Purdue University

Purdue e-Pubs
Department of Psychological Sciences Faculty
Department of Psychological Sciences
Publications

2012

A Five-Factor Measure of Obsessive-Compulsive

Personality Traits.
Douglas B. Samuel
Purdue University, [email protected]

Ashley D.B. Riddell

Donald R. Lynam
Purdue University, [email protected]

Joshua D. Miller

Thomas A. Widiger

Follow this and additional works at: http://docs.lib.purdue.edu/psychpubs

Part of the Psychology Commons

Recommended Citation
Samuel, Douglas B.; Riddell, Ashley D.B.; Lynam, Donald R.; Miller, Joshua D.; and Widiger, Thomas A., "A Five-Factor Measure of
Obsessive-Compulsive Personality Traits." (2012). Department of Psychological Sciences Faculty Publications. Paper 6.
http://dx.doi.org/10.1080/00223891.2012.677885

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for
additional information.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
1

A Five-Factor Measure of Obsessive-Compulsive Personality Traits

Douglas B. Samuel

Ashley D. B. Riddell

Donald R. Lynam

Joshua D. Miller

Thomas A. Widiger

In press; Journal of Personality Assessment

Author’s Notes:

Douglas B. Samuel, Department of Psychological Sciences, Purdue University; Ashley D. B.

Riddell, Department of Psychology, University of Kentucky; Donald R. Lynam, Department of

Psychological Sciences, Purdue University; Joshua D. Miller, Department of Psychology, University of

Georgia; Thomas A. Widiger, Department of Psychology, University of Kentucky.

Correspondence regarding this article should be directed to Douglas B. Samuel, Department of

Psychological Sciences, Purdue University 703 Third St, West Lafayette, IN 47906.

Email: [email protected]
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
2
Abstract

The current study provides convergent, discriminant, and incremental validity data for the Five-Factor

Obsessive-Compulsive Inventory (FFOCI), a newly-developed measure of traits relevant to obsessive-

compulsive personality disorder (OCPD) from the perspective of the five-factor model (FFM). Twelve

scales were constructed as maladaptive variants of specific FFM facets (e.g., Perfectionism as a

maladaptive variant of FFM competence). On the basis of data from 407 undergraduates (oversampled

for OCPD symptoms) these 12 scales demonstrated convergent correlations with established measures

of OCPD and the FFM. Further, they obtained strong discriminant validity with respect to facets from

other FFM domains. Most importantly, the individual scales and total score of the FFOCI obtained

incremental validity beyond existing measures of the FFM and OCPD for predicting a composite

measure of obsessive-compulsive symptomatology. The findings support the validity of the FFOCI as a

measure of obsessive-compulsive personality traits, as well as of maladaptive variants of the FFM.

Keywords: five-factor model, obsessive-compulsive personality disorder, validation, maladaptive,

perfectionism.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
3
A Five-Factor Measure of Obsessive-Compulsive Personality Traits

Obsessive-compulsive personality disorder (OCPD) is a condition characterized by such features as

perfectionism; devotion to work to the exclusion of other important activities; preoccupation with the

details, order, and organization of activities and tasks; rigidity; and difficulty expressing warmth or

affection. Obsessive-compulsive is the most prevalent personality disorder (PD) within community

samples (Torgersen, 2009) and has a lengthy history within the clinical literature, having been included

in all previous diagnostic manuals and tracing its roots to Freud’s “anal character” (Pfohl & Blum,

1995; Costa, Samuels, Bagby, Daffin, & Norton, 2005).

The current nomenclature of the American Psychiatric Association (APA), however, has received

considerable criticism for its assumption that PDs are categorically distinct entities (Clark, 2007; First

et al., 2002; Livesley, 2003; Trull & Durrett, 2005; Widiger & Trull, 2007). The limitations of this

categorical approach include excessive diagnostic co-occurrence, arbitrary and inconsistent diagnostic

boundaries, and insufficient coverage. One additional difficulty is the provision of a single diagnostic

term to describe a construct characterized by a heterogeneous constellation of maladaptive personality

traits. For instance, in the DSM-IV-TR (APA, 2000), any four of eight criteria are required for the

diagnosis of OCPD. Therefore, there are 163 different combinations of criteria that yield an OCPD

diagnosis. Moreover, because only half of the criteria are required, it is possible that two individuals

could both meet the diagnosis, yet not even share a single feature. This problematic heterogeneity has

prompted some researchers to propose subtypes of OCPD in order to develop more specific treatment

guidelines (Ansell et al., 2010).

The heterogeneity of OCPD has been further verified through factor analysis, which has supported

the position that OCPD is more fruitfully considered a constellation of maladaptive personality traits

(e.g., Baer, 1994; Grilo, 2004; Hummelen, Wilberg, Pedersen, & Karterud, 2008; Pinto, Ansell, Grilo,

& Shea, 2007). This situation has created a number of difficulties in the assessment of OCPD. For
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
4
example, existing measures typically evince problematic levels of reliability (e.g., Cronbach’s alpha

values < .50; Samuel & Widiger, 2010). Furthermore, these same measures often obtain such weak

convergence with one another that one might question if they are even assessing the same constructs

(Widiger & Boyd, 2009). Given the limitations of the categorical approach, researchers have suggested

that the DSM-IV-TR PDs, including OCPD, are best understood and assessed as maladaptive variants of

the traits identified within existing dimensional models of personality (Clark, 2007; Krueger & Eaton,

2010; Widiger & Simonsen, 2005). What is needed is a measure that parses the construct of OCPD into

components that would allow clinicians and researchers to more usefully assess, study, and treat those

specific traits that have previously been lumped into the heterogeneous category of OCPD.

One such framework for identifying those components is the five-factor model (FFM; McCrae &

Costa, 2003). The FFM consists of the broad domains of neuroticism (emotional instability or negative

affectivity) versus emotional stability, extraversion (surgency or positive affectivity) versus

introversion, openness (intellect or unconventionality) versus closedness to experience, agreeableness

versus antagonism, and conscientiousness (constraint) versus disinhibition. Each of these five broad

domains were further differentiated into six more specific facets by Costa and McCrae (1995) as they

developed and conducted research with the NEO Personality Inventory-Revised (NEO PI-R; Costa &

McCrae, 1992), the predominant measure of the FFM. For instance, the six facets of conscientiousness

are competence, order, dutifulness, achievement-striving, self-discipline, and deliberation.

A considerable body of research has also indicated that the DSM-IV-TR PDs, including OCPD, can

be understood as maladaptive variants of the domains and facets of the FFM (Clark & Livesley, 2002;

O’Connor, 2005; Samuel & Widiger, 2008). Nonetheless, existing measures of the FFM, which were

designed to assess normative personality traits, might be insufficient for assessing those pathological

aspects of OCPD pathology (Haigler & Widiger, 2001; Reynolds & Clark, 2001). As such it would be

useful to develop a FFM measure that focuses explicitly on the maladaptive aspects and levels of the
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
5
trait that are relevant to OCPD. A first step in developing such a measure is to determine the FFM traits

that best define the core components of OCPD. The existing literature provides useful data for this

decision in the form of expert opinions as well as empirical research.

Lynam and Widiger (2001) asked OCPD researchers to describe a prototypic case of OCPD in terms

of the 30 facets of the FFM, using the Five-Factor Model Rating Form (FFMRF; Mullins-Sweatt,

Jamerson, Samuel, Olson, & Widiger, 2006). These researchers suggested that an individual with a

prototypic case of OCPD would be high in competence, order, dutifulness, achievement-striving, self-

discipline, and deliberation from the conscientiousness domain; high in anxiety (from neuroticism);

low in openness to feelings, actions, ideas, and values; low in warmth and excitement-seeking (from

extraversion); and low in impulsivity (from neuroticism). This trait profile was quite consistent with a

comparable survey of practicing clinicians by Samuel and Widiger (2004). The only notable difference

was that the clinicians’ description of OCPD did not include low impulsivity nor low openness to ideas

and feelings (the 2.22 score for low openness to feelings fell just above the rationally derived 2.00 cut

point for a low score). Finally, Widiger, Trull, Clarkin, Sanderson, and Costa (2002) coded each of the

DSM-IV-TR diagnostic criteria (APA, 2000) in terms of the FFM. Their coding identified considerably

fewer facets of the FFM (e.g., excluding high anxiousness, low excitement-seeking, and high self-

discipline) due largely to being confined to the symptoms contained within the DSM-IV-TR criterion

set.

Saulsman and Page (2004) meta-analyzed the correlations between measures of OCPD and the FFM

and reached the surprising conclusion that OCPD was only weakly related to FFM conscientiousness,

producing a weighted mean effect size of only .23 (p < .0001). However, they suggested that this effect

size varied across OCPD measures such that the relationship between OCPD and conscientiousness

was particularly strong when the Millon Clinical Multiaxial Inventory (MCMI; Millon, 1994) was

used. Furthermore, they suggested that this effect size varied across OCPD measures such that the
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
6
relationship between OCPD and conscientiousness was heavily influenced by a version of the Millon

Clinical Multiaxial Inventory (MCMI-III; Millon, 1994). The MCMI-III obtained a weighted mean

effect size of .52 with conscientiousness, whereas all other self-report PD scales showed an effect size

of only .03; interview-based measures evinced an effect size of -.05. This difference is not surprising as

the MCMI-III often fails to converge with other measures of OCPD (Widiger & Boyd, 2009).

However, an additional explanation for the effect size noted by Saulsman and Page (2004) was a

potential limitation in the NEO Personality Inventory-Revised’s (NEO PI-R; Costa & McCrae, 1992)

assessment of maladaptively high conscientiousness. When Haigler and Widiger (2001) experimentally

manipulated NEO PI-R items by inserting words to make them more excessive, extreme, or

maladaptive variants of the same content, the correlations between conscientiousness and OCPD

increased substantially.

Samuel and Widiger (2008) further investigated the relationship between OCPD and FFM

conscientiousness when they replicated and extended the Saulsman and Page (2004) meta-analysis to

consider the 30 FFM facets. They found positive relationships between OCPD and the

conscientiousness facets of order, dutifulness, achievement striving, self-discipline, and deliberation. In

addition, when they tested for moderation by instrument, they indicated that the high convergence with

FFM conscientiousness noted previously for the MCMI-III was also evident with the Schedule for

Nonadaptive and Adaptive Personality - 2 (SNAP-2; Clark, Simms, Wu, & Casillas, in press).

In the absence of a gold standard for the assessment of these constructs, Samuel and Widiger (2011)

administered multiple measures of conscientiousness, OCPD, and specific components of OCPD. This

multi-faceted approach allowed for an examination of the overall relationship rather than specific

conceptualizations offered by individual instruments. Importantly, they reported that FFM

conscientiousness was particularly strongly related with the more specific components of OCPD, such

as the Compulsivity subscale from the Dimensional Assessment of Personality Pathology-Basic

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
7
Questionnaire (DAPP-BQ; Livesley & Jackson, 2009), and the Workaholism and Propriety subscales

from the SNAP-2 (Clark et al., in press).

Based on the surveys of researchers (Lynam & Widiger, 2001), surveys of clinicians (Samuel &

Widiger, 2004), and empirical research (Samuel & Widiger, 2008, 2011; Saulsman & Page, 2004)

twelve facets of the FFM were identified as being particularly relevant for the assessment of OCPD

from the perspective of the FFM. As discussed above, the six facets of conscientiousness have been the

most frequently identified for describing OCPD. Nevertheless, the surveys and empirical research has

also identified additional facets from the domains of extraversion, neuroticism, and openness domains.

Specifically, an adequate FFM description of obsessive-compulsive personality traits (OCPT) should

include low warmth and excitement-seeking from extraversion; high anxiety from neuroticism; and low

openness to feelings, actions, and values, in addition to the facets of conscientiousness.

In order to assess these FFM obsessive-compulsive personality traits we developed brief scales,

including Perfectionism (a variant of FFM competence), Fastidiousness (FFM order), Punctiliousness

(FFM dutifulness), Workaholism (FFM achievement-striving), Doggedness (FFM self-discipline),

Ruminative Deliberation (FFM deliberation), Detached Coldness (low FFM warmth), Risk Aversion

(low FFM excitement-seeking), Excessive Worry (high FFM anxiety), Constricted (low FFM openness

to feelings), Inflexibility (low FFM openness to actions), and Dogmatism (low FFM openness to

values).

The purpose of the current study was to investigate these twelve FFM OCPT scales (collectively

referred to as the Five-Factor Obsessive-Compulsive Inventory [FFOCI]), in terms of internal

consistency, convergent and discriminant validity with existing measures, and incremental validity over

the NEO PI-R and existing measures of OCPD.

Method

Procedures
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
8
The study’s undergraduate participants were drawn from the introductory psychology student

participant pool at the University of Kentucky. To ensure the inclusion of elevated levels of OCPD

pathology in our sample we administered the OCPD scale from the Personality Diagnostic

Questionnaire-4 (PDQ-4; Bagby & Farvolden, 2004) to a pool of over 1400 potential participants and

selected the top-scoring 100 individuals (each endorsed at least five of the eight PDQ-4 items) over two

consecutive semesters and invited them to participate in the study. Once a number of these participants

had been sampled, the study was opened to the entire subject pool to expand the range.

All measures were administered via SurveyMonkey, a secure online survey service. Given the

online format, individuals indicated their informed consent by selecting the appropriate box. After

providing informed consent, participants completed selected scales from personality and PD

instruments; the order of administration was standard across all participants. Participants were allowed

as much time as necessary to complete the materials (which required approximately 2.5 hours), and

could temporarily suspend participation whenever necessary. Upon completion, each participant

received a debriefing document and research participation credits.

Participants

A total of 500 responses were obtained (including 1001 from the oversampled group). Of these, 51

(6 oversampled) were eliminated due to incomplete or missing data for a total of 449 participants (95

oversampled). Finally, 42 participants (6 oversampled) were excluded from the study due to elevated

scores on the survey’s validity scale (described below) yielding a useable sample of 407 participants of

whom 89 were pre-screened for elevated OCPD scores.

The sample of participants was split such that 204 (45 oversampled) were included in the item

selection process and 203 (44 oversampled) were utilized for the convergent, discriminant, and

incremental validity analyses. These two groups did not differ significantly on any demographic

variables. Any remaining missing data were assumed to be missing at random and were imputed using
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
9
the expectation maximization (EM) procedure. We chose EM because it has been shown to produce

more accurate estimates of population parameters than other methods (Enders, 2006).

The entire sample was predominantly female (61%) and Caucasian (84%), but some participants did

indicate their race as African American (5%), Asian (3%), multiracial (3%), or “other” (4%). In

addition, (2%) identified their ethnicity as Hispanic. Finally, the mean age was 19.1 years (sd = 3.4).

The level of OCPD pathology in the validation sample was high according to the PDQ-4. The

number of criteria endorsed ranged from zero to eight, with a mean of 3.3 (sd = 1.8). The oversampling

strategy appeared to be successful in obtaining significant levels of OCPD pathology as evidenced by

the mean values on other measures, as well. For example, the mean item-score on the WISPI-IV OCPD

scale was 4.4 (sd = 1.5) in the current sample compared with 3.8 (sd = 1.3) among psychiatric

inpatients in a validation sample (Smith, Klein, & Benjamin, 2003). In addition, the mean dimensional

t-score on the SNAP-2 OCPD scale in the current sample was 52.4 (sd = 12.4), which is comparable to

the mean (53.1, sd = 11.1) among a small sample of outpatients (n = 63) reported in the SNAP-2

manual (Clark et al., in press). Finally, the DAPP-BQ Compulsivity scale had a mean of 52.7 (sd =

12.0) in our sample, the norm for a clinical sample was 51.4 (sd = 12.3) in the manual (Livesley &

Jackson, 2009).

Materials

The current study includes a number of psychometric measures; namely, the FFOCI, three

alternative measures of the domain of conscientiousness, four alternative measures of OCPD, and two

scales assessing specific components of OCPD.

Five-Factor Obsessive-Compulsive Inventory (FFOCI). The initial item pool for the FFOCI

consisted of 298 items, with approximately 22 items per subscale (range of 20-24 items per FFM facet),

answered on a 5-point scale ranging from strongly disagree to strongly agree. Items were written to

assess obsessive-compulsive maladaptive variants of each respective FFM facet, modeled after the
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
10
development of the Elemental Psychopathy Assessment (Lynam et al., 2011). For example, the items

for FFOCI did not concern simply deliberation but more specifically the ruminative deliberation that is

characteristic of OCPD. Items were written so as to be maladaptive in nature (i.e., a positive response

would be likely to impede general functioning) while remaining true to the spirit of the FFM facet

description.

Specifically, six subscales in the item pool assess obsessive-compulsive variants of FFM

Conscientiousness: Perfectionism (e.g., “People often think I work too long and hard to make things

perfect” and “I like my work to be flawless and unblemished”), Fastidiousness (e.g., “I probably spend

more time than is needed organizing and ordering things” and “I need to consider every little detail”),

Punctiliousness (e.g., “Some persons suggest I can be excessive in my emphasis on being proper and

moral” and “I have such a strong sense of duty that I sometimes become over-committed”),

Workaholism (e.g., “My drive to succeed keeps me going when others have stopped” and “I get so

caught up in my work that I lose time for other things”), Doggedness (e.g., “I have a strong, perhaps at

times even excessive, single-minded determination” and “If I start something I work until it is

complete”), and Ruminative Deliberation (e.g., “I think things over and over and over before I make a

decision” and “I often dwell on every possible thing that might go wrong”). Two subscales assess

OCPD facets of low Extraversion: Detached Coldness (e.g., “I often come across as formal and

reserved” and “Warmth and intimacy are not my strengths”), and Risk Aversion (e.g., “I would always

sacrifice fun and thrills for the security of my future” and, reverse-scored, “If it sounds exciting, I'd try

anything once”). One subscale assesses an OCPD variant of Neuroticism: Excessive Worry (e.g., “I am

often concerned, even nervous, about things going wrong” and “I am a worrier”). Three subscales

assess OCPD facets of low Openness to Experience: Constricted (e.g., “I am a thinker, not a feeler” and

“Strong emotions are not that important in my life”), Inflexibility (e.g., “I like to keep to the ‘tried and

true’ rather than try new things” and “I much prefer predictability than exploring the unknown”), and
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
11
Dogmatism (e.g., “It troubles me how society is losing its strong moral core” and “I live my life by a

set of tough, unyielding moral principles”).

Validity Scale. In the current study, a five-item validity scale was used. Each item describes a

behavior that was very unlikely to be true (e.g., “I am currently in the Guinness Book of World

Records” and, reverse coded, “I have used a computer in the past 2 years”), thus an endorsement

suggested the individual was not attending to the item’s content. The items were rated on a five-point

Likert scale whose values ranged from strongly disagree to strongly agree.

Conscientiousness-Related Scales.

NEO Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992). The NEO PI-R is a 240-

item self-report inventory designed to assess normal personality domains according to the FFM,

including conscientiousness. It uses a 5-point Likert scale (ranging from strongly disagree to strongly

agree). Alpha coefficients ranged from .47 (activity) to .87 (warmth) for the facets.

Experimentally Manipulated NEO PI-R (EXP-NEO; Haigler & Widiger, 2001). The EXP-NEO is an

experimental version of the NEO PI-R in which items were altered to refer to extreme and/or

maladaptive variants of the existing content. For example, the item “I am known for my prudence and

common sense” was altered to “I have been told that I may at times display an excessive prudence and

rigid common sense” and the item “I tend to be somewhat fastidious or exacting” became “I tend to be

overly fastidious or exacting”). The EXP-NEO uses the same 5-point Likert scale as the NEO PI-R and

the EXP-NEO Conscientiousness facet scales have evinced large, significant correlations with NEO PI-

R Conscientiousness facet scales (Haigler & Widiger, 2001; Samuel & Widiger, 2011). Only the 48

conscientiousness items from the EXP-NEO were included in the present study. Cronbach’s alpha for

this domain was .90.

HEXACO Personality Inventory – Revised (HEXACO PI-R; Ashton & Lee, 2008). The HEXACO-

PI-R is a 200-item self-report inventory designed to assess six domains of normal personality
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
12
functioning (Honesty-Humility, Emotionality, Extraversion, Agreeableness, Conscientiousness, and

Openness to Experience). The broad domains are further subdivided into four facets, each of which is

assessed using an 8-item subscale. The HEXACO PI-R uses a 5-point Likert scale (ranging from

strongly disagree to strongly agree). Only the 32 items from the HEXACO-PI Conscientiousness scale

were included in the present study. Alpha value for this domain was .91.

Obsessive-Compulsive Personality Disorder and OCPD Component Scales.

Dimensional Assessment of Personality Pathology-Basic Questionnaire (DAPP-BQ; Livesley &

Jackson, 2009). The DAPP-BQ is a 290-item self-report inventory consisting of 18 scales designed to

measure aspects of personality pathology (e.g., compulsivity and affective instability). Responses are

given using a 5-point Likert-type scale ranging from strongly disagree to strongly agree. The present

study included only the 16-item DAPP Compulsivity scale, which consists of items such as “I do jobs

thoroughly even if no one will ever see them.” The alpha value for the compulsivity scale was .94.

Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon, 1994). The MCMI-III is a 175-item

true-false self-report inventory designed to assess DSM-IV-TR (APA, 2000) PDs. The present study

included only the 17 MCMI-III items pertaining to OCPD, as well as the seven Grossman facet items

for OCPD. The alpha for the MCMI-III scale was .77.

Personality Diagnostic Questionnaire-4 (PDQ-4; Bagby & Farvolden, 2004). The PDQ-4 is a 99-

item true-false self-report inventory intended to measure the 10 DSM-IV-TR (APA, 2000) PDs. The

present study included only the eight items pertaining to OCPD. The alpha for the PDQ-4 OCPD scale

was .54.

Schedule for Nonadaptive and Adaptive Personality -2 (SNAP-2; Clark et al., in press). The SNAP-

2 is a 390-item factor analytically derived true-false, self-report inventory designed to measure both

normal and abnormal personality functioning through dimensional scales. It includes 12 scales to

measure maladaptive personality traits (e.g., manipulativeness), three scales to assess broad personality
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
13
temperaments (e.g., disinhibition), six validity scales, and diagnostic scales for DSM-IV TR (APA,

2000) PDs. The present study included only the 25 items pertaining to OCPD and the 35 items forming

the Constraint trait scale. Although mostly independent, five items are scored for both of these scales,

for a total of 55 unique items. The alphas were .79 and .87 for the OCPD and Constraint scales,

respectively.

Wisconsin Personality Disorder Inventory (WISPI-IV; Klein et al., 1993). The WISPI-IV is a 204-

item questionnaire designed to measure the DSM-IV PDs. Using a ten-point Likert scale (ranging from

not at all; never applies to me to extremely; always applies to me) participants rate how often

statements have applied to them in the past five years. The present study included only the 20 WISPI-

IV items pertaining to OCPD. The alpha for the WISPI-IV was .93.

Results

FFOCI Item Selection

Using one half of the data set, the final item selections were made using a criterion-keying approach

(Clark & Watson, 1995). Each potential FFOCI item was correlated with the OCPD scales, their

respective NEO PI-R facet scales and, for FFOCI conscientiousness items, their respective facet scale

of the EXP-NEO and the full HEXACO PI-R Conscientiousness scale. We selected items for each

subscale that obtained the maximal correlations with these criterion measures. For example, the item “I

like my work to be flawless and unblemished” was retained for the Perfectionism subscale (C1) as it

evinced correlations with the nine criteria ranging from .18 (PDQ-4 OCPD) to .46 (HEXACO PI-R

Perfectionism), with a median value of .31. Similarly, the item “People consider me a rather serious and

reserved person” was retained for the Detached Coldness subscale (low E1) as it correlated significantly

with the five relevant criteria including -.35 with NEO PI-R warmth and .39 with the WISPI-IV OCPD

scale.

There was not a strict cutoff in terms of the magnitude of the relationships with the criteria as they
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
14
varied across the FFOCI subscales (e.g., not all measures of OCPD would be expected to include all 12 of

the FFM components of OCPD). At times an item selection could be somewhat arbitrary, as multiple

items would be expected to obtain comparably strong results. We also examined all candidate items to

avoid explicitly redundant items. Finally, we aimed for approximately 30% of the items on the final

scales to be reverse-scored, but this was not always possible. Indeed, reverse-scored items tended, on

average, to function less effectively (Rodebaugh, Woods, & Heimberg, 2007) and nine of the twelve

FFOCI scales include only two reverse-scored items, while the other scales had three.

Descriptive Properties of the FFOCI scales

The properties of each FFOCI scale are presented in Table 1. Cronbach’s alpha values ranged from

.77 to .87. The average corrected item-total correlations, which index the relationship between each

individual item and all others in a given scale, ranged from .45 to .60. The minimum, maximum, and

mean inter-item correlations for the items in each scale are also presented in Table 1. The means ranged

from .28 to .42 and no inter-item correlations were above .80.

Table 2 presents the intercorrelations among the FFOCI scales. All but five of these correlations

were significant at p > .01, with a majority of the effect sizes greater than r = .40. Not surprisingly,

correlations were higher among scales within the same FFM domain (e.g., the correlation among the

FFOCI conscientiousness scales ranged from .59 to .74, with a median of .67). However, there were

also exceptions. For example, the Constricted (O3) scale related rather weakly with most other FFOCI

scales (a correlation of .61 with Detached Coldness was the exception).

Convergence of FFOCI Scales with Related Personality Traits

Convergent validity, reported in Table 3, was examined by correlating each of the 12 FFOCI

subscales with their corresponding NEO PI-R facets (e.g., FFOCI Perfectionism correlated with NEO

PI-R Competence). Significant convergent validity correlations were obtained for all 12 FFOCI

subscales with their respective NEO PI-R facet scales. In fact, all of these correlations are considered
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
15
large (i.e., r > .50) according to Cohen (1992), except for FFOCI Perfectionism, which obtained a

correlation of .45. Some of these values are negative because the FFOCI scale is keyed in the opposite

direction as the NEO PI-R facet. For example, FFOCI Detached Coldness correlated -.74 with the NEO

PI-R facet of Warmth.

The FFOCI Conscientiousness scales were also correlated with their corresponding EXP-NEO

Conscientiousness facets, as well as with HEXACO PI-R Conscientiousness and SNAP-2 Constraint

scales. These convergent validity correlations were again significant in all instances as the FFOCI

scales related strongly with the respective EXP-NEO facets (ranging from .52 for Perfectionism to .76

for Fastidiousness), HEXACO PI-R Conscientiousness (ranging from .66 for Perfectionism and

Punctiliousness to .75 for Fastidiousness), and SNAP-2 Constraint (.41 for Perfectionism to .60 for

Ruminative Deliberation).

Discriminant validity was investigated by analyzing the relationships between each of the 12 FFOCI

subscales and all other, non-corresponding, NEO PI-R facet scales. The second row of Table 3 provides

the mean of the discriminant correlations between the FFOCI subscale and the NEO PI-R facet scales

within the same domain. The third row provides the mean correlation of the FFOCI subscale with the

24 NEO PI-R facet scales outside the domain. Consistent with expectations, the within-domain

discriminant correlations were substantial, whereas those outside the domain were small. For example,

the FFOCI Workaholism subscale obtained an average correlation of .51 with the five non-

corresponding facets within the Conscientiousness domain and -.05 with the 24 facets from all other

domains. Although the within-domain discriminant correlation was large, it was still typically lower

than the convergent correlation between this FFOCI subscale and its parent NEO PI-R facet (i.e., r =

.69). This pattern held for all but two of the FFOCI scales; the two exceptions being FFOCI

Perfectionism and Punctiliousness, both from conscientiousness.

Convergent Validity of FFOCI with Measures of OCPD and its Components

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
16
Table 4 reports the correlations between the FFOCI total score and four measures of OCPD as well

as the DAPP-BQ Compulsivity scale. The convergent correlations for the FFOCI were all significant

and large, ranging from.50 (PDQ-4) to .71 (WISPI-IV). In fact, the relevant scores from the WISPI-IV,

SNAP-2, MCMI-III, and DAPP-BQ correlated as highly, or higher, with the FFOCI as they did with

any of the other measures of OCPD. This suggests that that the FFOCI captures the construct encoded

in existing OCPD measures and can be considered a common ground among OCPD measures, which

often have displayed problematic levels of convergence (Widiger & Boyd, 2009).

Next we correlated each of the individual 12 FFOCI subscales with these same five measures. The

results in Table 5 indicate that the individual FFOCI subscales, like the combined FFOCI score, each

converged significantly with the established OCPD measures. Nonetheless, the magnitude of these

relationships varied across the FFOCI subscales. For example, FFOCI Perfectionism manifested

correlations that ranged from .43 (MCMI-III) to .64 (SNAP-2), yet FFOCI Constricted evinced

correlations ranging from .07 (MCMI-III) to .23 (SNAP-2). Overall, ten of the twelve subscales

evinced significant correlations with all five of the OCPD scales.

Incremental Validity of the FFOCI

The incremental validity of the FFOCI subscales was examined by testing whether each scale could

account for OCPD variance over and above that provided by its respective NEO PI-R facet scale. For

these analyses we employed a composite of the four DSM-IV-TR OCPD scales as the criterion. After

we standardized scores from the WISPI-IV, SNAP-2, PDQ-4, and MCMI-III to place them on the same

metric, we then averaged across these measures to produce the criterion. We entered the NEO PI-R

facet score in the first step of a hierarchical linear regression. These values are presented in the first

columns of Table 6. All but one of the NEO PI-R facets (i.e., openness to feelings) predicted a

significant portion of the variance in the criterion measure. We next added the relevant FFOCI subscale

in a second step to determine the increment of prediction it offered. As can be seen in Table 6, 11
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
17
FFOCI subscales accounted for a notable portion of the variance in the criterion, with R2Δ values

ranging from .12 (Excessive Worry) to .39 (Perfectionism). Nonetheless, the magnitude was quite small

for FFOCI Constricted (R2Δ = .03) suggesting it was not much better than the NEO PI-R facet of

openness to feelings for assessing OCPD.

Finally, the incremental validity of the FFOCI total score was examined to test whether it would

account for variance beyond established measures of OCPD. In order to avoid criterion overlap,

separate criteria were computed for each of four comparisons by averaging the standardized scores

from the other OCPD measures. For example, the criterion for the FFOCI comparison with the SNAP-2

was the mean of the standardized scores from the MCMI-III, WISPI-IV, and PDQ-4. Using these

criteria, the selected OCPD scale (e.g., SNAP-2) was entered in the first step of a hierarchical linear

regression with the FFOCI total score entered in the second step. Not surprisingly, the results in Table 7

indicate that each of the established OCPD measures accounted for significant portions of the variance

in the composite of the remaining measures (although the value for the MCMI-III was quite small at

.05). Most importantly, the FFOCI accounted for a significant increment over each of the OCPD

measures with R2Δ values ranging from .18 (beyond the WISPI-IV) to .52 (beyond the MCMI-III). In

each case, the total R2 was substantial and accounted for more than 50% of the variance in the criteria.

Discussion

The purpose of the present study was to examine the convergent, discriminant, and incremental

validity of the FFOCI and the results yielded promising support for its validity. For instance, each of

the twelve subscales displayed strong internal consistency, indicating they are homogenous in content.

Additionally, the FFOCI subscales correlated significantly with the corresponding facet scales from the

NEO PI-R, as well as with selected scales from other personality inventories. Importantly, these

convergent correlations were, for the most part, large in size, demonstrating that the FFOCI subscales
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
18
remain true to the FFM traits from which they were derived, despite their emphasis on aspects and

levels of the traits that are relevant to OCPD. In this regard, we suggest the FFOCI provides a “bridge”

between the general personality traits assessed by the NEO PI-R and the maladaptive traits encoded

within existing OCPD scales.

The relationship between FFM conscientiousness and OCPD has not always been substantial. The

first version of the DSM-5 dimensional trait model proposal included a domain of compulsivity that

Clark and Krueger (2010) and Krueger et al. (2011) suggested was distinct from FFM

conscientiousness, citing the meta-analysis of Saulsman and Page (2004). Saulsman and Page did

report correlations that were significant, but small. Nevertheless, a subsequent meta-analysis by Samuel

and Widiger (2008) indicated that the failure to confirm the hypothesis may have reflected limitations

of some existing measures of conscientiousness, not limitations within the conscientiousness construct

itself. Samuel and Widiger (2011) indicated strong support when FFM conscientiousness was

correlated with particular components of OCPD, such as Compulsivity as assessed by the DAPP-BQ;

Livesley & Jackson, 2009), and Workaholism and Propriety assessed by the SNAP (Clark et al., in

press). Support for the relationship of FFM conscientiousness and OCPD is also provided in the meta-

analytic study of O’Connor (2005), as well as the factor analyses of Clark, Livelsey, and Schroeder

(1996), Markon, Krueger, and Watson (2005), Shroeder, Wormworth, and Livesley (1992), and Watson,

Clark, and Chmielewski (2008). In the current study, the FFOCI scales assessing maladaptive variants

of FFM conscientiousness (i.e. Perfectionism, Fastidiousness, Punctiliousness, Workaholism,

Doggedness, and Ruminative Deliberation) all correlated highly with DAPP-BQ Compulsivity, as well

as with FFM conscientiousness as assessed by the NEO PI-R, the experimentally altered version of the

NEO PI-R, SNAP-2 Constraint (Clark et al., in press), and with HEXACO-PI Conscientiousness

(Ashton & Lee, 2008).

In addition to the convergent validity, each of the FFOCI subscales manifested significant
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
19
incremental validity over its corresponding facet scale from the NEO PI-R in accounting for variance in

a sum of four measures of OCPD. This particular finding is not surprising as the FFOCI subscales were

constructed to provide assessments of maladaptive variants of their respective NEO PI-R facets. In

addition, this is consistent with prior studies that have reported incremental validity of measures of

maladaptive personality functioning relative to the NEO PI-R’s assessment of general personality

functioning. For example, Reynolds and Clark (2001) reported that the maladaptive personality trait

scales from the SNAP-2 (e.g., Workaholism and Propriety) obtain incremental validity over the scales

from the NEO PI-R in accounting for PD symptomatology. As indicated by Reynolds and Clark,

however, these findings do not necessarily suggest that the FFM, per se, lacks adequate coverage of the

PD symptomatology. Instead, they merely suggest that the NEO PI-R lacks adequate coverage, which

is to be expected given that it was designed to provide an assessment of the normal variants of

personality functioning. They in fact suggested the development of new measures of the FFM that

would provide better fidelity for the assessment of the maladaptive traits.

The total score on the FFOCI (i.e., the sum of all 12 FFOCI subscales) also correlated highly with

and obtained incremental validity over each of the OCPD scales by accounting for additional variance

in a composite of three other measures of OCPD. These findings indicate that the effort to develop a

measure of obsessive-compulsive personality traits has, indeed, proven fruitful. Although each subscale

is brief, the FFOCI has a large number of subscales. In this regard, the FFOCI has a particular

advantage over existing OCPD scales in that it provides separate, homogenous subscales with which to

assess each of the specific components of OCPD. This is similar to the approach taken by the SNAP-2

(Clark et al., in press), which includes two subscales to assess components of OCPD (i.e., Workaholism

and Propriety). OCPD is a heterogeneous construct (Ansell et al., 2010) and without separate subscales

it will not always be clear why or how OCPD relates to external validators (Smith & Combs, 2010).

The FFOCI enables the disambiguation of the OCPD construct and allows researchers and clinicians to
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
20
assess specific traits of fastidiousness, workaholism, excessive worry, and/or Risk Aversion rather than

a global match to a monolithic yet heterogeneous construct. Finally, another potential advantage of the

FFOCI is its conceptual and empirical alignment with a broader dimensional model of general

personality structure, providing thereby a connection with a considerable body of construct validity

research (Widiger & Trull, 2007).

Areas for Future Study

The FFOCI Perfectionism subscale correlated less strongly with its corresponding NEO PI-R and

EXP-NEO facet, Competence, than did other FFOCI conscientiousness subscales with their respective

facets. Nevertheless, Perfectionism did correlate well with HEXACO PI-R Conscientiousness,

suggesting that the lower convergent validity of Perfectionism with NEO and EXP-NEO Competence

may reflect a limitation of the NEO PI-R itself, rather than of the FFOCI Perfectionism scale. It is

possible that the NEO PI-R does not express the idea of “competence” in a manner that can easily be

construed as perfectionism when taken to its maladaptive extreme. In fact, NEO PI-R Competence

appears to center on the idea of successfully “getting things done” and, as a result, tends to assess an

individual’s perceived ability to get things done rather than his or her general concern with or

orientation towards being competent and getting things done well. It may be that it is not competence

itself, but rather the desire for competence that may become warped to create something as maladaptive

as perfectionism. Similar distinctions have been made between the assessment of adaptive and

maladaptive perfectionism (Bieling, Israeli, & Antony, 2004; Flett & Hewitt, 2006; Page, Bruch, &

Haase, 2008). It will be of interest in future research to explore the relative relationship of the NEO PI-

R Conscientiousness facet scales (such as Competence) and FFOCI obsessive-compulsive personality

trait scales with measures of adaptive and maladaptive perfectionism, as well as additional measures of

OCPD traits, such as the Pathological Obsessive-Compulsive Personality Scale (Pinto, 2011), that also

aims to deconstruct OCPD into its components.

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
21
The FFOCI scales also varied in the extent of their correlation with measures of DSM-IV-TR

OCPD, likely reflecting that some FFM traits are more central to the historical construct and

conceptualization of OCPD than others (Samuel & Widiger, 2010). A scale that obtained somewhat

weak findings in this regard was the FFOCI Constricted subscale that assessed a maladaptive, extreme

variant of low Openness to Feelings. Although the Constricted subscale correlated strongly with the

NEO PI-R facet (r = -.78), its relationships with OCPD scales, including other scales from the FFOCI,

were mostly small and even non-significant in some cases. In addition, although the Constricted scale

did obtain a significant increment over the NEO PI-R facet for predicting the OCPD composite, this

effect size was also quite small. Taken together, it would appear that a trait defined by being closed to

one’s own feelings, whether assessed by the NEO PI-R or the FFOCI, is only moderately related to

existing measures of OCPD or its components. This, of course, does not necessarily suggest that such a

scale is not useful, but does indicate it is somewhat distinct from other aspects of the OCPD construct

and is not encoded within existing OCPD instruments. This particular FFM facet was nominated as

prototypic by the experts surveyed in Lynam and Widiger (2001) but only marginally identified as such

by the clinicians in Samuel and Widiger (2004). It will be useful for future research to determine

whether low openness to feelings is indeed a useful and/or valid component of a measure of obsessive-

compulsive personality traits.

More generally, the FFOCI assesses 12 trait components of OCPD. It is possible that only a small

subset of this list is really needed to provide a useful or valid assessment. For example, in the more

recent proposal for DSM-5, the diagnosis of OCPD includes only two traits: rigid perfectionism and

perseveration (APA, 2011). It will be of interest for future research to determine whether such

additional traits as workaholism, ruminative deliberation, constricted, detached coldness, dogmatism,

and/or risk aversion assessed by the FFOCI are really necessary for a valid assessment.

Finally, it might be informative in future research to utilize item-response theory analyses to

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
22
investigate the properties of the FFOCI items and/or scales. As the FFOCI scales are designed to assess

extreme and/or maladaptive variants of FFM traits (particularly as assessed by the NEO PI-R) one

might predict that FFOCI scales will have better fidelity for the extreme range of traits whereas the

respective NEO PI-R facet scales will have better fidelity for the lower range, as indicated in previous

FFM IRT studies by Samuel, Simms, Clark, Livesley, and Widiger (2010) and Stepp et al (in press).

However, in some cases a respective FFOCI subscale is perhaps best understood as assessing an OCPD

maladaptive variant of an FFM trait rather than a more extreme variant. For example, NEO PI-R low

Warmth is a scale that is also assessing a maladaptive trait of introversion (Haigler & Widiger, 2001).

FFOCI Detached Coldness might not be assessing a more extreme variant of low warmth, but simply

an OCPD variant of low warmth.

Limitations

One potential limitation of the current study was the use of online data collection. The latter does

not provide as much control over the validity of questionnaire completion as would be provided by a

group administration in a classroom setting or other uniform environment. This concern can be further

exacerbated by the number of items that the participants were asked to complete. However, several

factors offset these concerns. First, the participants were free to use as much time as needed to

complete the questionnaires. Second, an exclusion threshold was used to ensure that invalid protocols

were deleted. In addition, the findings were largely consistent with theoretical expectations, suggesting

that the results were not substantially negatively affected by random, careless responding. Indeed, a

number of recent studies attest to reliability and validity of data that are collected online (e.g., Wilt,

Condon, & Revelle, in press; Witt, Donnellan, & Orlando, 2011).

An additional potential limitation was that the data were collected within a student population. It

will be important to replicate the current findings within a clinical sample in which a number of

persons with OCPD are known to be present. Anticipating this concern, the authors of the current study
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
23
ensured that the PDQ-4 OCPD scale was administered to over 1,000 potential participants; inviting 100

of them with the very highest scores to participate in the study was intended to provide the student

sample with a sufficient range of OCPD symptomatology. The mean scores on measures of OCPD

included within the current study were comparable to mean scores obtained within clinical samples.

Finally, it is important to note that the validity evidence presented here is confined to other self-

report questionnaires. It will be important for future studies to go beyond this methodology and

compare the FFOCI against semi-structured interviews and informant reports of OCPD to determine

how well these scales predict specific problematic behaviors and important clinical outcomes, such as

psychosocial functioning. For example, one area of fruitful research might be the distinction between

adaptive and maladaptive career, work, and/or academic performance. Competence and achievement-

striving are unambiguously related to successful work, but traits such as perfectionism and

workaholism can be both advantageous and problematic (Bieling et al., 2004; Flett & Hewitt, 2006).

Conclusions

The FFOCI obtained strong convergent validity with existing measures of normal personality,

suggesting that the subscales provide robust assessments of the same FFM personality traits. However,

the FFOCI also correlates highly with, yet provide incremental validity over, established measures of

OCPD pathology. Taken together, these findings support the understanding of OCPD as a constellation

of maladaptive traits that are variants of general personality structure and suggest the FFOCI provides a

promising tool for assessing of these traits in clinical and research settings.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
24

References

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders. (3rd

ed., rev.). Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders. Text

Revision. (4th ed., rev.). Washington, DC: Author.

American Psychiatric Association. (June 21, 2011). Personality disorders. Retrieved from

http://www.dsm5.org/PROPOSEDREVISIONS/Pages/PersonalityandPersonalityDisorders.aspx

Ansell, E. B., Pinto, A., Crosby, R. D., Becker, D. F., Anez, L. M., Paris, M., & Grilo, C. M. (2010).

The prevalence and structure of obsessive-compulsive personality disorder in Hispanic

psychiatric outpatients. Journal of Behavior Therapy and Experimental Psychiatry, 41, 275-281.

Ashton, M. C., & Lee, K. (2008). The prediction of Honesty-Humility-related criteria by the

HEXACO and Five-Factor models of personality. Journal of Research in Personality, 42, 1216-

1228.

Baer, L. (1994). Factor analysis of symptom subtypes of obsessive compulsive personality disorder and

their relationship to personality and tic disorders. Journal of Clinical Psychiatry, 55, 18-23.

Bagby, R.M., & Farvolden, P. (2004). The Personality Diagnostic Questionnaire-4 (PDQ-4). In M.J.

Hilsenroth, D.L. Segal, & M. Hersen (Eds.), Comprehensive handbook of psychological

assessment, Volume 2. Personality assessment (pp. 122-133). NY: John Wiley.

Bieling, P. J., Israeli, A L., & Antony, M. M. (2004). Is perfectionism good, bad, or both? Examining

models of the perfectionism construct. Personality and Individual Differences, 36, 1373-1385.

Clark, L.A., Simms, L.J., Wu, K.D., & Casillas, A. (in press). Manual for the Schedule for Nonadaptive

and Adaptive Personality (SNAP-2). Minneapolis, MN: University of Minnesota Press.

Clark, L.A. (2007). Assessment and diagnosis of personality disorder: Perennial issues and an emerging
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
25
reconceptualization. Annual Review of Psychology, 58, 227-257.

Clark, L.A., & Krueger, R.F. (2010, February 10). Rationale for a six-domain trait dimensional

diagnostic system for personality disorder. Retrieved from

http://www.dsm5.org/ProposedRevisions/Pages/RationaleforaSix-

DomainTraitDimensionalDiagnosticSystemforPersonalityDisorder.aspx

Clark, L. A., & Livesley, W. J. (2002). Two approaches to identifying the dimensions of personality

disorder: Convergence on the five-factor model. In P.T. Costa & T.A. Widiger (Eds.),

Personality disorders and the five-factor model of personality (2nd ed., pp. 161-176).

Washington, DC: American Psychological Association.

Clark, L. A., Livesley, W. J., Schroeder, M.L., & Irish, S. L. (1996). Convergence of two systems for

assessing personality disorder. Psychological Assessment, 8, 294-303.

Clark, L. A. & Watson, D. (1995). Constructing validity: Basic issues in objective scale development.

Psychological Assessment, 7, 309-319.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.

Costa, P.T., & 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., & McCrae, R.R. (1995). Domains and facets: Hierarchical personality assessment using the

Revised NEO Personality Inventory. Journal of Personality Assessment, 64, 21-50.

Costa, P., Samuels, J., Bagby, M., Daffin, L., & Norton, H. (2005). Obsessive-compulsive

personality disorder: A review. In M. Maj, H. S. Akiskal, J. E. Mezzich, & A. Okasha (Eds.)

Personality Disorders. New York: Wiley and Sons.

Enders, C.K. (2006). A primer on the use of modern missing-data methods in psychosomatic medicine

research. Psychosomatic Medicine, 68, 427-736.

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
26
First, M. B., Bell, C. C., Cuthbert, B., Krystal, J. H., Malison, R., Offord, D. R., … & Wisner, K. L.

(2002). Personality disorders and relational disorders: A research agenda for addressing crucial

gaps in DSM. In D. J. Kupfer, M. B. First, & D. A. Regier (Eds.), A research agenda for DSM-V

(pp. 123-199). Washington, DC: American Psychiatric Association.

Flett, G. L., & Hewitt, P. L. (2006). Positive versus negative perfectionism in psychopathology:

A comment on Slade and Owens’s dual process model. Behavior Modification, 30, 472-495.

Grilo, C. M. (2004). Factor structure of DSM-IV criteria for obsessive-compulsive personality disorder

in patients with binge eating disorder. Acta Psychiatrica Scandinavica, 109, 64-69.

Haigler, E.D., & Widiger, T.A. (2001). Experimental manipulation of NEO PI-R items. Journal of

Personality Assessment, 77, 339-358.

Hummulen, B., Wilberg, T., Pedersen, G., & Karterud, S. (2008). The quality of the DSM-IV obsessive-

compulsive personality disorder construct as a prototype category. Journal of Nervous &

Mental Disease, 196, 446-455.

Klein, M.H., Benjamin, L.S., Rosenfeld, R., Treece, C., Husted, J., & Greist, J.H. (1993). The

Wisconsin Personality Disorders Inventory: I. Development, reliability, and validity. Journal of

Personality Disorders, 7, 285-303.

Krueger, R. F., & Eaton, N. R. (2010). Personality traits and the classification of mental disorders:

Toward a more complete integration in DSM-5 and am empirical model of psychopathology.

Personality Disorders: Theory, Research & Treatment, 2, 97-118.

Krueger, R. F., Eaton, N. R., Clark, L. A., Watson, D., Markon, K. E., Derringer, J., …Livesley, W. J.

(2011). Deriving an empirical structure of personality pathology for DSM-5. Journal of

Personality Disorders, 25, 170-191.

Livesley, W.J. (2003). Diagnostic dilemmas in classifying personality disorder. In K. A. Phillips, M. B.

First, & H. A. Pincus (Eds.) Advancing DSM: Dilemmas in psychiatric diagnosis (pp. 153-189).
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
27
Washington, DC: American Psychiatric Association.

Livesley, W.J., & Jackson, D. (2009). Manual for the Dimensional Assessment of Personality

Pathology—Basic Questionnaire. Port Huron, MI: Sigma Press.

Lynam, D.R., Gaughan, E.T., Miller, J.D., Miller, D.J., Mullins-Sweatt, S., & Widiger, T.A. (2011).

Assessing the basic traits associated with psychopathy: Development and validation of the

Elemental Psychopathy Assessment. Psychological Assessment, 23, 108-124.

Lynam, D.R., & Widiger, T.A. (2001). Using the five factor model to represent the DSM-IV personality

disorders: An expert consensus approach. Journal of Abnormal Psychology, 110, 401-412.

Markon, K. E., Krueger, R. F., & Watson, D. (2005). Delineating the structure of normal and abnormal

personality: An integrative hierarchical approach. Journal of Personality and Social

Psychology, 88, 139-157.

McCrae, R.R., & Costa, P.T. (2003). Personality in adulthood. A five factor theory perspective (2nd

ed.). NY: Guilford.

Millon, T. (1994). Millon Clinical Multiaxial Inventory-III. Manual. Minneapolis, MN: National

Computer Systems.

Mullins-Sweatt, S.N., Jamerson, J.E., Samuel, D.B., Olson, D.R., & Widiger, T.A. (2006).

Psychometric properties of an abbreviated instrument of the five-factor model. Assessment, 13,

119-137.

O’Connor, B. P. (2005). A search for consensus on the dimensional structure of personality disorders.

Journal of Clinical Psychology, 61, 323-345.

Page, J., Bruch, M. A., & Haase, R. F. (2008). Role of perfectionism and five-factor model traits in

career indecision. Personality and Individual Differences, 45, 811-815.

Pfohl, B. & Blum, N. (1995). Obsessive-compulsive personality disorder. In W.J. Livesley (Ed.) The

DSM-IV Personality Disorders (pp. 261-276). New York: Guilford.

FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
28
Pinto, A. (2011, March). Introducing the POPS (Pathological Obsessive-Compulsive Personality

Scale): Derivation and exploratory factor analysis. Annual meeting of the Society for

Personality Assessment. Cambridge, MA.

Pinto, A., Ansell, E. B., Grilo, C. M., & Shea, M.T. (2007). A multidimensional model of obsessive-

compulsive personality disorder. Paper presented at the Annual meeting of the American

Psychiatric Association, San Diego, California.

Reynolds, S. K., & Clark, L. A. (2001). Predicting dimensions of personality disorder from domains

and facets of the five-factor model. Journal of Personality, 69, 199-222.

Rodebaugh, T. L., Woods, C. M., & Heimberg, R. G. (2007). The reverse of social anxiety is not always

the opposite: The reverse-scored items of the social interaction anxiety scale do not belong.

Behavior Therapy, 38, 192-206.

Samuel, D. B., Simms, L. J., Clark, L. A., Livesley, W. J., & Widiger, T. A. (2010). An item response

theory integration of normal and abnormal personality scales. Personality disorders: Theory,

Research, and Treatment, 1, 5-21.

Samuel, D. B. & Widiger, T. A. (2004). Clinicians’ personality descriptions of prototypic personality

disorders. Journal of Personality Disorders, 18 (3), 286-308.

Samuel, D.B., & Widiger, T.A. (2008). A meta-analytic review of the relationships between the Five-

Factor Model and DSM-IV-TR personality disorders: A facet level analysis. Clinical Psychology

Review, 28, 1326-1342.

Samuel, D.B., & Widiger, T.A. (2010). A comparison of obsessive-compulsive personality

disorder scales. Journal of Personality Assessment, 92, 232-240.

Samuel, D.B., & Widiger, T.A. (2011). Conscientiousness and obsessive-compulsive

personality disorder. Personality Disorders: Theory, Research, and Treatment, 2, 161-174.

Saulsman, L.M., & Page, A.C. (2004). The five-factor model and personality disorder empirical
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
29
literature: a meta-analytic review. Clinical Psychology Review, 23, 1055-1085.

Schroeder, M. L., Wormworth, J. A., & Livesley, W. J. (1992). Dimensions of personality disorder and

their relationship to the Big Five dimensions of personality. Psychological Assessment, 4, 47-

53.

Smith, G. T., & Combs, J. (2010). Issues of construct validity in psychological diagnoses. In T. Millon,

R. F. Krueger, and E. Simonsen (Eds.), Contemporary Directions in Psychopathology: Scientific

Foundations of the DSM-V and ICD-11 (pp.205-222). New York: Guilford Press.

Smith, T. L., Klein, M. H., & Benjamin, L. S. (2003). Validation of the Wisconsin Personality Disorders

Inventory-IV with the SCID-II. Journal of Personality Disorders, 17(3), 173-187.

Stepp, S. D., Yu, L., Miller, J. D., Hallquist, M. N., Trull, T. J., & Pilkonis, P. A. (in press). Integrating

competing dimensional models of personality: Linking the SNAP, TCI, and NEO using item

response theory. Personality Disorders: Theory, Research, and Treatment.

Torgersen, S. (2009). The nature (and nurture) of personality disorders. Scandinavian Journal of

Psychology, 50, 624-632.

Trull, T.J., & Durrett, C.A. (2005). Categorical and dimensional models of personality disorder. Annual

Review of Clinical Psychology, 1, 355-380.

Watson, D., Clark, L. A., & Chmielewski, M. (2008). Structures of personality and their relevant to

psychopathology: II. Further articulation of a comprehensive unified trait structure. Journal of

Personality, 76, 1485-1522.

Widiger, T. A, & Boyd, S. (2009). Assessing personality disorders. In J. N. Butcher (Ed.), Oxford

handbook of personality assessment (3rd ed., pp. 336-363). New York: Oxford University Press.

Widiger, T.A. & Simonsen, E. (2005). Alternative dimensional models of personality disorder: Finding

a common ground. Journal of Personality Disorders, 19(2), 110-130.

Widiger, T.A., & Trull, T.J. (2007). Plate tectonics in the classification of personality disorder: Shifting
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
30
to a dimensional model. American Psychologist, 62, 71-83.

Widiger, T.A., Trull, T.J., Clarkin, J.F., Sanderson, C., & Costa, P.T. (2002). A description of DSM-IV

personality disorders with the five-factor model of personality. In P.T. Costa & T.A. Widiger

(Eds.), Personality disorders and the five-factor model of personality (2nd ed., pp. 89-99).

Washington, DC: American Psychological Association.

Wilt, J., Condon, D. M., & Revelle, W. (in press). Telemetrics and online data collection: Collecting

data at a distance. To appear in T. D. Little, N. Card, & B. Laursen (Eds.), Handbook of

Developmental Research Methods. New York: Guilford

Witt, E. A., Donnellan, M. B., & Orlando, M. J. (2011). Timing and selection effects within a

psychology subject pool: Personality and sex matter. Personality and Individual Differences, 50,

355-359.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
31
Table 1
Properties of FFOCI Scales from Validation Sample
Avg. Inter-item Correlations
m sd α CITC mean min max
Perfectionism (C1) 32.6 6.0 .84 .54 .35 .09 .68
Fastidiousness (C2) 31.4 6.8 .87 .60 .38 .19 .77
Punctiliousness (C3) 31.2 5.7 .80 .47 .28 .09 .58
Workaholism (C4) 29.8 6.2 .82 .52 .29 -.01 .65
Doggedness (C5) 31.6 6.2 .86 .57 .35 .03 .56
Ruminative Deliberation (C6) 31.1 6.6 .86 .56 .37 .07 .61
Detached Coldness (E1) 24.8 6.5 .84 .54 .31 .08 .73
Risk Aversion (E5) 26.9 6.8 .87 .58 .41 .17 .65
Excessive Worry (N1) 35.0 7.4 .87 .60 .42 .14 .72
Constricted (O3) 25.4 6.5 .85 .55 .37 .03 .66
Inflexibility (O4) 26.1 5.7 .77 .45 .25 .02 .52
Dogmatism (O6) 25.2 6.1 .83 .51 .32 .04 .64
Notes: FFOCI = Five-Factor Obsessive Compulsive Inventory; Avg. CITC = Averaged Corrected Item-total
correlations. n = 203
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
32

Table 2
Intercorrelations Among the Final FFOCI scales in the Validation Sample (n = 203)
C1 C2 C3 C4 C5 C6 E1 E5 O3 O4 O6
Perfectionism (C1)
Fastidiousness (C2) .73**
Punctiliousness (C3) .62** .69**
Workaholism (C4) .72** .63** .70**
Doggedness (C5) .62** .59** .67** .74**
Ruminative Deliberation (C6) .60** .68** .64** .67** .64**
Detached Coldness (E1) .23** .35** .36** .36** .34** .42**
Risk Aversion (E5) .38** .51** .54** .47** .46** .68** .48**
Constricted (O3) .10 .13 .14* .21** .22** .20** .61** .21**
Inflexibility (O4) .44** .55** .57** .56** .50** .57** .59** .66** .32**
Dogmatism (O6) .41** .45** .65** .44** .48** .47** .41** .50** .31** .51**
Excessive Worry (N1) .32** .35** .29** .31** .15* .40** .19** .32** -.09 .41** .20**
Notes: FFOCI = Five Factor Obsessive Compulsive Inventory.
** = p < .01 (2-tailed), * = p < .05 (2-tailed)
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
33
Table 3
Convergent and discriminant validity of the FFOCI subscales with measures of general personality.
FFOCI Subscales
Other
Measures (C1) (C2) (C3) (C4) (C5) (C6) (E1) (E5) (N1) (O3) (O4) (O6)
NEO faceta .45** .74** .51** .69** .77** .76** -.74** -.68** .82** -.78** -.53** -.62**
Disc Sameb .49 .46 .49 .51 .59 .45 -.46 -.36 .49 -.26 -.22 -.31
Disc Otherc -.01 -.04 -.06 -.05 .11 -.08 -.07 .06 -.06 -.15 .01 .01
Exp Cd .52** .76** .55** .69** .72** .70**
HEX Ce .66** .75** .66** .69** .72** .69**
SNAP-Cf .41** .50** .55** .56** .51** .60**
Note. n = 203. FFOCI=Five-Factor Obsessive-Compulsive Inventory; FFOCI subscales are denoted as follows: C1:
Perfectionism, C2: Fastidiousness, C3: Punctiliousness, C4: Workaholism, C5: Doggedness, C6: Ruminative Deliberation,
E1: Detached Coldness, E5: Risk Aversion, N1: Excessive Worry, O3: Constricted, O4: Inflexibility, O6: Dogmatism.
a
Corresponding NEO Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992) facet for each FFOCI subscale;
b
Discriminant validity between the FFOCI and the average correlation of non-corresponding NEO PI-R facets within the
same domain; cDiscriminant validity between the FFOCI and the average correlation of non-corresponding NEO PI-R
facets outside of each subscale’s domain; dCorresponding Experimental NEO PI-R (Haigler & Widiger, 2001)
Conscientiousness facet; eTotal HEXACO PI-R Conscientiousness scale (Ashton & Lee, 2008); fSNAP-2 = Schedule for
Nonadaptive and Adaptive Personality, Constraint scale (Clark et al., in press).
*p < .05, **p < .01
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
34

Table 4
Convergence of FFOCI and OCPD Scales
MCMI-III PDQ-4 SNAP-2 WISPI-IV DAPP-BQ
FFOCI .58** .50** .66** .71** .66**
MCMI-III .00 .27** .30** .53**
PDQ-4 .67** .61** .38**
SNAP-2 .63** .58**
WISPI-IV .60**
Notes. n = 203. FFOCI = Five Factor Obsessive Compulsive Inventory; OCPD = Obsessive
Compulsive Personality Disorder; MCMI-III = Millon Clinical Multiaxial Inventory - III
(Millon, 1994); PDQ-4 = Personality Diagnostic Questionnaire (Bagby & Farvolden, 2004);
SNAP-2 = Schedule for Nonadaptive and Adaptive Personality (Clark et al., in press); WISPI-
IV = Wisconsin Personality Inventory (Klein et al., 1993); DAPP-BQ (Livesley &
Jackson, 2009).
*p < .05, **p < .01
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
35

Table 5
Convergent Validity of FFOCI Subscales with OCPD and Related Measures
FFOCI Subscales

Other Measures C1 C2 C3 C4 C5 C6 E1 E5 N1 O3 O4 O6
MCMI-III .43** .50** .53** .52** .62** .60** .26** .59** .09 .07 .35** .36**
PDQ-4 .48** .44** .38** .39** .22** .35** .26** .23** .48** .14* .39** .40**
SNAP-2 .64** .56** .52** .54** .49** .49** .33** .41** .41** .23** .51** .47**
WISPI-IV .62** .63** .58** .59** .49** .60** .33** .46** .42** .14* .57** .53**

DAPP-BQ .70** .77** .59** .61** .60** .60** .21** .43** .29** .02 .38** .41**
Note. n = 203. FFOCI=Five-Factor Obsessive-Compulsive Inventory; FFOCI subscales are denoted as follows: C1:
Perfectionism, C2: Fastidiousness, C3: Punctiliousness, C4: Workaholism, C5: Doggedness, C6: Ruminative
Deliberation, E1: Detached Coldness, E5: Risk Aversion, N1: Excessive Worry, O3: Constricted, O4: Inflexibility,
O6: Dogmatism. MCMI=Millon Clinical Multiaxial Inventory III (Millon, 1994); PDQ-4 = Personality Diagnostic
Questionnaire (Bagby & Farvolden, 2004); SNAP-2 = Schedule for Nonadaptive and Adaptive Personality-2 (Clark et
al., in press); WISPI-IV=Wisconsin Personality Inventory (Klein et al., 1993); DAPP-BQ = Dimensional Assessment
of Personality Pathology-Basic Questionnaire Compulsivity scale (Livesley and Jackson, 2009).
*p < .05, **p < .01.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
36
Table 6
Incremental Validity of FFOCI Subscales over NEO PI-R Facets for Predicting
OCPD Composite
Step 1 Step 2
NEO PI-R β R2 NEO PI-R β FFOCI β R2 ∆
C1 .37** .14** .06 .70** .39**
C2 .51** .26** -.04 .74** .25**
C3 .41** .17** .10 .62** .29**
C4 .55** .30** .14* .58** .18**
C5 .38** .15** -.22* .78** .24**
C6 .58** .34** .15 .57** .14**
E1 -.20** .04** .20* .54** .13**
E5 -.36** .13** .05 .60** .19**
N1 .33** .11** -.16 .59** .12**
O3 -.10 .01 .12 .29* .03*
O4 -.44** .19** -.16* .52** .20**
O6 -.38** .15** -.03 .56** .20**
Note. n = 203. FFOCI = Five-Factor Obsessive-Compulsive Inventory; OCPD Composite =
The mean standardized scores of the OCPD scales from the Millon Clinical Multiaxial
Inventory - III (Millon, 1994), Schedule for Nonadaptive and Adaptive Personality (Clark et
al, in press), Wisconsin Personality Inventory (Klein et al., 1993), and Personality Diagnostic
Questionnaire (Bagby & Farvolden, 2004). FFOCI subscales are denoted as follows: C1:
Perfectionism, C2: Fastidiousness, C3: Punctiliousness, C4: Workaholism, C5: Doggedness,
C6: Ruminative Deliberation, E1: Detached Coldness, E5: Risk Aversion, N1: Excessive
Worry, O3: Constricted, O4: Inflexibility, O6: Dogmatism.
* p < .05, ** p < .01.
FIVE-FACTOR OBSESSIVE-COMPULSIVE INVENTORY
37

Table 7
Incremental validity of FFOCI Over Established OCPD Scales
SNAP-2 MCMI-III PDQ-4 WISPI-IV
2 2
β R∆ β R∆ β R2∆ β R2∆
Step 1 .51** .05** .30** .49**
OCPDa .71** .22** .55** .70**

Step 2 .21** .52** .43** .18**

OCPDa .31** -.30** .18** .28**
FFOCIb .61** .89** .75** .59**

Total R2 .71** .57** .73** .66**

Note. Values down each column indicate the comparison of the FFOCI with the scale
listed on the column header. Criterion measure for each analysis was the mean of the
standardized scores from the remaining OCPD scales. FFOCI = Five-Factor Obsessive-
Compulsive Inventory; MCMI=Millon Clinical Multiaxial Inventory - III (Millon,
1994); SNAP-2 = Schedule for Nonadaptive and Adaptive Personality (Clark et al, in
press); WISPI-IV = Wisconsin Personality Inventory (Klein et al., 1993); PDQ-4 =
Personality Diagnostic Questionnaire (Bagby & Farvolden, 2004). a Total score of the
individual OCPD scale listed at the top each column; bFFOCI total score.
* p < .05, ** p < .01.