, is recommended in most cases, and particularly in clinical, educational, and occupational assessments. It is conceivable in research applications that only the domain scales are relevant to the issue under study, and consequently, there is no reason to score and interpret the facet scales. It is very important to consider computer scoring the NEO PI-R when all the domain and facet scales are to be interpreted, because of the high probability of some scoring error in making that many calculations. Computer scoring also allows for the factor score for each domain to be computed directly rather than using the formulas presented in the Manual (Costa & McCrae, 1992, p. 8) to estimate them.

APPLICATIONS

As a self-report inventory, the NEO PI-R is easily administered in a wide variety of settings and for a variety of purposes. It is the most widely used self-report measure of personality in countries around the world. Costa and McCrae (2003) reported that there are 9 published translations, 25 validated translations, 8 research translations, and 3 more translations in progress. This 60-page, single-spaced, bibliography illustrates the variety of issues and

Revised NEO Personality Inventory 337

research on the NEO PI-Rand NEO-FFI. Any comprehensive review of this literature is beyond the scope of this Handbook.

There are numerous settings in which the NEO PI-R·is appropriate for use: clinical, educational, medical, organizational, and research. The NEO PI-R is primarily used in educational, organizational, and research settings. The NEO PI-R is probably underutilized in clinical and medical settings and would seem worthy of wider usage in these settings. The NEO PI-R comes out of a long line ofresearch on the five-factor model of personality described earlier (p. 315) and will not be reiterated. The use of the NEO PI-R is discussed for each of these other four settings in turn.

In clinical settings, the NEO PI-R can serve at least six useful purposes. First, it can provide a positive or nonpathological description of the person that can compensate for the heavy focus on psychopathology in most assessment tools and techniques. Most of the self-report inventories discussed in this Handbook have few, if any, positive statements to make about the person. Second, the focus on the more positive aspects of the person can help establish rapport and build the therapeutic alliance, and serves as an easy means of starting the feedback of the results of the assessment process before getting into the psychopathological issues. Third, there is a fairly extensive literature on the use of the NEO PI-R in the treatment of personality disorders (cf. Costa & Widiger, 2002). Fourth, the assessment of validity as described should be carried out routinely in clinical settings because of the higher probability of some type of response distortion. Fifth, knowledgeable others' ratings of the person using Form R can make an important contribution to under­ standing him or her, particularly when there is some reason to suspect that may be some type ofresponse distortion. Finally, the NEO PI-R is neither a diagnostic instrument nor a measure of psychopathology and cannot be used as the sole assessment tool or technique in a clinical setting.

In educational settings, the NEO PI-R can be used in advising students about personality characteristics that will facilitate or impede their academic progress. There are areas of study, such as chemistry or accounting, where careful attention to detail is mandatory for success, and other areas, such as philosophy or literature, where the focus is on more abstract or larger conceptual issues, and careful attention to detail is much less necessary. Persons with high scores on the Conscientiousness (C) domain are more likely to be successful in chemistry or accounting, while persons with high scores on the Openness ( 0) domain are more likely to be successful in philosophy or literature. In neither example is academic success foreclosed in the other area, but these individuals may have to work harder to recognize how their natural personality style affects their academic performance and they may need to find methods for coping with them to increase the probability of success. The NEO PI-R also can be used in counseling students in academic settings, which would be considered a clinical setting and was discussed earlier.

In medical settings, the NEO PI-R can be used to identify personality characteristics that might facilitate or impede treatment. The NEO PI-R will be better accepted by medical patients than other self-report inventories that have a heavy focus on psychopathology. Medical patients, particularly pain patients, are frequently upset at the thought of psy­ chological assessment because they think that it implies that the problem "is all in their head."

Medical patients with high scores on the Neuroticism (N) domain can alert the examiner to review their background and history for the potential impact of psychopathology on the

338 Self-Report Inventories

medical treatment. Medical patients with high scores on the Conscientiousness ( C) domain would be expected to be more likely to follow through on the recommended steps for treatment, particularly as the treatment process becomes more complex or long-term. An interesting line of research has used the NEO PI-R in predicting risk for coronary heart disease (cf. Costa, McCrae, & Dembroski, 1989), and Vollrath and Torgersen (2002) used the NEO PI-R to predict risky health behavior in college students. Costa and McCrae (2003) have listed the multiple areas in behavioral medicine in which the NEO PI-R is being used.

In occupational settings, the NEO PI-R can be used to identify personality characteristics that might facilitate or impede success in a specific occupation. As with educational settings, certain personality dimensions are more important in some occupations than others. These personality dimensions can be used in selecting candidates for specific occupations or in ad­ vising individuals on what occupations might be better suited for them. When the NEO PI-R is used to select potential candidates for specific occupations, the examiner must be aware that because examinees may simulate their scores on the appropriate domains, evaluating the validity of the NEO PI-R will be important (cf. Griffin et al., 2004).

When an occupation requires significant amounts of interpersonal interactions, individ­ uals with higher scores on the Extraversion (E) and Agreeableness (A) domains will be more likely to be successful than individuals with lower scores on these same domains. Conversely, when an occupation requires a significant amount of time by oneself, indi­ viduals with lower scores on the Extraversion and Agreeableness domains will be more likely to be successful than individuals with higher scores on these same domains. Again, the examiner is reminded that when individuals do not have the optimal scores on the personality dimensions for a specific occupation, their success is not precluded, but they need to be aware of the potential impact these personality dimensions may have on their performance.

PSYCHOMETRIC FOUNDATIONS

Demographic Variables

Age

Specific norms are not provided by age for adults on the NEO PI-R. There are some differences in young adults ( <20) and a separate profile form and norms are used for them. The items on the NEO PI-3 are easier to read than those on the NEO PI-R, and the NEO PI-3 can be used for adolescents 12 years of age and older. Further research currently is being conducted to determine whether the NEO Pl-3 can be considered as a replacement for the NEO PI-Rat all ages.

Terracciano, McCrae, Brant, and Costa (2005) examined age trends on the NEO PI-R in a sample of nearly two thousand adults in the Baltimore Longitudinal Study on Aging. There was a gradual curvilinear decline of slightly over one-half of a standard deviation in the Neuroticism (N) and Extraversion (E) domains from age 30 to age 90. There was a linear decline in the Openness (0) domain and linear increase in the Agreeableness (A) domain. There was a parabolic change in the Conscientiousness (C) domain with scores increasing until about age 70 and then slightly declining thereafter. All these changes in adulthood across the five domains were about one T score point per decade or slightly more than one­ half of a standard deviation across the entire 60-year age span. A cross-sectional analysis

Revised NEO Personality Inventory 339

of these data produced results that are similar to the longitudinal analysis. Terracciano et al. also provide similar information on all 30 of the facet scales on the NEO PI-R.

Gender

Gender does not create any general issues in NEO Pl-R interpretation because separate norms (profile forms) are used for men and women. Any gender differences in how individ­ uals responded to the items on each scale are removed when the raw scores are converted

to T scores. Consequently, men and women with a T score of 60 (84th percentile) on

Agreeableness (A) are one standard deviation above the mean, although women have a

slightly higher raw score (~142) than men (~136; Costa & McCrae, 1992, Appendix C,

p. 79). Costa, Terracciano, and McCrae (2001) analyzed gender differences in 26 cultures and found that these gender differences were typically less than one-half of a standard deviation (5 T points), and most were closer to one-quarter of a standard deviation, relative to variations within gender.

Education

The potential effects of education have not been investigated in any systematic manner on the NEO PI-R, It is not apparent that such research would yield any significant findings given the ease with which the NEO PI-R is read and the similar findings in factor structure across multiple cultures.

Ethnicity

The effects of ethnicity per se on NEO PI-R performance have not been studied, if ethnicity as construed as being different from culture. However, the prolific literature on the cross­ cultural use of the NEO PI-R is discussed briefly in the next section.

Cross-Cultural Implementation

Costa and McCrae (2003) reported that there are 9 published translations, 25 validated translations, 8 research translations, and 3 more translations in progress of the NEO PI-R. The breadth of the use of the NEO PI-R across various cultures can be seen by the fact that there are 79 contributing members to the Personality Profiles of Cultures Project who represent 51 cultures from six continents (McCrae, Terracciano, et al., 2005). This project is looking at the aggregate personality profiles of different cultures to assess whether they can provide insight into cultural differences and the stereotypes of national character (McCrae & Terracciano, 2006). The robustness of the factor structure of the NEO PI-R across these various cultures not only speaks to the usefulness of the NEO PI-R cross-culturally, but it allows for comparisons to be made into the actual differences in aggregate personality profiles. As would be expected, stereotypes of national character are erroneous (McCrae & Terracciano, 2006), similar to the erroneous conceptualization that all patients within a specific diagnostic category are alike (pp. 60-61). There are small differences in these aggregate personality profiles across the different cultures, but much larger variability within cultures. These variations in aggregate personality profiles appear to reflect real differences that warrant further investigation.

In summary, it appears that demographic variables have minimal impact on the

NEO PI-R profile in most individuals. The fact that the NEO PI-R can be read to indi­ viduals and is available in many different languages makes it applicability even broader.

340 Self-Report Inventories

Reliability

The NEO PI-R Manual (Costa & McCrae, 1992, table 5, p. 44) reports the reliability (coefficient alpha) data for 1,539 individuals for Form S. Coefficient alpha ranged from

.56 to .81 for the facet scales and .86 to .92 for the domain scales. The reliability data are quite good for the domain scales that contain 48 items each. As expected, the reliability data are somewhat lower, though still very respectable, for the facet scales that only have eight items each.

A subset of the college students (N = 208) in the normative sample for the NEO PI-R were retested after an average of nearly 3 months with the NEO-FFI, which allowed

determination of the reliability of the five domain scores. The test-retest correlations ranged from .75 to .83 across the five scales and averaged .79. The standard error of measurement is about 4 T points for the domain scales; that is, the individual's true score on the domain scales will be within ±4 T points two-thirds of the time.

Stability

There is impressive research on the long-term stability of NEO PI-R scores. Costa and McCrae (1988) reported that the stability coefficients over a 6-year period in a large sample of adults for the domains of N (Neuroticism), E (Extraversion), and O (Openness) were .83, .82, and .83, respectively. The stability coefficients over a 3-year period for the domains of A (Agreeableness) and C (Conscientiousness) were .63 and .79, respectively. These stability coefficients are higher and over a longer time period than for any of the other self-report inventories reviewed in this Handbook.

CONCLUDING COMMENTS

The voluminous literature on the five-factor model of personality provides solid underpin­ nings for the NEO PI-R (Costa & McCrae, 1992). The Personality Profiles of Cultures Project that represents 51 cultures from six continents (McCrae, Terracciano, et al., 2005) shows how well regarded the NEO PI-R is internationally. More widespread use of the NEO PI-R in clinical and medical settings to provide a positive perspective on the person is warranted given the heavy bias toward psychopathology in virtually all other assessment tests and techniques. The existence of a parallel form for rating of the person by a knowl­ edgeable other (Form R) is an invaluable source of information any time there is reason to suspect any type of response distortion that seems particularly helpful in clinical and medical settings.

REFERENCES

Bagby, R. M., Rector, N. A, Bindseil, K., Dickens, S. F., Levitan, R. D., & Kennedy, S. H. (1998). Self-reports and informant ratings of personalities of depressed outpatients. American Journal ofPsychiatry, 155, 437-438.

Ballenger, J. F., Caldwell-Andrews, A., & Baer, R. A. (2001). Effects of positive impression man­ agement on the NEO PI-Rina clinical population. Psychological Assessment, 13, 254-260.

Revised NEO Personality Inventory 341

Berry, D. T. R., Bagby, R. M., Smerz, J., Rinaldo, J.C., Caldwell-Andrews, A., & Baer, R. A. (2001). Effectiveness of NEO PI-R research validity scales for discriminating analog malingering and genuine psychopathology. Journal ofPersonality Assessment, 76, 496-516.

Caldwell-Andrews, A. (2001). Relationships between MMPI-2 validity scales and NEO PI-R exper­ imental validity scales in police candidates. Unpublished doctoral dissertation, University of Kentucky.

Caldwell-Andrews, A., Baer, R. A., & Berry, D. T. R. (2000). Effects of response sets on NEO PI-R scores and their relation to external criteria. Journal of Personality Assessment, 74, 472- 488.

Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources.

Costa, P. T., Jr., & McCrae, R.R. (1988). Personality in adulthood: A six-year longitudinal study of self-reports and spouse ratings on the NEO PI. Journal of Personality and Social Psychology, 54, 853-863.

Costa, P. T., Jr., & McCrae, R.R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.

Costa, P. T., Jr., & McCrae, R.R. (2003). Bibliographyfor the NEO Pl-Rand NEO FF!. Lutz, FL: Psy­ chological Assessment Resources. Available at www3.parinc.com/uploads/pdfs/NEO_bib.pdf.

Costa, P. T., Jr., McCrae, R.R., & Dembroski, T. M. (1989). Agreeableness vs. antagonism: Explica­ tion of a potential risk factor for CHD. In A. Siegman & T. M. Dembroski (Eds.), In search of coronary-prone behavior: Beyond Type A (pp. 41-63). Hillsdale, NJ: Erlbaum.

Costa, P. T., Jr., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality traits across cultures: Robust and surprising findings. Journal of Personality and Social Psychology, 81, 322-331.

Costa, P. T., Jr., & Widiger, T. A. (2002). Personality disorders and the five-factor model ofpersonality (2nd ed.). Washington, DC: American Psychological Association.

Fiedler, E. R., Oltmanns, T. F., & Turkheimer, E. (2004). Traits associated with personality disorders and adjustment to military life: Predictive validity of self and peer reports. Military Medicine, 169, 32-40.

Finn, S. (1996). Using the MMPI-2 as a therapeutic intervention. Minneapolis: University of Min­ nesota Press.

Fischer, C. T. (1994). Individualizing psychological assessment. Hillsdale, NJ: Erlbaum.

Griffin, B., Hesketh, B., & Grayson, D. (2004). Applicants faking good: Evidence of bias in the NEO PI-R. Personality and Individual Differences, 36, 1545-1558.

McCrae, R. R., Costa, P. T., Jr., & Martin, T. A. (2005). The NEO PI-3: A more readable revised NEO Personality Inventory. Journal of Personality Assessment, 84, 260-270.

McCrae, R. R., Costa, P. T., Jr., Parker, W. D., Mills, C. J., Terracciano, A., De Fruyt, F., et al. (2002). Personality trait development from age 12 to age 18: Longitudinal, cross-sectional, and cross-cultural analyses. Journal of Personality and Social Psychology, 83, 1456-1468.

McCrae, R.R., & Terracciano, A. (2006). National character and personality. Current Directions in Psychological Science, 15, 156-161.

McCrae, R.R., Terracciano, A., & 79 Members of the Personality Profiles of Cultures Project. (2005). Personality profiles of cultures: Aggregate personality traits. Journal of Personality and Social Psychology, 89, 407-425.

Morey, L. C. (1991). Personality Assessment Inventory: Professional manual. Odessa, FL: Psycho­ logical Assessment Resources.

Morey, L. C., Quigley, B. D., Sanislow, C. A., Skodol, A. E., McGlashan, T. H., Shea, M. T., et al. (2002). Substance or style? An investigation of the NEO PI-R validity scales. Journal of Per­ sonality Assessment, 79, 583-599.

342 Self-Report Inventories

Pauls, C. A., & Crost, N. W. (2005). Effects of different instructional sets on the construct validity of the NEO PI-R. Personality and Individual Differences, 39, 297-308.

Piedmont, R. L., McCrae, R. R., Riemann, R., & Angleitner, A. (2000). On the invalidity of va­ lidity scales: Evidence from self-report and observer ratings in volunteer samples. Journal of Personality and Social Psychology, 78, 582-593.

Schink.a, J. A., Kinder, B. N., & Kremer, T. (1997). Research validity scales for the NEO PI-R: Development and initial validation. Journal ofPersonality Assessment, 68, 127-138.

Terracciano, A., McCrae, R. R., Brant, L. J., & Costa, P. T., Jr. (2005). Hierarchical linear modeling analyses of the NEO PI-R scales in the Baltimore Longitudinal Study of Aging. Psychology and Aging, 20, 493-506.

Vollrath, M., & Torgersen, S. (2002). Who takes health risks? A probe into eight personality types. Personality and Individual Differences, 32, 1185-1198.

Wiggins, J. S. (Ed.). (1996). The five-factor model ofpersonality. New York: Guilford Press.

Yang, J., Bagby, R. M., & Ryder, A.G. (2000). Response style and the NEO PI-R: Validity scales and spousal ratings in a Chinese psychiatric sample. Assessment, 7, 389-402.

Young, M. S., & Schink.a, J. A. (2001). Research validity scales for the NEO PI-R: Additional evidence for reliability and validity. Journal ofPersonality Assessment, 76, 412-420.