BRIEF REPORT
Marijuana-Related Problems and Social Anxiety: The Role of Marijuana
Behaviors in Social Situations
Julia D. Buckner
Louisiana State University
Richard G. Heimberg
Temple University
Russell A. Matthews and Jose Silgado
Louisiana State University
Individuals with elevated social anxiety appear particularly vulnerable to marijuana-related problems. In
fact, individuals with social anxiety may be more likely to experience marijuana-related impairment than
individuals with other types of anxiety. It is therefore important to determine whether constructs
particularly relevant to socially anxious individuals play a role in the expression of marijuana-related
problems in this vulnerable population. Given that both social avoidance and using marijuana to cope
with negative affect broadly have been found to play a role in marijuana-related problems, the current
study utilized a new measure designed to simultaneously assess social avoidance and using marijuana to
cope in situations previously identified as anxiety-provoking among those with elevated social anxiety.
The Marijuana Use to Cope with Social Anxiety Scale (MCSAS) assessed behaviors regarding 24 social
situations: marijuana use to cope in social situations (MCSAS-Cope) and avoidance of social situations
if marijuana was unavailable. In Study 1, we found preliminary support for the convergent and
discriminant validity and internal consistency of the MCSAS scales. In Study 2, we examined if MCSAS
scores were related to marijuana problems among those with (n 44) and without (n 44) clinically
elevated social anxiety. Individuals with clinically meaningful social anxiety were more likely to use
marijuana to cope in social situations and to avoid social situations if marijuana was unavailable. Of
importance, MCSAS-Cope uniquely mediated the relationship between social anxiety group status and
marijuana-related problems. Results highlight the importance of contextual factors in assessing
marijuana-related behaviors among high-risk populations.
Keywords: social anxiety, social phobia, marijuana, cannabis, coping motives, social avoidance
Individuals with elevated social anxiety appear particularly vulnerable
to marijuana-related problems (Buckner et al., 2008).
Nearly one third of people with cannabis dependence also have
social anxiety disorder (SAD), a rate higher than for any other
anxiety disorder (Agosti, Nunes, & Levin, 2002). Social anxiety is
related to faster transition from first use to experiencing marijuanarelated
problems among adolescent boys (Marmorstein, White,
Loeber, & Stouthamer.Loeber, 2010). Further, adolescents with
SAD were nearly 5 times more likely to develop cannabis dependence
in young adulthood after controlling for other Axis I disorders
(Buckner et al., 2008). No other mood or anxiety disorder
remained significantly related to subsequent cannabis dependence
after controlling for Axis I disorder comorbidity, suggesting that
clinically elevated social anxiety is an important risk factor for
marijuana-related problems.
Consistent with tension-reduction models (Conger, 1956), socially
anxious individuals may use marijuana to manage chronically
elevated anxiety, and using marijuana in this way may place
them at risk for developing marijuana-related problems. In partial
support of this hypothesis, elevated social anxiety was related to
using marijuana to cope with negative affect, which mediated the
relationship between social anxiety and marijuana-related problems
(Buckner, Bonn.Miller, Zvolensky, & Schmidt, 2007). Yet,
social anxiety and SAD were unrelated to the expectation that
using marijuana would result in reductions in negative affect
(Buckner & Schmidt, 2008, 2009), suggesting the link between
social anxiety and marijuana-related problems may be more complex
than simply using marijuana to decrease anxiety. Another
limitation of this hypothesis is that it does not address the question
as to why people with elevated social anxiety (as opposed to other
types of negative affect) have such high rates of marijuana-related
problems. Also, if socially anxious individuals use marijuana for
tension reduction, it follows that they would use marijuana more
frequently than nonsocially anxious individuals. Yet, findings regarding
the relationship between social anxiety and frequency of
This article was published Online First October 17, 2011.
Julia D. Buckner, Russell A. Matthews, and Jose Silgado, Department of
Psychology, Louisiana State University; Richard G. Heimberg, Department
of Psychology, Temple University.
Correspondence concerning this article should be addressed to Julia D.
Buckner, Department of Psychology, Louisiana State University, 236
Audubon Hall, Baton Rouge, LA 70803. E-mail: jbuckner@lsu.edu
Psychology of Addictive Behaviors c 2011 American Psychological Association
2012, Vol. 26, No. 1, 151.156 0893-164X/12/$12.00 DOI: 10.1037/a0025822
151
marijuana use are mixed (Buckner et al., 2007; Buckner, Ecker, &
Cohen, 2010; Fergusson, Horwood, & Beautrais, 2003; Griffin,
Botvin, Scheier, & Nichols, 2002; Oyefeso, 1991).
It may be that marijuana-related behaviors specifically concerning
social situations place individuals with clinically elevated social anxiety
at particular risk for experiencing marijuana-related problems. In
partial support of this hypothesis, marijuana users with SAD (but not
those without SAD) reported increases in marijuana craving during
periods of elevated state social anxiety (Buckner, Silgado, & Schmidt,
2011). Yet although using marijuana to manage social anxiety was
related to marijuana problems, this link was reduced to nonsignificance
once other marijuana use motives were considered (Lee, Neighbors,
Hendershot, & Grossbard, 2009).
One limitation to the extant research is that the majority used
measures developed to examine global marijuana use behaviors.
For instance, the Marijuana Motives Measure (Simons, Correia,
Carey, & Borsari, 1998) used in prior work (Buckner et al., 2007)
assesses marijuana use to manage negative affect broadly. Yet,
individuals with elevated social anxiety may use marijuana to
manage negative affect related specifically to social anxietyprovoking
situations and may not be especially likely to use
marijuana to manage negative affect in other situations. Although
Lee et al. (2009) attempted to address this limitation by examining
marijuana use in more specific situations, they did not directly
assess whether marijuana was used to manage anxiety in social
situations. Furthermore, avoidance of social situations may be
especially related to marijuana-related problems (Buckner, Heimberg,
& Schmidt, 2011), suggesting the need to examine both
marijuana use to cope in social situations as well as avoidance of
social situations if marijuana is unavailable. It may also be necessary
to measure marijuana use behaviors related to situations
known to be associated with elevated social anxiety and social
avoidance among socially anxious individuals. This approach has
been used to understand the high rates of alcohol-related problems
among those with clinically elevated social anxiety (Buckner &
Heimberg, 2010; Thomas, Randall, & Carrigan, 2003).
The present study tested whether using marijuana to cope specifically
in social situations and avoidance of social situations if
marijuana was not available were associated with marijuanarelated
problems among socially anxious individuals. Given the
lack of measures designed to assess these constructs, two studies
were conducted. In Study 1, we examined the psychometric properties
of the Marijuana Use to Cope with Social Anxiety Scale
(MCSAS), a self-report measure developed for this study to assess
using marijuana to cope in social situations and avoidance of social
situations in the absence of marijuana. In Study 2, we examined if
socially anxious individuals used marijuana to cope in more social
situations and avoided more social situations if marijuana was
unavailable. We also tested whether these constructs mediated the
link between social anxiety and marijuana problems. We examined
these variables in undergraduates given this age cohort is especially
vulnerable to marijuana problems. Specifically, age of cannabis
use disorder (CUD) onset peaks at this age followed by a
sharp decline (Stinson, Ruan, Pickering, & Grant, 2006) and
marijuana use rates are similar between undergraduates and noncollege
peers (Johnston, OfMalley, Bachman, & Schulenberg,
2007).
Study 1
Sample and Procedures
Participants (N 35; 60% female; 82.9% Caucasian) were
recruited through the psychology student participant pool based on
responses to an online screening question assessing current (past 3
months) marijuana use. This sample was composed of current
marijuana users, with 57.1% endorsing weekly marijuana use. The
mean [M] age was 19.8 (standard deviation [SD] 2.9). Participants
completed computerized versions of measures using surveymonkey.
com and received referrals to university-affiliated psychological
outpatient clinics and research credit for study completion.
This study was approved by the universityfs Institutional Review
Board.
Measures
MCSAS. The MCSAS is a modification of a similar scale of the
use of drinking-related behaviors (see Buckner & Heimberg, 2010).
Items were modified from the Liebowitz Social Anxiety Scale
(LSAS; Liebowitz, 1987), a highly reliable and valid measure of
anxiety in specific social situations (e.g., Fresco et al., 2001; Heimberg
& Holaway, 2007), thereby providing items ideally suited for
assessing social situations that have been found to be related to greater
distress and avoidance among socially anxious individuals. Participants
rated the degree to which they use marijuana to cope in the 24
LSAS social situations (e.g., participating in small groups, going to a
party, being the center of attention) as 0 never, 1 occasionally
(1.33%), 2 often (34.67%), or 3 usually (68.100%). The same
scale was used to assess the degree of avoidance if marijuana was
unavailable in each of the 24 social situations. Consistent with prior
work (Thomas et al., 2003), each item was scored dichotomously
indicating whether an individual did (1) or did not (0) endorse each
item. The gdidh responses were summed to provide the total number
of situations in which marijuana was used to cope (MCSAS-Cope) or
were avoided if marijuana was unavailable (MCSAS-Avoid).
Other measures. Participants completed the self-report version
of the LSAS (Liebowitz, 1987). Frequency of past 3-month marijuana
use was assessed with the Marijuana Use Form (Buckner et al., 2007)
on a 0 (never) to 10 (at least 21 times per week) rating scale.
Frequency of past-week tobacco smoking was assessed with the
Smoking History Questionnaire (SHQ; Brown, Lejuez, Kahler, &
Strong, 2002) and typical drinking quantity was assessed by asking
participants to indicate how much alcohol they drank on a typical
weekend evening in the past month. Participants completed the Marijuana
Problems Scale (MPS; Stephens, Roffman, & Curtin, 2000), a
list of 19 negative consequences related to marijuana use in the past
90 days. Participants rated each marijuana problem as 0 (no problem),
1 (minor problem), or 2 (serious problem). The Marijuana Effect
Expectancies Questionnaire (MEEQ) is a list of 48 expectations
regarding marijuana use rated from 1 (strongly disagree) to 5
(strongly agree; Aarons, Brown, Stice, & Coe, 2001). It is composed
of six subscales: Cognitive and Behavioral Impairment, Relaxation
and Tension Reduction, Social and Sexual Facilitation, Perceptual and
Cognitive Enhancement, Global Negative Effects, and Craving and
Negative Effects. These subscales have demonstrated adequate reliability
(e.g., Aarons et al., 2001; Buckner & Schmidt, 2008). The
Marijuana Motives Measure (MMM; Simons et al., 1998) is a 25-item
152 BUCKNER, HEIMBERG, MATTHEWS, AND SILGADO
measure assessing the following marijuana use motives: enhancement,
coping, social, conformity, and expansion. Participants indicate
the degree to which they have smoked marijuana for particular reasons
from 1 (almost never/never) to 5 (almost always/always).MMM
subscales have demonstrated excellent internal consistency (Chabrol,
DucongeL, Casas, Roura, & Carey, 2005).
Study 1 Results
See Table 1 for means, standard deviations, alphas, and correlations
of MCSAS scales with LSAS and substance use variables.
MCSAS-Cope was significantly related to MEEQ-Social and Sexual
Facilitation and MEEQ-Relaxation and Tension Reduction.
MCSAS-Avoid was related to MEEQ-Social and Sexual Facilitation
and MEEQ-Perceptual and Cognitive Enhancement. MCSASCope
was significantly related to MMM-Social, MMM-Coping,
and MMM-Enhancement. MCSAS-Avoid was related to MMMSocial,
MMM-Coping, and MMM-Expansion. Marijuana problems
were related to MCSAS-Cope but not MCSAS.Avoid.
Weekly marijuana users (M 7.00, SD 7.30) had higher
MCSAS-Cope scores than infrequent users (M 0.60, SD
1.24), F(1, 34) 11.21, p .002, d 1.18. Weekly marijuana
users (M 8.55, SD 15.12) did not significantly differ from
infrequent users (M 3.53, SD 9.16) on MCSAS-Avoid, F(1,
34) 1.28, p .266, d .40. Both MCSAS scales were unrelated
to alcohol and tobacco use (see Table 1).
Study 1 Discussion
Results provide preliminary support for the internal consistency
and convergent validity of the MCSAS scales given that both
scales were significantly related to MEEQ-Social and Sexual Facilitation,
MMM-Coping, and MMM-Social. MCSAS-Cope was
also significantly related to MEEQ-Relaxation and Tension Reduction.
Of note, the magnitude of these correlations suggests that
although these constructs were related, the MCSAS subscales
appear to assess constructs that differ from those assessed by the
MEEQ and MMM. MCSAS-Cope (but not Avoid) scores were
related to marijuana problems and were higher among weekly
users. There was some support for discriminant validity.although
MCSAS-Cope was related to marijuana use frequency, it was
unrelated to frequency of tobacco or alcohol use.
Study 2 Method
Sample and Participant Selection
To test if MCSAS scores play a role in the relation between
social anxiety and marijuana problems, 1,156 potential participants
were recruited through the psychology student participant pool to
complete an online survey. Of these, 252 (22%) endorsed current
(past 3 months) marijuana use and were included in the present
study. This sample was predominantly female (63.6%) and non-
Hispanic/Latino (96.4%). Racial composition was 8.0% African
American, 0.4% American Indian, 2.4% Asian American, 84.0%
Caucasian, 4.0% gmixed,h and 1.2% gother.h Ages ranged from
18.35 (M 19.93, SD 2.17) and 34.7% endorsed weekly
marijuana use. Participants completed measures using surveymonkey.
com and received referrals to university-affiliated psychological
outpatient clinics and research credit for study completion.
This study was approved by the universityfs Institutional Review
Board.
Table 1
Relations Between MCSAS Scales and Measures of Social Anxiety and Marijuana Behaviors in
Study 1
M SD
MCSAS MCSAS
Cope Avoid
MCSAS-Cope 2.94 3.96 .89
MCSAS-Avoid 3.49 6.32 .96 .43
LSAS-Anxiety 17.31 8.40 .89 .32 .67
LSAS-Avoidance 15.26 11.13 .89 .49 .79
MEEQ cognitive/behavioral impairment 31.71 6.01 .76 .00 .30
MEEQ relaxation/tension reduction 29.20 4.81 .75 .37 .23
MEEQ social and sexual facilitation 27.20 5.92 .74 .54 .48
MEEQ perceptual/cognitive enhancement 26.57 5.04 .71 .20 .35
MEEQ global negative effects 15.43 4.45 .65 .21 .15
MEEQ craving/physical effects 24.74 3.55 .74 .08 .24
MMM social motives 11.57 5.14 .86 .65 .36
MMM coping motives 9.40 4.76 .86 .54 .34
MMM enhancement motives 17.91 5.73 .89 .39 .13
MMM conformity motives 6.17 2.16 .61 .04 .03
MMM expansion motives 9.37 4.89 .94 .21 .34
Marijuana problems 2.77 2.81 .79 .48 .20
Tobacco usea 2.67 5.27 NA .14 .32
Typical drinking quantity 6.57 3.33 NA .19 .15
Note. MCSAS Marijuana Use to Cope with Social Anxiety Scale; LSAS Liebowitz Social Anxiety Scale;
MEEQ Marijuana Effect Expectancy Questionnaire; MMM Marijuana Motives Measure; NA not
applicable (single item measure).
a Assessed for current tobacco users only (n 20).
p .05. p .01. p .001.
MARIJUANA PROBLEMS AND SOCIAL ANXIETY 153
Measures
Social anxiety. The Social Interaction Anxiety Scale (SIAS;
Mattick & Clarke, 1998) is a measure of general social interaction
fears that demonstrates high levels of internal consistency and
test.retest reliability across clinical, community, and student samples
(Mattick & Clarke, 1998; Osman, Gutierrez, Barrios, Kopper,
& Chiros, 1998). To increase generalizability to individuals with
SAD, an empirically supported cutoff score (34; see Heimberg,
Mueller, Holt, Hope, & Liebowitz, 1992) was used to identify
those with clinically meaningful levels of social anxiety (high
social anxiety [HSA] n 44). To facilitate the comparison of
those with clinically meaningful social anxiety to those with normative
levels of social anxiety, a randomly selected unmatched
group of 44 participants scoring below the Heimberg et al. (1992)
SIAS community sample mean (20) was selected to comprise the
lower social anxiety (LSA) group. The SIASfs internal consistency
was adequate for the entire sample ( .93) and our clinical
analogue sample ( .96). Information regarding demographic
characteristics, social anxiety, and marijuana use of the HSA and
LSA groups is presented in Table 2.
Substance use measures. Internal consistency was adequate
for the MCSAS Scales for entire sample (MCSAS-Cope .95;
MCSAS-Avoid .96) and our clinical analogue sample
(MCSAS-Cope .95; MCSAS-Avoid .96). It was also
adequate for the Marijuana Problems Scale (Stephens et al., 2000)
for entire sample ( .84) and our clinical analogue sample (
.87). Participants also completed the Marijuana Use Form (Buckner
et al., 2007).
Study 2 Results
Among all current marijuana users, social anxiety was positively,
significantly correlated with both MCSAS-Cope (r .19,
p .002) and MCSAS-Avoid (r .25, p .001) as well as
marijuana problems (r .18, p .005). However, the magnitude
of the relation of social anxiety with marijuana problems was
small. The magnitude of the relation between social anxiety and
marijuana problems was larger in our clinical analogue sample
(see Table 2). The most common problems endorsed by HSA
participants were procrastination (endorsed by 45.5% of HSA
participants), lower productivity (45.5%), and lower energy
(36.4%). Marijuana problems were correlated with MCSAS-Cope
(r .56, p .001) and MCSAS-Avoid (r .27, p .011). It is
interesting that both LSA and HSA participants reported using
marijuana to cope in social situations and avoiding social situations
if marijuana was unavailable (see Table 2). As compared
with the LSA group, a significantly larger percentage of the HSA
group was likely to use marijuana to cope in social situations (p
.006) and avoid social situations if marijuana was unavailable (p
.055). HSA participants reported use of marijuana to cope in
significantly more social situations (p .033) and avoiding a
significantly greater number of social situations if marijuana was
unavailable (p .007) as compared with the LSA group.
Given that the relation between social anxiety and marijuana
problems was larger in our clinical analogue sample, we tested
whether MCSAS scalesf total scores mediated this relationship
using maximum likelihood bootstrapping (5,000 samples were
drawn) within the structural equation modeling program AMOS 17
(McCabe et al., 2004); estimated standard errors and confidence
intervals (90%) were calculated for all indirect, direct, and total
effects. Three fully mediated models were tested (see Figure 1).
Specifically, we tested the mediational effects of MCSAS-Avoid
(Model A), MCSAS-Cope (Model B), and given the bivariate
relationship between MCSAS-Avoid and MCSAS-Cope, we examined
the two proposed mediators simultaneously (Model C) to
better understand their additive contributions.
For each model, three measures of model fit were calculated; 2,
comparative fit index (CFI), and standardized root mean square
residual (SRMR). A nonsignificant 2 indicates good model fit;
however, 2 is sensitive to sample size. A CFI value of .95 or
higher and an SRMR value of .08 or lower are indicative of good
model fit (Lasser et al., 2000). As reported in Figure 1, Models B
and C demonstrated acceptable fit; Model A demonstrated poor fit.
Standardized path estimates are reported for each model. In Model
A, social anxiety group had an unstandardized indirect effect (via
MCSAS-Avoid) on marijuana-related problems of .62 (p .05),
suggesting that when social anxiety group goes up by one (from
LSA to HSA), marijuana-related problems goes up by .62; how-
Table 2
Demographic Characteristics, Social Anxiety, and Marijuana Use Behaviors by Social Anxiety Group
Classification (Study 2)
LSA (n 44) HSA (n 44)
F or 2 M or % SD M or % SD p dor
Age (years) 20.14 1.42 20.05 1.84 0.07 .791 0.06
Sex (% female) 67.4 63.6 0.14 .709 0.04
Race (% Caucasian) 86.0 84.1 0.07 .798 0.03
Employed (%) 55.8 56.8 0.01 .925 0.01
Typical drinking frequency 2.75 1.63 3.05 1.64 0.72 .399 0.18
Marijuana use (% use marijuana weekly) 29.5 38.6 0.81 .368 0.10
Marijuana problem severity 2.32 2.49 4.23 4.95 5.21 .025 0.49
Social anxiety 13.36 4.69 41.27 8.40 370.38 .001 4.15
% Use marijuana to cope in social situations 36.4 65.9 7.69 .006 0.30
% Avoid social situations if marijuana unavailable 40.9 60.0 3.68 .055 0.21
# Social situations in which marijuana used to cope 2.07 4.46 4.57 6.20 4.71 .033 0.47
# Social situations avoided if marijuana unavailable 2.00 3.56 5.59 7.78 7.75 .007 0.60
Note. Social anxiety was assessed with the Social Interaction Anxiety Scale. LSA lower social anxiety; HSA
higher social anxiety.
154 BUCKNER, HEIMBERG, MATTHEWS, AND SILGADO
ever, this indirect effect should be interpreted with caution given
the CFI value. In Model B, social anxiety group had an unstandardized
indirect effect (via MCSAS-Cope) on marijuana-related
problems of 1.01 (p .05). When these two mediators were
combined in Model C, only MCSAS-Cope demonstrated a significant
direct effect on marijuana-related problems ( .56, p
.01). In this model, the social anxiety group again demonstrated an
indirect effect on marijuana-related problems; the unstandardized
indirect effect was 1.00 (p .01). These results suggest that the
primary mediational effect is via MCSAS-Cope.1
Given the limitations of conducting mediational analyses using
cross-sectional data, one method of increasing confidence in the
observed effects is to reverse the proposed mediator with the
criterion variable (Sheets & Braver, 1999). We evaluated whether
marijuana problems mediated the relation between social anxiety
group and each MCSAS scale. The MCSAS-Avoid analysis was
not consistent with mediation in this direction as the model was a
poor fit, 2 5.08, p .024, CFI .71, SRMR .09. However,
the MCSAS-Cope model was a good fit, 2 1.21, p .271,
CFI .99, SRMR .04. In this model, social anxiety group had
an unstandardized indirect effect (via marijuana problems) on
MCSAS-Cope of .13 (p .001).
Study 2 Discussion
Compared to LSA individuals, HSA individuals used marijuana to
cope in a greater number of social situations and avoided a greater
number of social situations if marijuana was unavailable. Importantly,
using marijuana to cope specifically in social situations (more so than
avoidance of social situations if marijuana was not available) at least
partially accounts for marijuana-related problems among individuals
with HSA, a group at particular risk for marijuana-related problems
and CUD (Buckner et al., 2008). Yet, it is unclear why using marijuana
to cope in more social situations was related to more marijuanarelated
problems among HSA participants. One possibility is that
HSA marijuana usersf reliance on marijuana to help them cope in
social situations may interfere with the learning or use of more
adaptive coping strategies. Furthermore, they may come to believe
they need marijuana to cope with these situations and be particularly
likely to continue to use marijuana despite possible negative consequences.
This is consistent with the especially high rates of cannabis
dependence among HSA individuals (Buckner et al., 2008). However,
it is also feasible (and consistent with our reversed mediational model)
that HSA participants have marijuana problems and that the experience
of these problems for some reason increases the likelihood that
they will use marijuana to cope in social situations. Future work is
necessary to delineate the temporal relations between these variables.
Our finding that marijuana-related problems were significantly
related to avoidance of social situations if marijuana is unavailable
is consistent with prior work finding that social avoidance is
especially related to marijuana-related problems (Buckner, Heimberg,
et al., 2011). Although the mediational model was a poor fit,
our data overall suggest that avoidance of marijuana-free social
events may be problematic for HSA individuals. HSA individuals
avoided more social situations in which marijuana was unavailable.
By avoiding situations in which marijuana is unavailable,
HSA individuals may disproportionately choose to attend social
situations in which marijuana is available. Considered in combination
with our finding that HSA individuals are more likely to
rely on marijuana to cope during social events, the choice to attend
those social events involving marijuana may place them at risk for
using marijuana to cope in these situations thereby increasing their
risk for marijuana-related problems.
Findings should be considered in light of limitations. The studyfs
cross-sectional nature precludes delineation of causal relationships,
and prospective work will be an important next step. Also, the
samples were primarily female and undergraduate; thus replication in
other populations is necessary. Although an empirically supported
clinical cut-off score was used to identify participants with clinically
elevated social anxiety, replication with patients with SAD is necessary,
especially given that the strength of the link between social
anxiety and marijuana problems was greater in our clinical analogue
sample than that obtained using continuous SIAS scores. We did not
include measures of marijuana expectancies or marijuana use motives
in Study 2, and future work could benefit from testing if MCSAS
scales are more strongly related to marijuana-related problems than
these other constructs among HSA individuals. Similarly, future work
could test whether HSA is uniquely related to MCSAS scale scores or
whether HSA individuals are vulnerable to also using marijuana to
cope in nonsocial situations.
1 A similar pattern of findings emerged when analyses were conducted
with the entire sample (N 254) using continuous SIAS scores, 2 2.23,
p .135, CFI .99, SRMR .03. When both mediators entered
simultaneously, only MCSAS-Cope demonstrated a significant direct effect
on marijuana-related problems.
Model A: Mediational Effects of MCSAS-Avoid
Social Anxiety
Group
MCSASAvoid
Marijuana-Related
Problems
.29** .27**
ƒÔ2(1) = 2.69, p > .05, CFI = .88, SRMR = .07
Model B: Mediational Effects of MCSAS-Cope
Social Anxiety
Group
MCSASCope
Marijuana-Related
Problems
.23** .56**
ƒÔ2(1) = 1.69, p > .05, CFI = .98, SRMR = .05
Model C: Additive Mediational Effects of MCSAS-Avoid & MCSAS-Cope
Social Anxiety
Group
MCSASAvoid
Marijuana-Related
Problems
.29** -.01
ƒÔ2(1) = 1.81, p > .05, CFI = .99, SRMR = .04
MCSASC.
23** ope
.45**
.56**
Figure 1. Standardized direct effects and fit statistic information for
proposed mediational models in Study 2 for MCSAS-Avoid (Model A),
MCSAS-Cope (Model B), and MCSAS-Avoid and MCSAS-Cope combined
(Model C).
您必须登录才能发表评论。