Content uploaded by Shannon Sauer
Author content
All content in this area was uploaded by Shannon Sauer
Content may be subject to copyright.
http://asm.sagepub.com
Assessment
DOI: 10.1177/1073191107313003
2008; 15; 329 originally published online Feb 29, 2008; Assessment
Danielle Duggan and J. Mark G. Williams
Ruth A. Baer, Gregory T. Smith, Emily Lykins, Daniel Button, Jennifer Krietemeyer, Shannon Sauer, Erin Walsh,
Samples
Construct Validity of the Five Facet Mindfulness Questionnaire in Meditating and Nonmeditating
http://asm.sagepub.com/cgi/content/abstract/15/3/329
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
can be found at:Assessment Additional services and information for
http://asm.sagepub.com/cgi/alerts Email Alerts:
http://asm.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://asm.sagepub.com/cgi/content/refs/15/3/329
SAGE Journals Online and HighWire Press platforms):
(this article cites 40 articles hosted on the Citations
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
Construct Validity of the Five Facet
Mindfulness Questionnaire in Meditating
and Nonmeditating Samples
Ruth A. Baer
Gregory T. Smith
Emily Lykins
Daniel Button
Jennifer Krietemeyer
Shannon Sauer
Erin Walsh
University of Kentucky
Danielle Duggan
J. Mark G. Williams
University of Oxford
Previous research on assessment of mindfulness by self-report suggests that it may include
five component skills: observing, describing, acting with awareness, nonjudging of inner
experience, and nonreactivity to inner experience. These elements of mindfulness can be
measured with the Five Facet Mindfulness Questionnaire (FFMQ). The authors investi-
gated several aspects of the construct validity of the FFMQ in experienced meditators and
nonmeditating comparison groups. Consistent with predictions, most mindfulness facets
were significantly related to meditation experience and to psychological symptoms and
well-being. As expected, relationships between the observing facet and psychological
adjustment varied with meditation experience. Regression and mediation analyses showed
that several of the facets contributed independently to the prediction of well-being and
significantly mediated the relationship between meditation experience and well-being.
Findings support the construct validity of the FFMQ in a combination of samples not
previously investigated.
Keywords: self-report assessment; mindfulness; meditation; psychological well-being
The development of tools for the assessment of mind-
fulness is critically important for several reasons. Although
the empirical literature on the efficacy of mindfulness-based
interventions has been growing rapidly (Baer, 2003;
Grossman, Neimann, Schmidt, & Walach, 2004; Hayes,
Masuda, Bissett, Luoma, & Guerrero, 2004; Robins &
This project was supported by a grant to Ruth Baer from the College of Arts and Sciences at the University of Kentucky. The
authors thank their participants for their generous contributions to this research. Correspondence concerning this article may be
addressed to Ruth A. Baer, PhD, Department of Psychology, 115 Kastle Hall, University of Kentucky, Lexington, KY 40506-0044;
e-mail: rbaer@email.uky.edu.
Assessment, Volume 15, No. 3, September 2008 329-342
DOI: 10.1177/1073191107313003
© 2008 Sage Publications
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
Chapman, 2004; Salmon et al., 2004), fewer studies have
examined the mechanisms or processes by which the prac-
tice of mindfulness leads to beneficial outcomes. Several
authors have suggested that clarifying these processes
requires psychometrically sound methods for assessing
mindfulness (Baer, Smith, & Allen, 2004; Bishop et al.,
2004; Brown & Ryan, 2004; Dimidjian & Linehan, 2003a).
Such measures are necessary for examining whether indi-
viduals who practice mindfulness become more mindful
over time and whether these changes mediate the effects
of mindfulness training on psychological health.
Recently developed measures of mindfulness include
the Freiburg Mindfulness Inventory (FMI; Buchheld,
Grossman, & Walach, 2001), the Mindful Attention
Awareness Scale (MAAS; Brown & Ryan, 2003), the
Kentucky Inventory of Mindfulness Skills (KIMS; Baer
et al., 2004), the Cognitive and Affective Mindfulness
Scale (CAMS; Feldman, Hayes, Kumar, Greeson, &
Laurenceau, 2007), and the Mindfulness Questionnaire
(MQ; Chadwick, Hember, Mead, Lilley, & Dagnan, 2005).
All use self-report methods to assess a general tendency
to be mindful in daily life and have shown promising psy-
chometric characteristics. However, differences in their
content and structure suggest a lack of consensus among
researchers about whether mindfulness should be con-
ceptualized as a multifaceted construct, and if so, how the
facets should be defined and operationalized. For example,
Brown and Ryan’s (2003, 2004) work suggests that mind-
fulness consists of a single factor, which they describe as
awareness of and attention to present events and experi-
ences. Accordingly, the MAAS is a unidimensional instru-
ment yielding a single total score. In contrast, the KIMS
(Baer et al., 2004) is based largely on dialectical behav-
ioral therapy (DBT) in which mindfulness is conceptual-
ized as a set of interrelated skills (Dimidjian & Linehan,
2003b). The KIMS provides subscale scores for each of
four mindfulness skills (observing, describing, acting
with awareness, and accepting without judgment). The
other instruments are designed to capture multiple com-
ponents of mindfulness (including attention, awareness,
openness, letting go, nonjudging, acceptance, and non-
aversion) but do not measure them separately and yield
only a total score.
Several authors have argued that assessment of complex
constructs at the facet level is essential for clarifying their
relationships with other variables (Schneider, Hough, &
Dunnette, 1996; Smith et al., 2007; Smith, Fischer, &
Fister, 2003; Smith & McCarthy, 1995). Investigating
facets of mindfulness is likely to improve our under-
standing of the specific skills that are cultivated through
the practice of mindfulness and how these are related to
psychological adjustment. In a recent study of facets of
mindfulness, Baer, Smith, Hopkins, Krietemeyer, and
Toney (2006) conducted exploratory factor analysis in a
large sample (N = 613) of students, who had completed
all five of the mindfulness questionnaires just described.
This analysis allowed items from different instruments to
combine to form factors, providing an empirical integra-
tion of these independent attempts to operationalize
mindfulness. Findings suggested a five-factor solution.
Observing includes noticing or attending to internal and
external experiences, such as sensations, cognitions,
emotions, sights, sounds, and smells. Describing refers to
labeling internal experiences with words. Acting with
awareness includes attending to one’s activities of the
moment and can be contrasted with behaving mechani-
cally while attention is focused elsewhere (often called
automatic pilot). Nonjudging of inner experience refers
to taking a nonevaluative stance toward thoughts and
feelings. Nonreactivity to inner experience is the ten-
dency to allow thoughts and feelings to come and go,
without getting caught up in or carried away by them.
Scales for each of these five factors were created by
selecting the seven or eight items with the highest loadings
on their respective factors (and low loadings on all other
factors). These were combined to form the 39-item Five
Facet Mindfulness Questionnaire (FFMQ; Baer et al.,
2006). Example items can be seen in Table 1. The five
facet scales demonstrated adequate to good internal con-
sistency, with alpha coefficients ranging from .75 to .91,
and relationships between the facet scales and other vari-
ables were consistent with predictions in most cases.
However, some unexpected findings for the observing
facet were noted. For example, hierarchical confirmatory
factor analysis (CFA) examined whether the five facets
should be viewed as elements of a general mindfulness
construct or are better understood as five separate con-
structs. Four of the facets were found to be clear indica-
tors of an overarching mindfulness construct. However,
the observing facet did not fit this model. This finding
was surprising, because observing is typically described
330 ASSESSMENT
TABLE 1
Example Items for Mindfulness Facets
Facet Example Item
Observing I notice the smells and aromas of things.
Describing I am good at finding words to describe my
feelings.
Acting with I find myself doing things without paying
awareness attention. (R)
Nonjudging of inner I think some of my emotions are bad or
experience inappropriate and I should not feel them. (R)
Nonreactivity to inner I perceive my feelings and emotions
experience without having to react to them.
NOTE: R = reverse-scored item (higher scores represent higher levels
of mindfulness).
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
as a central element of mindfulness. In addition, because
mindfulness training consistently yields positive outcomes,
mindfulness facets were expected to be positively corre-
lated with adaptive characteristics (e.g., openness to
experience, emotional intelligence) and negatively corre-
lated with maladaptive variables (e.g., thought suppression,
experiential avoidance). Although this was true in most
cases, the observing facet was unexpectedly found to be
modestly but positively correlated with several maladap-
tive constructs, including dissociation, absentmindedness,
psychological symptoms, and thought suppression.
Earlier findings reported by Baer et al. (2004) in the
development of the KIMS had suggested that the observ-
ing facet may operate differently in samples with and
without meditation experience. To examine this possibil-
ity, Baer et al. (2006) selected from their student samples
those participants reporting some degree of experience
with meditation (N = 190). In this subsample, CFA showed
good fit for a hierarchical model in which all five facets
are elements of mindfulness. In addition, correlations
between observing and maladaptive constructs were no
longer unexpectedly positive but were nonsignificant.
The observing facet was the only facet to show a different
pattern of findings in the meditating subsample compared
to the total sample.
These results suggested that the observing facet may be
sensitive to changes with meditation practice that alter its
relationships with other variables. This idea is plausible in
light of the literature on self-focused attention, which has
been defined as the awareness of internally generated stim-
uli such as sensations, cognitions, and emotions (Ingram,
1990). Recent reviews suggest that self-focused attention
can be maladaptive. It has been associated with negative
emotion in clinical and nonclinical samples (Mor &
Winquist, 2002) and is common in many psychological
disorders (Harvey, Watkins, Mansell, & Shafran, 2004).
Mindfulness training includes close observation of internal
stimuli, but teaches participants to observe them with an
accepting, nonjudging, and nonreactive stance, even if
they are unpleasant. Responding in these ways to negative
thoughts and feelings appears to be uncommon in Western
culture (Hayes, Strosahl, & Wilson, 1999; Segal, Williams,
& Teasdale, 2002). Thus, close observation of internal
experience may be maladaptive in the general population
but adaptive when it is done mindfully. Because learning
to observe internal stimuli mindfully may require more
meditation experience than is typically found in student
samples, individuals with more extensive histories of
meditation practice should be studied.
The purpose of this project, therefore, was to investi-
gate several important aspects of the construct validity of
the FFMQ that previous work (Baer et al., 2006) could
not address because of its reliance on student samples
with little meditation experience. In the Buddhist tradi-
tions from which current mindfulness practices originate,
it is widely believed that the long-term practice of medi-
tation cultivates mindfulness skills and that these skills
promote psychological well-being (PWB; Goldstein &
Kornfield, 1987; Kabat-Zinn, 2003; Walsh & Shapiro,
2006). We tested this general idea by examining relation-
ships between meditation experience, FFMQ scores, and
psychological adjustment in experienced meditators and
comparison groups of nonmeditating individuals. Several
hypotheses were tested. We predicted that mindfulness
facet scores would be positively correlated with medita-
tion experience and that meditators would score higher
than nonmeditators. We also predicted that in general,
mindfulness facet scores would be negatively associated
with psychological symptoms and positively associated
with well-being. However, based on previous findings
(Baer et al., 2004, 2006), we also expected that correla-
tions for the observing facet would vary with meditation
experience. In addition, we predicted that the hierarchical
five-factor structure of mindfulness facets reported by
Baer et al. (2006) for individuals with meditation experi-
ence would be replicated. Finally, we expected that mind-
fulness facets would show incremental validity in the
prediction of PWB and that increases in mindfulness skills
would mediate a positive relationship between meditation
experience and well-being.
METHOD
Participants and Procedures
We collected data from four samples of participants,
including regular meditators, demographically similar
nonmeditators, a general community sample, and a student
sample.
Regular meditators. A sample of individuals with a
currently ongoing regular meditation practice were
recruited in several ways. All individuals (N = 278) who
had attended an international conference on mindfulness
at the University of Massachusetts Medical School in
2005 were mailed a packet containing a cover letter invit-
ing them to participate in a study of mindfulness, medi-
tation, and psychological functioning, a demographic
form, the FFMQ, measures of several other constructs
(some described later, others for a separate project), and
a return envelope. No compensation was offered, other
than a teabag included as a token of appreciation. A total
of 132 packets were completed and returned, for a response
rate of 48%. Of these, 119 reported a currently ongoing
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 331
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
regular meditation practice (at least once or twice per
week) and therefore were included in data analyses. The
others were excluded because they did not respond to this
question (N = 2), had never meditated regularly (N = 3),
or had discontinued a regular meditation practice (N = 8).
Potential differences between individuals who have main-
tained or discontinued a regular meditation practice were
beyond the scope of this project.
An additional 61 regular meditators with an ongoing
practice were recruited using announcements or flyers
that were posted to listservs and other Internet-based
groups focused on mindfulness or meditation, distributed
in mindfulness meditation and yoga centers in Lexington,
KY, and other cities in the Midwest, or posted in the
Lexington community. Interested individuals were asked
to contact one of the experimenters to request a packet.
Approximately 60% of those who expressed interest and
were sent a packet completed and returned it. These
participants also were offered only a teabag (included in
the packet) as compensation.
As described in the following paragraphs, additional
regular meditators were identified through recruitment of
the other groups and added to the meditating sample.
Demographically similar nonmeditators. Many of the
experienced meditators just described held graduate
degrees and some reported working in a mental health
field. We used several strategies to recruit a group of non-
meditating individuals who were demographically similar
to the meditating sample. Some of these participants were
offered no compensation and were asked to complete only
the FFMQ and a demographic questionnaire. These forms
were mailed, along with a cover letter, to all individuals
listed in the University of Kentucky telephone directory
under several departments chosen to represent a broad
range of fields, including mental health disciplines (coun-
seling, psychiatry, social work) and other fields (natural
sciences, humanities, journalism, law, and music). These
listings include faculty and professional or administrative
staff (but not students), ensuring that most respondents are
highly educated. Responses were returned by 197 (35%)
individuals, 54 (27%) of whom worked in mental health
fields. Of these 197, 174 reported never having engaged in
regular meditation. The remaining 23 (12%) respondents
reported currently engaging in a regular meditation prac-
tice and therefore were coded as regular meditators and
added to the previously described meditating group.
Other highly educated but nonmeditating individuals
(some in mental health fields) were offered $50 to com-
plete a longer packet of questionnaires containing many
additional measures (some described later and others for a
separate project). To recruit nonmeditating mental health
professionals, letters were mailed to clinicians listed in the
local telephone directory and distributed to professional
staff mail boxes in mental health clinics and hospitals.
Letters stated that individuals who had never practiced
meditation on a regular basis were eligible to participate
(although those who had tried it once or a few times were
eligible). We also stipulated that those using DBT or accep-
tance and commitment therapy (ACT) in their professional
work were not eligible to participate. Although DBT and
ACT do not require a regular meditation practice, mind-
fulness-related concepts and exercises are central to these
interventions. Whether the effects of working with these
are similar to the effects of regular meditation practice is
unknown. To recruit nonmeditating non–mental health
professionals, similar letters (also offering $50 for com-
pletion of a packet of questionnaires) were sent to faculty
and staff (excluding psychology and social work depart-
ments) of a local college. Individuals with undergraduate
or graduate degrees in disciplines other than mental health,
and who had never engaged in a regular meditation prac-
tice or worked in a mental health field, were eligible to par-
ticipate. In response to these recruitment letters, 50 mental
health professionals and 48 individuals in other fields
requested and were sent packets. Completed packets were
returned by 40 mental health professionals (80%) and 41
others (85%). Of these 81 participants, 2 (both mental
health professionals) unexpectedly reported currently
engaging in a regular meditation practice and therefore
were coded as regular meditators. Another respondent
reported using DBT in her work, and her data were
excluded from analyses. This left 78 paid participants who
were combined with the 174 unpaid participants previ-
ously described, bringing the total for the highly educated
nonmeditating group to 252. Data from all 252 of these
participants were used for analyses of relationships
between mindfulness and demographic variables and for
group comparisons of mindfulness scores. Data from the
78 paid participants, who had completed the full packet of
measures, were used in analyses involving symptoms and
well-being.
Nonmeditating community sample. Adults with a level
of education more typical of the general population were
recruited from a community sample of volunteers in the
United Kingdom, who had participated in a previous
questionnaire study and had given permission to be con-
tacted again for participation in other research. Packets
containing a brief measure of psychological symptoms
(described later) and the FFMQ (demographic data had
been obtained previously) were mailed to 659 individu-
als. No compensation was offered. Completed packets
were returned by 313 individuals, for a response rate of
47.5%. Of these, 293 reported never having engaged in a
regular meditation practice and were therefore included
332 ASSESSMENT
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
in data analyses. The others were excluded for having
discontinued a regular meditation practice (N = 14), for
providing an unclear response or no response to this
question (N = 5), or for engaging in transcendental rather
than mindfulness meditation (N = 1; this distinction is
discussed later).
Nonmeditating student sample. Undergraduate students
in psychology classes at the University of Kentucky were
recruited using an online registration system for partici-
pation in research. Data from this sample have also been
reported in Baer et al. (2006). They received credit in
their classes in exchange for participation. In 1-hour
group sessions in a university classroom, 269 students
completed the FFMQ, a demographic form, and a packet
of other measures. Eight reported engaging in the regular
practice of mindfulness meditation and were added to the
meditating group. Two others reported practicing tran-
scendental meditation and were not included in analyses,
leaving 259 to make up the nonmeditating student sample
reported here.
With the additional meditators identified in recruiting
the other groups, the final meditating sample included
213 individuals (119 from the mindfulness conference,
61 from listservs and flyers, 25 from university faculty
and other professionals, and 8 students). All these had
been asked to exclude practices such as yoga, tai chi, chi
gong, and prayer when describing their experience with
meditation. The more traditional forms of meditation
often are divided into two general categories (Kabat-
Zinn, 1982): concentration-based practices (such as tran-
scendental meditation) and Buddhist-based approaches.
According to Kabat-Zinn (2003), mindfulness is “the
fundamental stance underlying all streams of Buddhist
meditative practice” (p. 146) and is the core teaching in the
many variations of the Buddhist tradition. Thus, Buddhist-
based practices are more likely to cultivate mindfulness
skills. In our meditating sample, 86% identified their pri-
mary practice as Buddhist-based, whereas 14% inadver-
tently had not been asked to discriminate between
concentrative and Buddhist-based and therefore could
have been engaged in either type. Most of this subgroup
came from the university faculty recruits and had com-
pleted only the FFMQ and demographic form. Analyses
involving these variables (described later) were con-
ducted both with and without this subgroup, and the pat-
tern of findings was identical. Therefore, to increase both
the size and the diversity of sources for the meditating sam-
ple, these participants were retained.
Participants’ demographic characteristics can be seen
in Table 2. All samples were largely White (race was not
reported in the community sample) and included more
females than males. The student sample had a significantly
lower proportion of males and were significantly younger
than in the other samples. Mean education levels for the
student and community samples were similar (about 13
years or slightly more than a high school education).
Education levels for the other two samples were signifi-
cantly higher. The meditating sample had a significantly
higher proportion of mental health professionals. Of the
experienced meditators, 80% resided in the United States
and 20% in other countries—primarily Canada, the United
Kingdom, or other Western European countries. Among
the meditators, there were no significant differences in
mindfulness scores between residents of the United States
and residents of other countries. When controlling for
demographic variables, there were no significant differ-
ences in mindfulness scores between meditators recruited
from the conference and those recruited in other ways.
Among the highly educated nonmeditators, there were no
significant differences in mindfulness scores related to
receipt of monetary compensation for participating in the
study. Other relationships between mindfulness scores
and demographic variables are described later.
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 333
TABLE 2
Demographic Characteristics of Four Samples of Participants
Students
a
Community
a
Highly Educated
a
Meditators
N 259 293 252 213
Age in years
M (SD) 18.9 (3.2) 49.5 (6.7) 44.2 (11.9) 48.8 (12.9)
Range 18-53 34-66 22-71 18-83
% Male 22 40 42 32
% White 92 — 91 94
Years of education
M (SD) 13.4 (0.8) 12.9 (3.2) 18.2 (2.1) 18.3 (1.8)
% MH profl 0 — 34 63
NOTE: M = mean; SD = standard deviation; MH profl = mental health professional.
a. nonmeditating.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
For participants in the meditating sample, characteris-
tics of their meditation practice are shown in Table 3.
Most of these participants (N = 180) were asked how long
they had meditated regularly, how many times per week
and for how many minutes each time they typically med-
itate, and how many days they had spent on meditation
retreats. Participants recruited from sources expected to
be primarily nonmeditating were asked whether they had
ever meditated and, if regularly, for how long they had
done so. These data show that the participants in the med-
itating sample had extensive experience. Nearly half
(45%) had meditated regularly for more than 10 years
and only 8% had done so for less than 1 year. Substantial
majorities reported meditating between three and six
times per week, for 21-45 minutes each time. More than
60% had spent more than 10 total days on meditation
retreats, with 44% reporting over 30 such days.
Measures
Mindfulness. All participants completed the FFMQ
(Baer et al., 2006), which assesses five facets of a general
tendency to be mindful in daily life: observing, describing,
acting with awareness, nonreactivity to inner experience,
and nonjudging of inner experience. Items are rated on a
5-point Likert-type scale ranging from 1 (never or very
rarely true) to 5 (very often or always true).
Symptoms. The student sample completed the Brief
Symptom Inventory (BSI; Derogatis, 1992), which
includes 53 items and measures a range of psychological
symptoms and somatic complaints. The community sam-
ple completed a version of the Profile of Mood States
(POMS; McNair, Lorr, & Droppelman, 1971) containing
24 items measuring several types of negative affect and
bodily symptoms. The highly educated nonmeditators
who had been paid to complete the longer packet (N = 78)
completed the Depression Anxiety Stress Scales (DASS;
Lovibond & Lovibond, 1995), as did 61 of the regular
meditators. (To reduce response burden for the meditators
who were not compensated, some measures were
included in only a subset of packets.) The DASS includes
42 items measuring negative affect and bodily symptoms.
For all three of these instruments, participants provide
Likert-type scale ratings of their symptoms over the last
week. We used the total score for all instruments (on the
POMS, vigor items were reverse-scored). These measures
are widely used and have good psychometric properties
(Crawford & Henry, 2003; Derogatis, 1992; Lovibond &
Lovibond, 1995; Nulty, Wilkins, & Williams, 1987).
PWB. Meditation traditions suggest that, in addition to
the reduction of symptoms, the practice of mindfulness
cultivates positive qualities such as wisdom and compas-
sion (Shapiro, Schwartz, & Santerre, 2002). Similarly,
much current thinking in psychology suggests that psy-
chological health is broader than the absence of symptoms
(Hayes et al., 1999; Keyes, 2007; Snyder & Lopez, 2002).
After reviewing many theories of psychological health,
Ryff (1989) conceptualized PWB as having six elements:
self-acceptance (positive attitude toward one’s self, life,
and past, including good and bad qualities), positive rela-
tions with others (warm, satisfying, trusting relationships),
autonomy (independence, ability to resist social pressures
and follow own standards), environmental mastery (com-
petence in managing life’s demands), purpose in life (goals
and direction, sense of meaning in life), and personal
growth (view of self as growing and developing, openness
to new experiences). The scales of PWB measure these
elements and are available in several lengths. We used the
54-item version, which includes 9 items per scale and
has good psychometric properties (Ryff, 1989). We used a
total score, summing the six elements of well-being. Most
of the meditators and demographically similar nonmedita-
tors completed this measure. Because the student sample
described earlier had not completed it, we used a different
student sample (N = 175) that had completed both the
PWB and the FFMQ for these analyses only. The demo-
graphic characteristics of this student sample are indistin-
guishable from those shown in Table 2. The community
sample did not complete this measure.
334 ASSESSMENT
TABLE 3
Characteristics of Meditation Practice in
Regularly Meditating Individuals
Duration of regular practice (N = 213)
Less than 1 year 8%
1-5 years 28%
6-10 years 18%
More than 10 years 45%
Frequency of meditation sessions (N = 180)
1-2 per week 14%
3-4 per week 39%
5-6 per week 30%
7 or more per week 17%
Length of typical meditation session (N = 180)
<10 minutes 1%
10-20 minutes 18%
21-30 minutes 44%
31-45 minutes 28%
46-60 minutes 8%
>60 minutes 2%
Total number of days on meditation retreats (N = 180)
None 7%
1-4 days 8%
5-10 days 16%
11-30 days 24%
31-90 days 24%
>90 days 20%
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
RESULTS
Analyses tested a series of important questions about
the construct validity of the FFMQ. After verifying ade-
quate-to-good internal consistency across all samples, we
examined the FFMQ’s factor structure in the meditating
sample. If the five subscales measure facets of an overar-
ching mindfulness construct, they should be significantly
intercorrelated, and a hierarchical factor structure should
be confirmed. Because this question has already been
studied in nonmeditators (Baer et al., 2006), we exam-
ined factor structure in the regular meditators only. The
remaining analyses tested whether FFMQ findings are
consistent with a central claim of Buddhist-based medi-
tation traditions: that meditation cultivates the tendency
to be mindful in daily life, which enhances PWB. If this
is true, and if the FFMQ measures mindfulness, then sev-
eral patterns should be observed. Meditation experience
should be positively associated with mindfulness facets
and with PWB, and meditators should score higher than
nonmeditators. If the facets measure important and dis-
tinct elements of mindfulness, they should show incre-
mental validity over other facets in predicting well-being.
Finally, mindfulness facets should mediate the relationship
between meditation experience and well-being.
Internal Consistency of Mindfulness Facets
Alpha coefficients for all facets in all samples were
adequate-to-good (range .72 to .92), with the exception
of the nonreactivity to internal experience facet in the
student sample, for which alpha was .67. For this facet,
alpha coefficients in the other three samples were good,
ranging from .81 to .86.
Facet Intercorrelations and Factor
Structure in the Meditating Sample
Intercorrelations of the five mindfulness facets and
factor structure of the FFMQ were examined in the expe-
rienced meditators only. Intercorrelations of the facets
are shown in Table 4. Correlations range from .32 to .56
(all p < .01), suggesting that the facets represent related
but distinct constructs. Their factor structure in the medi-
tating sample was examined using CFA. In the few cases
of missing data from participants’ responses to the FFMQ,
values were imputed using the expectation maximization
method (Allison, 2003). In CFA, fit indices indicate the
extent to which the covariances among the items are
accounted for by the hypothesized factor model. We used
four fit indices for these analyses: the comparative fit
index (CFI; Bentler, 1990), the Tucker-Lewis Index (TLI;
Tucker & Lewis, 1973), the root mean square error of
approximation (RMSEA; Marsh, Balla, & Hau, 1996),
and the maximum likelihood (ML)–based standardized
root mean squared residual (SRMR; Hu & Bentler, 1999).
By rule of thumb, CFI and TLI values greater than .90 are
thought to indicate good fit between a model and the
data; for the RMSEA, a value of .05 is thought to indicate
close fit, .08 a fair fit, and .10 a marginal fit (Browne &
Cudeck, 1993); for the SRMR, values less than .08 indicate
good fit (Hu & Bentler, 1999).
For several reasons, the following CFAs were con-
ducted using item parcels (groups of items) rather than
individual items. For each facet, items were assigned
sequentially to parcels in the order that they appear on the
instrument (first item to Parcel 1, next item to Parcel 2,
etc.). Item responses within each parcel then were aver-
aged. This yielded a total of 15 parcels (3 per facet), each
the average of two or three items. Little, Cunningham,
Shahar, and Widamon (2002) and Rushton, Brainerd, and
Pressley (1983) have described several advantages of
item parceling. First, the reliability of a parcel of items is
greater than that of a single item, so parcels can serve as
more stable indicators of a latent construct. Second, as
combinations of items, parcels provide more scale points,
thereby more closely approximating continuous mea-
surement of the latent construct. Third, risk of spurious
correlations are reduced, both because fewer correlations
are being estimated and because each estimate is based
on more stable indicators. Fourth, parcels have been shown
to provide more efficient estimates of latent parameters
than do items. Fifth, the object of investigation is not the
performance of specific items but rather the relations
among the scales.
First, we tested a nonhierarchical five-factor model
in which the five factors were allowed to intercorrelate.
Although this model fit the data well (CFI = .97, TLI = .95,
RMSEA = .07 [90% confidence interval: .05 to .08],
SRMR = .05), it does not test whether the five factors are
elements of an overall mindfulness construct. To examine
this question, we tested a hierarchical model in which the
five factors are indicators of an overarching mindfulness
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 335
TABLE 4
Intercorrelations of Five Mindfulness Facets
in Experienced Meditators
Facet Describe Act Aware Nonjudge Nonreact
Observe .40** .43** .49** .56**
Describe — .32** .38** .36**
Act aware — — .39** .49**
Nonjudge — — — .52**
**p < .01.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
factor. No loss of fit was seen with this more parsimonious
model (CFI = .97, TLI = .96, RMSEA = .06 [90% confi-
dence interval: .05 to .08], and SRMR = .05). Loadings
of the five factors on the overall mindfulness construct
were: observing = .82, describing = .53, acting with
awareness = .63, nonjudging = .69, and nonreactivity = .84.
Thus, the hypothesis that the hierarchical, five-factor model
described by Baer et al. (2006) would be replicated in the
sample of experienced meditators was supported by the
model’s good fit to the data.
Finally, the extent to which the five facets are nonover-
lapping was examined using a regression analysis for each
facet in which all four of the other facets were entered as
predictors. The obtained value for adjusted R
2
represents
the variance in each facet accounted for by its relationships
with the other four facets. These values ranged from .24
to .41, suggesting that, although the facets are intercorre-
lated elements of a general mindfulness construct, a sub-
stantial portion of the variance in each facet is distinct
from the other four.
Relationships Between Mindfulness
Facets, Meditation Experience, and
Demographic Variables
If meditation cultivates mindfulness skills and the facet
scores measure these skills, then positive correlations
between meditation experience and each of the facets
should be observed. In addition, regular meditators should
obtain higher mean scores than nonmeditators. Given the
wide range of demographic characteristics in our combi-
nation of samples, it also seemed important to explore
whether mindfulness facet scores are related to demo-
graphic variables. For the following analyses, we combined
the four samples shown in Table 2 into a single data set
(total N = 1,017). Meditation experience was coded as the
number of months of regular practice. Sex and mental
health work were coded dichotomously (male or female and
no or yes, respectively). Race was not examined because
very few participants were non-White.
One-way analysis of variance showed no significant
differences between males and females for any of the mind-
fulness facets. Because meditation experience was signifi-
cantly intercorrelated with age and education, partial
correlations of each of these three variables (controlling
for the other two) with the mindfulness facets were com-
puted and can be seen in Table 5. Results show that medi-
tation experience is significantly and positively correlated
with four of the mindfulness facets (all but acting with
awareness) when the effects of age and education are
controlled. Findings also show that age alone is modestly
correlated with acting with awareness and that education
alone is modestly correlated with all facets.
Mental health professionals comprised large propor-
tions of our two highly educated samples. Effects of
mental health work on levels of mindfulness have not
been previously reported. Therefore, we combined the
two highly educated samples and computed partial corre-
lations between mental health work and mindfulness
facets, controlling for age, education, and meditation
experience. Small but significant relationships were
found for four of the facets, with mental health profes-
sionals scoring slightly higher than others. The following
partial rs were obtained: observing = .12, describing =
.13, nonjudging = .16, nonreactivity = .21 (all p < .01)
and acting with awareness = –.02 (ns). However, addi-
tional partial correlations showed that meditation experi-
ence is significantly related to the same four mindfulness
facets when age, education, and mental health work are
controlled: observing = .38, describing = .15, nonjudging =
.22, nonreactivity = .34 (all p < .01) and acting with
awareness = .05 (ns). Because meditation experience is
significantly related to these facets even when mental
health work is controlled, all remaining analyses were
conducted with mental health professionals and others
combined, within groups.
Next, we compared mean scores for the mindfulness
facets across samples using one-way analysis of variance.
Meditators were expected to score higher than nonmedi-
tators on all facets. M, SDs, and F tests can be seen in the
336 ASSESSMENT
TABLE 5
Partial Correlations Between Mindfulness Facets and Three Participant Characteristics
(Controlling for the Other Two) in all Samples Combined
Mindfulness Facets
IV Control Variables Observe Describe Act Aware Nonjudge Nonreact
Meditation experience Age, education .35** .14** .04 .22** .31**
Age Education, meditation experience .07 .02 .12** –.03 .08
Education Age, meditation experience .17** .29** .13** .18** .20**
NOTE: IV = independent variable.
**p < .01.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
first few columns of Table 6. For all facets, higher scores
represent higher levels of mindfulness. The F tests indi-
cate significant differences between groups for all five
mindfulness facets. We conducted two sets of planned
contrasts to investigate differences between meditators
and nonmeditators. The first set compared meditators to
all nonmeditators combined. Significant differences, with
meditators scoring higher, were seen for all facets. The
second set of planned contrasts compared meditators only
to the demographically similar nonmeditators. Meditators
scored higher on four of the five facets. However, for acting
with awareness the comparison was nonsignificant.
In combination with the partial correlations presented
earlier, these findings suggest that meditation experience
has significant effects on observing, describing, nonjudg-
ing, and nonreactivity, even with other demographic vari-
ables controlled. For acting with awareness, we found
no significant effect of meditation experience, but small
effects for age and education.
Relationships Between Mindfulness Facets
and Psychological Symptoms and Well-Being
A growing literature shows that mindfulness-based
interventions consistently have positive outcomes, sug-
gesting that increased mindfulness is related to decreases
in psychological symptoms. In general, therefore, mind-
fulness facets were expected to be negatively correlated
with symptoms. Based on previous findings (Baer et al.,
2004, 2006), we also predicted that correlations for the
observing facet would vary with level of meditation
experience. Table 7 shows correlations between mindful-
ness facets and symptom measures for all four samples.
(Because preliminary analyses within each of the two
highly educated groups showed no significant differences
between mental health professionals and others for either
symptoms or well-being, mental health professionals are
not examined separately.) As expected, most correlations
between mindfulness facets and symptoms were negative.
However, for the observing facet this was true only in the
meditating sample. In the student sample, this correlation
was positive, and in the other two samples it was non-
significant. This pattern suggests that the tendency to notice
internal and external stimuli is associated with lower
symptom levels in meditators but not in others. This find-
ing is consistent with the hypothesis of Baer et al. (2006)
that the observing facet may function differently in indi-
viduals with meditation experience.
Relationships between mindfulness facets and PWB,
using total scores from the PWB scale, can be seen in
Table 8. (The community sample did not complete the
PWB.) As expected, most of these correlations were sig-
nificant and positive. However, the correlation for the
observing facet was significant only in the meditating sam-
ple, again suggesting that the tendency to notice internal
and external stimuli is strongly related to well-being in
meditators but not in others.
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 337
TABLE 6
Means, SD’s, Univariate
F
Tests, and Planned Contrasts for Mindfulness Facets in Four Samples
Group 1 Group 2 Group 3 Highly Group 4 Planned Contrasts (t)
Students Community Educated Meditators F
Meditators vs. Meditators vs.
Facet MSDMSDMSDMSD(3, 987) All Others Group 3
Observe 24.32 4.84 24.32 5.48 27.04 5.63 31.96 4.16 114.07
†
16.99
†
10.27
†
Describe 26.46 6.01 24.63 7.06 30.01 5.63 31.84 5.30 71.78
†
10.18
†
3.21***
Act aware 25.31 5.77 24.57 6.57 28.32 5.21 28.08 5.10 28.00
†
4.52
†
–.45 (ns)
Nonjudge 27.75 5.90 23.85 7.33 29.13 5.79 32.44 5.63 79.72
†
11.34
†
5.62
†
Nonreact 20.50 3.82 19.53 4.88 22.82 4.19 25.70 4.01 97.04
†
14.24
†
7.11
†
NOTE: For the nonreact facet, possible range of scores is 7-35. For all other facets, possible range is 8-40. Groups 1-3 are nonmeditating.
***p < .001.
†
p < .0001.
TABLE 7
Correlations Between Mindfulness Facets and
Psychological Symptoms in Four Samples
Sample
Highly
Facet Students Community Educated Meditators
Observe .21** .07 .12 –.48**
Describe –.22** –.35** –.30** –.14
Act aware –.50** –.63** –.29** –.30*
Nonjudge –.53** –.58** –.47** –.58**
Nonreact –.22** –.46** –.26* –.43**
NOTE: Students (N = 253) completed the Brief Symptom Inventory, the
community sample (N = 267) completed the Profile of Mood States,
highly educated participants (N = 78) and meditators (N = 61) com-
pleted the Depression Anxiety Stress Scales.
*p < .05. **p < .01.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
Incremental Validity of Mindfulness
Facets in Predicting PWB
Next, we conducted a regression analysis to examine
whether mindfulness facets contribute independently to
the prediction of PWB as measured by the total score on
the PWB. To increase the range of variability for mindful-
ness facets and well-being, we combined the meditating
sample (169 of whom had completed the PWB) with the
demographically similar nonmeditators (75 of whom had
completed the PWB) for a total sample size of 244. All
five mindfulness facets were made available simultane-
ously as predictors. Results can be seen in Table 9. All
facets except observing were significant predictors, jointly
accounting for 39% of the variance. Observing had no
incremental validity over the other four facets. This may
not be surprising, because observing is related to well-
being in the experienced meditators but not in the demo-
graphically similar nonmeditators. These findings suggest
that, at least in a highly educated sample, four of the five
mindfulness facets have incremental validity over the
others in accounting for PWB.
Mediation Analyses
If long-term meditation is beneficial because it culti-
vates mindfulness skills, then mindfulness facet scores
should mediate the relationship between meditation expe-
rience and well-being. We conducted several mediation
analyses to examine this question, using the regression-
based methods described by Baron and Kenny (1986) in
conjunction with methods described by MacKinnon,
Krull, and Lockwood (2000) and MacKinnon, Lockwood,
Hoffman, West, and Sheets (2002) for testing the signifi-
cance of the mediation. To have an adequate range of med-
itation experience, for these analyses we again combined
the meditating sample with the demographically similar
nonmeditators. In each analysis, the independent variable
(IV) was meditation experience (months of regular prac-
tice) and the dependent variable (DV) was PWB total score.
Mindfulness facets served as potential mediators.
According to Baron and Kenny (1986), several condi-
tions must be met to support a mediational hypothesis. If
the data were collected at a single time point, there must
be a sound theoretical basis for the ordering of the vari-
ables. The idea that we tested (that meditation cultivates
mindfulness skills which in turn lead to improved well-
being) is entirely consistent with the mindfulness literature.
In addition, the IV, mediator, and DV must be significantly
intercorrelated. When the IV and mediator are entered
simultaneously into a model predicting the DV, the rela-
tionship between the IV and DV must be significantly
reduced. The significance of this reduction can be exam-
ined with a t test described by MacKinnon et al. (2000).
In addition, the significance of the mediated pathway can
be tested by calculating z’, which MacKinnon et al. (2002)
have shown to have good statistical power and accurate
Type I error rates.
First, we conducted a separate test of mediation for
each of the four mindfulness facets (observing, describing,
nonjudging, nonreactivity) that were significantly corre-
lated with both meditation experience and well-being.
Because of the small but significant relationships shown
earlier between demographic variables and mindfulness
facets, these analyses controlled for age, education, and
mental health work. Meditation experience was a signifi-
cant predictor of well-being (β=.23, p < .05). For each
potential mediator, this coefficient was reduced when the
mediator entered the model to values ranging from .02 (ns)
for nonreactivity to .12 (ns) for describing. These reduc-
tions were all significant at p < .001 (t-values ranged
338 ASSESSMENT
TABLE 8
Correlations Between Mindfulness Facets
and Psychological Well-Being (PWB)
in Three Samples
Sample
Facet Students
a
Highly Educated
b
Meditators
c
Observe .08 .08 .45**
Describe .34** .50** .36**
Act aware .34** .40** .51**
Nonjudge .52** .48** .50**
Nonreact .44** .51** .44**
NOTE: For all participants, the measure of well-being is the total score
from the scales of PWB.
a. n = 175.
b. n = 75.
c. n = 169.
**p < .01.
TABLE 9
Regression Analysis Showing Prediction of
PWB by Mindfulness Facets in Meditators and
Demographically Similar Nonmeditators
Predictor BSEBeta p
Observe –.02 .06 –.02 .77
Describe .17 .05 .20 .002
Act aware .18 .06 .19 .002
Nonjudge .22 .05 .26 .000
Nonreact .21 .08 .18 .011
NOTE: PWB = psychological well-being. Predictor variables entered
simultaneously into a single regression equation. R
2
for model = .39.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
from 4.02 for observing to 5.93 for nonreactivity). In
addition, values of z’ were significant for each of the indi-
rect pathways (for observing, z’ = 3.80; for describing,
z’ = 2.45; for nonjudging, z’ = 4.14; for nonreactivity,
z’ = 5.17; all p < .01). These findings are consistent with
the hypothesis that, when considered separately, each mind-
fulness facet examined (all but acting with awareness)
completely mediates the relationship between meditation
experience and well-being.
To examine the mediational properties of these four
facets in combination, we conducted a final regression
analysis in which the IV (meditation experience) and all
four mediators were entered simultaneously as predictors
of the DV (well-being). In this combined model, which is
depicted in Figure 1, the beta coefficient for the relation-
ship between the IV and DV was reduced to .001 (ns).
The observing facet also was a nonsignificant predictor
of well-being (β=.00), whereas describing, nonjudging
and nonreactivity each accounted for significant variance
in well-being (β=.22, .30, and .22, respectively, p < .05).
This finding suggests that each of these three facets con-
tributes independently to the mediation of the relationship
between meditation experience and well-being.
DISCUSSION
This project addressed several important questions
about the construct validity of the FFMQ in meditating and
nonmeditating samples. Results were largely consistent
with predictions. Four of the facets (all but acting with
awareness) were significantly correlated with meditation
experience (even with other demographic variables con-
trolled), and meditators scored significantly higher than in
other samples. Expected relationships between mindful-
ness facets and symptoms and well-being were found in
most cases, including relationships that vary with medita-
tion experience for the observing facet. The hierarchical
five-factor structure was confirmed in the meditating sam-
ple, and several of the facets significantly mediated the
relationship between meditation experience and well-
being. Although increased mindfulness scores in experi-
enced meditators have been seen with the MAAS (Brown
& Ryan, 2003) and the Toronto Mindfulness Scale (Lau
et al., 2006), this mediational relationship has not previ-
ously been shown. Overall, results suggest that meditation
cultivates several mindfulness skills, and these skills, as
measured by the FFMQ, encourage positive psychological
functioning in long-term practitioners.
We found no significant relationship between medita-
tion experience and the acting with awareness facet.
However, this facet was significantly correlated with the
other facets and with symptoms and well-being in the
expected directions. In addition, in the CFA this facet
loaded significantly on the overall mindfulness construct.
Our meditating sample had a high level of education and
included few individuals with short-term meditation
experience. It is possible that education cultivates the
ability to act with awareness, or that individuals who pur-
sue more education have higher levels of this skill, and
that meditation adds little to the effects of education. The
relationship between this facet and meditation experience
may be significant in samples with levels of education
more typical of the general population, in novice medita-
tors, and in clinical samples. Additional investigation of
this facet in these groups is warranted.
Relations Between the Observing Facet
and Psychological Adjustment
As expected, our findings showed that the relation-
ships between observing and psychological adjustment
(symptoms and well-being) varied with meditation expe-
rience. In meditators, higher levels of observing were
strongly associated with good adjustment. This finding is
striking, given the previously described literature showing
that self-focused attention can be maladaptive (Harvey
et al., 2004; Ingram, 1990; Mor & Winquist, 2002). In
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 339
Meditation
experience
Psychological
well-being
observing
describing
nonjudging
nonreactivity
.33*
.21*
.13*
.37*
.22*
.30*
.23* (−.001)
.22*
.00
FIGURE 1
Mediation by Mindfulness Facets of the
Relationship Between Meditation Experience
and Psychological Well-Being in the Combined
Sample of Meditators and Demographically
Similar Nonmeditators
NOTE: All values are beta coefficients. Values on arrows leading from
meditation experience to mindfulness facets are controlled for age, edu-
cation, and mental health work. Values on arrows leading from mind-
fulness facets to well-being show relationships between these variables
when other facets and meditation experience are included in the model.
The coefficient in parentheses shows the relationship between medita-
tion experience and psychological well-being with all mediating vari-
ables included in the model.
*p < .05.
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
contrast, in our nonmeditating samples, relations between
observing and psychological adjustment were nonsignifi-
cant or in the opposite direction. Several factors may
shed light on this pattern of findings. The observing facet
includes attention to both internal stimuli (thoughts, feel-
ings, sensations) and external stimuli (sights, sounds,
smells). Both were higher in meditators. Because medi-
tation teaches unbiased observation of all stimuli, it may
reduce maladaptive forms of selective attention. Thus,
high scores on the observing facet in meditators may
indicate a greater tendency to notice a wide range of
internal and external stimuli, rather than focusing selec-
tively on the threatening or unpleasant ones. Meditators
may be better able to shift their attention flexibly rather
than becoming rigidly absorbed in any particular class
of stimuli. In addition, higher scores for the describing,
nonjudging, and nonreactivity facets in meditators suggest
that meditators learn to respond differently to the internal
stimuli they observe. In particular, they appear more likely
to observe inner stimuli without judging them or reacting
to them in maladaptive ways and to label observed stimuli
with words. These results are consistent with recent
research in depressed patients showing that mindful self-
awareness can be adaptive, whereas ruminative self-
awareness is not (Watkins & Teasdale, 2004). Findings
for the describing facet also are consistent with recent
neuroscience data suggesting that verbal labeling of affect
modulates brain responses to emotional stimuli in normal
volunteers (Hariri, Bookheimer, & Mazziotta, 2000;
Lieberman et al., 2007).
Limitations and Future Directions
Because our purpose was to investigate in new samples
the construct validity of the five mindfulness facets derived
in previous research (Baer et al., 2006), we measured
mindfulness using only the FFMQ. Although this instru-
ment was derived from existing mindfulness measures, it
does not duplicate them. However, in a previous sample
of student participants (N = 613; Baer et al., 2006) for
whom both the FFMQ facets and the original mindfulness
instruments can be scored, correlations between FFMQ
facets and the other mindfulness measures are very high.
For example, the correlation (r) between the MAAS and
the acting with awareness facet of the FFMQ (which
includes five MAAS items) was .89. Similarly, the corre-
lation between the nonreactivity facet and the MQ (from
which four of the seven nonreactivity items were drawn)
was .75. Thus it seems unlikely that findings would differ
substantially if the original instruments from which the
FFMQ was derived had been used.
Demographically, our meditating sample is not repre-
sentative of the general population. However, it may be
similar in many ways to the population of experienced
meditators. Although we found no published data on the
demographic characteristics of persons who meditate, an
unpublished survey conducted in 2004 by the Insight
Meditation Society in Massachusetts, where more than
2,200 individuals per year participate in meditation
retreats, showed that the average age of their adult partici-
pants was 53 years, 34% were male, and 96% were White.
Although data on education level were not collected, anec-
dotal impression was that most participants have college
degrees (G. Gibson, personal communication, April 14,
2006). Thus, the primary demographic difference between
our meditating sample and other experienced meditators
may be the high percentage of mental health professionals.
We addressed this by recruiting a demographically similar
group of nonmeditators, some of whom work in mental
health fields. As noted earlier, partial correlations found
that mental health work was modestly related to several
of the mindfulness facets but that meditation experience
remained a significant predictor of these facets when
mental health work was controlled. Mental health work
was also unrelated to psychological symptoms or well-
being. Thus, it seems unlikely that the high percentage of
mental health professionals in our meditating sample has
distorted our findings. However, it is important for future
work to study samples of experienced meditators with a
wide range of educational and work histories.
Our findings support an idea that is well established in
Buddhist meditation traditions (Goldstein & Kornfield,
1987): that meditation practice leads to increased mind-
fulness in daily life, which in turn facilitates well-being.
However, because our design was cross-sectional, con-
clusions about these directional pathways must be made
cautiously. Alternative models (e.g., baseline levels of
mindfulness or well-being may influence the likelihood
of maintaining a meditation practice) cannot be ruled out
without longitudinal designs. In addition, although most
of the meditating sample reported that their practice was
primarily or exclusively Buddhist-based, and therefore
likely to involve some form of mindfulness meditation
rather than concentration-based methods, specific practices
can vary across Buddhist traditions (Kabat-Zinn, 2003).
Even within Buddhist traditions, practice may begin with
concentration-based methods before attention is expanded
to include all sensations, cognitions, and emotions that
arise. Therefore, in future studies it may be helpful to ask
participants for more detailed descriptions of their medi-
tation practices, to clarify specific relationships between
these practices and the skills that are cultivated.
340 ASSESSMENT
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
Persons with short-term meditation practice were under-
represented in our sample. Only 8% had meditated for less
than 1 year, and clinical populations were not studied. In
many clinical settings, meditation is taught to patients who
have never meditated before and have lower levels of edu-
cation, on the average, than our meditating sample. Among
empirically supported interventions, the most intensive
meditation training is provided by MBCT and MBSR, both
of which typically are 8 weeks in duration. Our data do not
address the changes in mindfulness facets that occur over 8
weeks of meditation practice, because most persons in our
meditating sample had been practicing for much longer
periods. Thus, it is important that future research examine
effects of meditation on all five mindfulness facets in pop-
ulations and settings more typical of current clinical use of
mindfulness-based interventions, because the facets may
function differently in these groups. Longitudinal research
with frequent assessment of participants would be espe-
cially helpful for examining patterns of skill development
as people new to meditation begin to practice.
Our findings provide good support for the construct
validity of the FFMQ and for continued study of multiple
facets of mindfulness. Measuring elements of mindful-
ness separately allows the investigation of differential
relationships with other variables and helps clarify the
specific skills that are cultivated by meditation practice
and their roles in promoting PWB. It may also contribute
to clarifying changes that occur over time as a meditation
practice is maintained over the long term.
REFERENCES
Allison, P. D. (2003). Missing data techniques for structural equation
modeling. Journal of Abnormal Psychology, 112, 545-557.
Baer, R. A. (2003). Mindfulness training as a clinical intervention:
A conceptual and empirical review. Clinical Psychology: Science
and Practice, 10, 125-143.
Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of mind-
fulness by self-report: The Kentucky Inventory of Mindfulness
Skills. Assessment, 11, 191-206.
Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L.
(2006). Using self-report assessment methods to explore facets of
mindfulness. Assessment, 13, 27-45.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
distinction in social–psychological research: Conceptual, strategic,
and statistical considerations. Journal of Personality and Social
Psychology, 51, 1173-1182.
Bentler, P. M. (1990). Comparative fit indices in structural models.
Psychological Bulletin, 107, 238-246.
Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. C.,
Carmody, J., et al. (2004). Mindfulness: A proposed operational
definition. Clinical Psychology: Science and Practice, 11, 230-241.
Brown, K. W., & Ryan, R. M. (2003). The benefits of being present:
Mindfulness and its role in psychological well-being. Journal of
Personality and Social Psychology, 84, 822-848.
Brown, K. W., & Ryan, R. M. (2004). Perils and promise in defining
and measuring mindfulness: Observations from experience. Clinical
Psychology: Science and Practice, 11, 242-248.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing
model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural
equation models (pp. 136-162). Newbury Park, CA: Sage.
Buchheld, N., Grossman, P., & Walach, H. (2001). Measuring mindful-
ness in insight meditation and meditation-based psychotherapy: The
development of the Freiburg Mindfulness Inventory (FMI). Journal
for Meditation and Meditation Research, 1, 11-34.
Chadwick, P., Hember, M., Mead, S., Lilley, B., & Dagnan, D. (2005).
Responding mindfully to unpleasant thoughts and images: Reliability
and validity of the Mindfulness Questionnaire. Manuscript under
review.
Crawford, J. R., & Henry, J. D. (2003). The Depression Anxiety
Stress Scales (DASS): Normative data and latent structure in a
large nonclinical sample. British Journal of Clinical Psychology,
42, 111-131.
Derogatis, L. R. (1992). The Brief Symptom Inventory (BSI):
Administration, scoring and procedures manual-II (2nd ed.).
Minneapolis, MN: National Computer Systems.
Dimidjian, S., & Linehan, M. M. (2003a). Defining an agenda for
future research on the clinical application of mindfulness practice.
Clinical Psychology: Science and Practice, 10, 166-171.
Dimidjian, S., & Linehan, M. M. (2003b). Mindfulness practice. In
W. O’Donohue, J. E. Fisher, & S. C. Hayes (Eds.), Empirically sup-
ported techniques of cognitive behavior therapy: A step-by-step
guide for clinicians. New York: John Wiley.
Feldman, G. C., Hayes, A. M., Kumar, S. M., Greeson, J. G., &
Laurenceau, J. P. (2007). Mindfulness and emotion regulation: The
development and initial validation of the Cognitive and Affective
Mindfulness Scale–Revised (CAMS-R). Journal of Psychopathology
and Behavioral Assessment, 29, 177-190.
Goldstein, J., & Kornfield, J. (1987). Seeking the heart of wisdom: The
path of insight meditation. Boston: Shambhala.
Grossman, P., Neimann, L., Schmidt, S., & Walach, H. (2004).
Mindfulness-based stress reduction and health benefits: A meta-
analysis. Journal of Psychosomatic Research, 57, 35-43.
Hariri, A. R., Bookheimer, S. Y., & Mazziotta, J. C. (2000). Modulating
emotional responses: Effects of a neocortical network on the limbic
system. NeuroReport, 11, 43-48.
Harvey, A., Watkins, E., Mansell, W., & Shafran, R. (2004). Cognitive
behavioural processes across psychological disorders: A transdiag-
nostic approach to research and treatment. Oxford, UK: Oxford
University Press.
Hayes, S. C., Masuda, A., Bissett, R., Luoma, J., & Guerrero, L. F.
(2004). DBT, FAP, and ACT: How empirically oriented are the new
behavior therapy technologies? Behavior Therapy, 35
, 35-54.
Hayes, S. C., Strosahl, K., & Wilson, K. G. (1999). Acceptance and
commitment therapy: An experiential approach to behavior change.
New York: Guilford.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covari-
ance structure analysis: Conventional criteria versus new alterna-
tives. Structural Equation Modeling, 6, 1-55.
Ingram, R. E. (1990). Self-focused attention in clinical disorders: Review
and a conceptual model. Psychological Bulletin, 107, 156-176.
Kabat-Zinn, J. (1982). An outpatient program in behavioral medicine
for chronic pain patients based on the practice of mindfulness medi-
tation: Theoretical considerations and preliminary results. General
Hospital Psychiatry, 4, 33-47.
Kabat-Zinn, J. (2003). Mindfulness-based interventions in context:
Past, present and future. Clinical Psychology: Science and Practice,
10, 144-156.
Keyes, C. L. M. (2007). Promoting and protecting mental health as
flourishing: A complementary strategy for improving national
mental health. American Psychologist, 62, 95-108.
Baer et al. / MINDFULNESS IN EXPERIENCED MEDITATORS 341
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
Lau, M. A., Bishop, S. R., Segal, Z. V., Buis, T., Anderson, N. D.,
Carlson, L., et al. (2006). The Toronto mindfulness scale: Development
and validation. Journal of Clinical Psychology, 62, 1445-1467.
Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M.,
Pfeifer, J. H., & Way, B. M. (2007). Putting feelings into words:
Affect labeling disrupts amygdala activity to affective stimuli.
Psychological Science, 18, 421-428.
Little, T. D., Cunningham, W. A., Shahar, G., & Widamon, K. F. (2002).
To parcel or not to parcel: Exploring the question, weighing the
merits. Structural Equation Modeling, 9, 151-173.
Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative
emotional states: Comparison of the Depression Anxiety Stress
Scales (DASS) with the Beck Depression and Anxiety Inventories.
Behaviour Research and Therapy, 33, 335-343.
MacKinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000).
Equivalence of the mediation, confounding, and suppression effect.
Prevention Science, 1, 173-181.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., &
Sheets, V. (2002). A comparison of methods to test mediation
and other intervening variable effects. Psychological Methods,
7, 83-104.
Marsh, H. W., Balla, J. R., & Hau, K. Y. (1996). An evaluation of incre-
mental fit indices: A clarification of mathematical and empirical
properties. In G. A. Marcoulides & R. E. Schumacker (Eds.),
Advanced structural equation modeling: Issues and techniques
(pp. 315-353). Mahwah, NJ: Lawrence Erlbaum.
McNair, D. A., Lorr, M., & Droppelman, L. F. (1971). Profile of mood
states. San Diego: Educational and Industrial Testing Service.
Mor, N., & Winquist, J. (2002). Self-focused attention and negative
affect: A meta-analysis. Psychological Bulletin, 128, 638-662.
Nulty, D. D., Wilkins, A. J., & Williams, J. M. G. (1987). Mood, pattern
sensitivity and headache: A longitudinal study. Psychological
Medicine, 17, 705-713.
Robins, C. J., & Chapman, A. L. (2004). Dialectical behavior therapy:
Current status, recent developments, and future directions. Journal
of Personality Disorders, 18, 73-89.
Rushton, J. P., Brainerd, C. J., & Pressley, M. (1983). Behavioral
development and construct validity: The principle of aggregation.
Psychological Bulletin, 94, 18-38.
Ryff, C. (1989). Happiness is everything, or is it? Explorations on the
meaning of psychological well-being. Journal of Personality and
Social Psychology, 57, 1069-1081.
Salmon, P., Sephton, S., Weissbecker, I., Hoover, K., Ulmer, C., &
Studts, J. (2004). Mindfulness meditation in clinical practice.
Cognitive and behavioral practice, 11, 434-446.
Schneider, R. J., Hough, L. M., & Dunnette, M. D. (1996). Broadsided
by broad traits: How to sink science in five dimensions or less.
Journal of Organization Behavior, 17, 639-655.
Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-
based cognitive therapy for depression: A new approach to preventing
relapse. New York: Guilford.
Shapiro, S. L., Schwartz, G. E., & Santerre, C. (2002). Meditation and
positive psychology. In C. R. Snyder & S. J. Lopez (Eds.), Handbook
of positive psychology. New York: Oxford University Press.
Smith, G. T., Fischer, S., Cyders, M. A., Annus, A. M., Spillane, N. S., &
McCarthy, D. M. (2007). On the validity and utility of discriminating
among impulsivity-like traits. Assessment, 14, 155-170.
Smith, G. T., Fischer, S., & Fister, S. M. (2003). Incremental valid-
ity principles in test construction. Psychological Assessment, 15,
467-477.
Smith, G. T., & McCarthy, D. M. (1995). Methodological considera-
tions in the refinement of clinical assessment instruments.
Psychological Assessment, 7, 300-308.
Snyder, C. R., & Lopez, S. J. (2002). Handbook of positive psychology.
New York: Oxford University Press.
Tucker, L. R., & Lewis, C. (1973). The reliability coefficient for maxi-
mum likelihood factor analysis. Psychometrika, 38, 1-10.
Walsh, R., & Shapiro, S. L. (2006). The meeting of meditative disci-
plines and western psychology: A mutually enriching dialogue.
American Psychologist, 61, 227-239.
Watkins, E., & Teasdale, J. D. (2004). Adaptive and maladaptive self-
focus in depression. Journal of Affective Disorders, 82, 1-8.
342 ASSESSMENT
© 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
at UNIV OF KENTUCKY on August 27, 2008 http://asm.sagepub.comDownloaded from
