Content uploaded by Kelly Harper
Author content
All content in this area was uploaded by Kelly Harper on May 17, 2020
Content may be subject to copyright.
Journal of Social and Clinical Psychology, Vol. 39, No. 2, 2020, pp. 117-140
PERFECTIONISM AND LONELINESS
HARPER ET AL.
PERFECTIONISM AND LONELINESS: THE ROLE
OF EXPECTATIONS AND SOCIAL HOPELESSNESS
IN DAILY LIFE
KELLY L. HARPER, KARI M. EDDINGTON, AND PAUL J. SILVIA
University of North Carolina at Greensboro
Introduction: Research suggests that social disconnection is one pathway that
explains why perfectionists have higher negative mental health outcomes. How-
ever, thus far research has not examined the role that perfectionists’ expectations,
about their social world, plays in explaining social disconnection. The current
study examined whether negative expectations about future social events and so-
cial hopelessness explained the relations between socially prescribed perfection-
ism (SPP), self-oriented perfectionism (SOP), and social disconnection. Method:
The current study used experience sampling methodology (ESM) with 145 partici-
pants to test multilevel structural equation models examining whether negative
expectations and social hopelessness mediated the relation between SPP (as well
as SOP) and loneliness at the following time point. Results: Results indicate that
having negative expectations for future social interactions and feeling socially
hopeless explains the relation between SPP and loneliness both when loneliness
was measured concurrently and loneliness measured later in the day. Having
lower levels of social hopelessness explained the relation between SOP and lone-
liness when loneliness was measured concurrently and later in the day. Discus-
sion: Findings suggest that SPP and SOP are differently associated with cognitions
about social interactions and relationships in daily life, which in turn explains
variations in loneliness. Results indicate that targeting socially prescribed per-
fectionists’ negative social cognitions in treatment may be effective for reducing
loneliness.
Keywords: perfectionism, socially prescribed perfectionism, self-oriented perfec-
tionism, loneliness, social disconnection, experience sampling methodology
A common distinction made across different conceptualizations
of perfectionism is where perfectionistic standards are attrib-
Address correspondence to Kelly L. Harper who is now at the Center for PTSD,
Behavioral Sciences Division, VA Boston Health Care System, 150 South Huntington
Avenue, Boston, MA, 02130; Email: Kelly.Harper@VA.gov.
118 HARPER ET AL.
uted (e.g., to self versus others). Specifically, one dimension of
perfectionism, socially prescribed perfectionism (SPP), involves
introjected standards from others and the perception that one
cannot live up to others’ expectations. Another dimension of
perfectionism, self-oriented perfectionism (SOP), involves in-
trinsic, personal, perfectionistic standards (Hewitt & Flett, 1991).
Socially prescribed perfectionism is consistently associated with
depressive symptoms, anxiety disorders, eating disorders (see
Shafran & Mansell, 2001 for a review) and suicidality (see Smith
et al., 2017 for a meta-analysis). Self-oriented perfectionism, in
comparison, has a less consistent and weaker relationship with
psychopathology (Shafran & Mansell, 2001) and is either unre-
lated to or negatively related to suicidality (see O’Connor, 2007
for a review).
One pathway theorized to explain the consistent link between
SPP and negative mental health outcomes, including suicidality,
is social disconnection (Hewitt, Flett, Sherry, & Caelian, 2006).
Research indicates that SPP is associated with lower perceived
social support and higher levels of loneliness (Chang, Sanna,
Chang, & Bodem, 2008; Flett, Druckman, Hewitt, & Wekerle,
2012). Moreover, these aspects of social disconnection play a
role in the association between SPP and negative mental health
outcomes. For example, loneliness moderates the association be-
tween SPP and depressive symptoms, such that having higher
levels of SPP and higher levels of loneliness is related to higher
depressive symptoms (Chang et al., 2008).
The association between SOP and social disconnection is less
clear. Although one study found that SOP was positively associ-
ated with social connection (Rice, Leever, Christopher, & Por-
ter, 2006), the majority of findings indicate that SOP or having
high personal standards is not directly associated with indica-
tors of social disconnection, such as loneliness (Chang et al.,
2008; Chang, 2013). Research suggests that social factors, such as
higher levels of loneliness or interpersonal stress, may moderate
the association with SOP and negative mental outcomes (Chang
et al., 2008; Joiner & Schmidt, 1995). These findings are consis-
tent with research that suggests that SOP’s relation to negative
mental health outcomes may be based on contextual factors and
social processes may be one contextual factor.
PERFECTIONISM AND LONELINESS 119
A potentially important factor in the relationship between
perfectionism and social disconnection may involve how per-
fectionists think about their social experiences. Specifically, per-
fectionist’s expectations about social interactions and perceived
social hopelessness may play a role in explaining experiences of
disconnection. Negative expectations, which can take the form
of expecting bad things to happen or having lower expectations
that good things will happen, are associated with lower well-
being (MacLeod & Byrne, 1996). Research on perfectionism and
future expectations suggests that people high on SPP have more
negative thinking and less positive thinking about the future. In
contrast, people high on SOP tend to have more positive future
thinking (O’Connor, O’Connor, O’Connor, Smallwood, & Miles,
2004; Stoeber & Corr, 2017). Regarding future thinking about so-
cial interactions, only one study examined socially prescribed
perfectionists’ beliefs about a future social interaction with a
stranger (Laurenti, Bruch, & Haase, 2008). Socially prescribed
perfectionism did not predict negative statements (e.g., self-
deprecating statements, fear of negative evaluation); however,
SPP moderated the relation between social anxiety and negative
statements such that people who are socially anxious and high
on SPP made more negative statements. This study did not mea-
sure SOP or personal perfectionistic standards; therefore, less is
known about how people high on SOP think about potential fu-
ture interactions.
Regarding social hopelessness, SPP, but not SOP, has been pos-
itively associated with ratings of social hopelessness, which in
turn may explain the association between SPP and suicide risk
(Roxborough et al., 2012). Because SOP was not directly associat-
ed with suicide risk or social hopelessness, this study did not test
a mediation model for SOP. Thus far, research has not examined
how expectations for future social interactions and social hope-
lessness relate to social disconnection, particularly in the context
of the participant’s daily life experiences.
Our understanding of social disconnection in perfectionism
may be enhanced by examining how perfectionists think about
social experiences in daily life, which is likely an important piece
in how perfectionists experience their social world. The use of
experience sampling methodology (ESM), which is when people
120 HARPER ET AL.
report on mood, cognitions, or behaviors at multiple time points
across days or weeks (Conner, Tennen, Fleeson, & Barrett, 2009),
can be useful in studying processes in daily life, such as social ex-
pectations and feeling disconnected from others. Using daily di-
ary methods, Dunkley, Zuroff, and Blankstein (2003) found that
self-critical perfectionism, a latent variable comprised in part by
SPP, was negatively related to daily perceived social support,
which in turn was associated with higher positive affect. Thus
far, one study has used multiple time points in a day to examine
SPP and mood (Mushquash & Sherry, 2012). Results suggested
that perfectionism was positively related to perfectionistic dis-
crepancies (perceiving one is not meeting others’ standards) and
negative affect. Additionally, self-defeating behaviors (i.e., binge
eating, procrastination, and interpersonal conflict) were associ-
ated with negative affect. Regarding daily diary methods with
perfectionistic personal standards, Dunkley, Ma, Lee, Preacher,
and Zuroff (2014) found that personal standards were not di-
rectly associated with perceived social support or daily positive
affect; however, people high on personal standards engaged in
positive reinterpretation of daily stressors, which in turn was as-
sociated with daily positive affect. These studies suggest that ex-
amining social and cognitive processes in daily life is useful for
understanding perfectionists’ daily affective experiences.
In summary, SPP is consistently associated with indicators of
social disconnection, such as loneliness (Chang et al., 2008). Ad-
ditionally, results suggest that loneliness moderates the associa-
tion between SPP and negative mental health outcomes (Chang
et al., 2008). However, prior research has primarily relied on self-
report measures administered at a single time point and has not
examined whether socio-cognitive processes in daily life may ex-
plain feelings of disconnection in people high on SPP. If people
high on SPP have more negative future thinking (O’Connor et
al., 2004; Stoeber & Corr, 2017) and higher levels of social hope-
lessness (Roxborough et al., 2012), this negative future thinking
within the social context may impact their feelings of connection
to others.
Self-oriented perfectionism, in contrast, has not been directly
associated with levels of loneliness (Chang et al., 2008; Chang,
2013). However, experiencing high levels of loneliness may be a
PERFECTIONISM AND LONELINESS 121
factor associated with depressive symptoms in people high on
SOP (Chang et al., 2008). Therefore, further research on SOP and
loneliness is needed. To our knowledge, research has not exam-
ined how people high on SOP think about their social experi-
ences, and whether socio-cognitive processes help explain any
indirect associations between SOP and daily levels of loneliness.
Previous research suggests that SOP is related to positive future
thinking O’Connor et al., 2004; Stoeber & Corr, 2017), which may
also extend to future thinking about social experiences. Addi-
tionally, using daily diary methods, a study found that holding
high personal standards was indirectly related to positive affect
through positive reinterpretation of events (Dunkley et al., 2014),
suggesting that cognitive processes in daily life may be especial-
ly important to examine in relation to SOP.
CURRENT STUDY
The aim of the present study was to examine whether expecta-
tions about future social interactions and social hopelessness ex-
plain the relation between perfectionism and loneliness, which
was used as an indicator of social disconnection. The purpose
of using three time points within a day across 7 days was to ex-
amine temporal effects of negative social expectations and so-
cial hopelessness on social disconnection. Based on previous
research supporting a positive association between SPP and
loneliness (Chang, 2013), it was expected that SPP would predict
daily levels of loneliness while controlling for trait loneliness.
Socially prescribed perfectionism was hypothesized to predict
higher daily negative expectations for future social interactions
(O’Connor et al., 2004; Stoeber & Corr, 2017) and daily social
hopelessness (Roxborough et al., 2012). Additionally, negative
expectations about future social interactions and feeling more
socially hopeless were hypothesized to mediate the relation be-
tween SPP and loneliness both when measured concurrently and
at the following time point.
SOP was also included to further contribute to the literature
examining personal, perfectionistic standards and loneliness
(Chang et al., 2008). Based on previous research, SOP was not ex-
pected to be directly associated either with loneliness (O’Connor
122 HARPER ET AL.
et al., 2004; Stoeber & Corr, 2017) or with social hopelessness
(Roxborough et al., 2012); however, based on self-oriented per-
fectionists’ tendency to have higher positive future thinking,
SOP was expected to be negatively related to negative expecta-
tions about future social interactions. Therefore, based on our
predictions, only expectations about future social interactions
would explain an indirect relationship between SOP and daily
loneliness.
METHODS
PARTICIPANTS
Participants (N = 160) were undergraduate students at a south-
eastern university. Participants were offered extra credit in their
course or received partial research credit for participating de-
pending on the semester of participation. To ensure that there
was representation in high scores on SPP, some participants were
contacted via E-mail and invited to participate if they scored 62
or higher on the SPP subscale of the Multidimensional Perfec-
tionism Scale (MPS; Hewitt & Flett, 1991). Ten participants who
completed less than 1/3 of the daily surveys (e.g., less than 7 sur-
veys) were excluded from all analyses. Three participants who
had high levels of inconsistent responding were also excluded
(see below for description on the inconsistent responding mea-
sure). Lastly, two people were excluded for selecting the Pacific
time zone (instead of Eastern time zone) when signing up for the
daily surveys.
Participants (N = 145) had a mean age of 20.1 (SD = 5.40, range:
18 to 56), 72% identified as female, 27% identified as male, and
1% declined to state their gender. Participants identified as Asian
(8.33%), African-American (33%), Hispanic/Latino (15.4%), Na-
tive Hawaiian or other Pacific Islander (<1%), Middle Eastern/
Arab (1.3%), White (46.2%), or Other (1.9%). Participants could
select more than one racial/ethnic identity.
PERFECTIONISM AND LONELINESS 123
MEASURES
BETWEEN-PERSON MEASURES
Perfectionism
The Multidimensional Perfectionism Scale (MPS; Hewitt & Flett,
1991) was used to measure perfectionism. The MPS is a 45-item
scale with three subscales that capture different dimensions
of perfectionism: SPP, SOP, and other-oriented perfectionism
(OOP). Items are rated on a 7-point scale from 1 (strongest dis-
agreement) to 7 (strongest agreement). The subscales of the MPS
had good internal consistency (SPP a = .88; OOP a = .76; SOP a
= .87).
Loneliness
Trait loneliness was measured with the University of California,
Los Angeles (UCLA) Loneliness Scale Version 3 (Russell, 1996).
The UCLA Loneliness Scale is a 20-item measure that assesses
trait loneliness. Items are rated on a 4-point scale from 1 (never)
to 6 (always). The UCLA Loneliness had good internal consis-
tency (a = .94).
Inconsistency Items
Ten items were used to assess for inconsistency in responding.
Five of the items were presented at the beginning of the initial
survey and the five paired items were presented at the end of
the survey. The items were rated on a 5-point scale from 1 (not
at all true) to 5 (very true). The difference scores for the matched
items were summed, and participants were excluded if they had
scores of 7 or higher (Maniaci & Rogge, 2014). Two people were
excluded from the study due to scores of 7 or higher.
124 HARPER ET AL.
WITHIN-PERSON MEASURES (ESM ITEMS)
Expectations
Expectations for future social interactions were measured with
three items that followed one item stem: “These questions are
about your expectations with others (e.g., friends, parents, ro-
mantic partners, siblings, classmates, and new people).” The
three items assessed how much people anticipated their inter-
actions in the future would be difficult, enjoyable, and unpleasant.
These items were rated on a 5-point scale from 1 (not at all) to
5 (extremely). Of note, these items captured both positive and
negative expectations for the future. Therefore, the enjoyable
item was reverse-scored, and the mean was computed from the
three items. Based on results from a multilevel confirmatory fac-
tor analysis (MCFA), the scale for expectations was reliable (ω =
.706).
Social Hopelessness
Social hopelessness was measured with a single item that was
used to assess social hopelessness in a previous study (Roxbor-
ough et al., 2012). The item wording was modified in the current
study to capture experiences of social hopelessness since the last
survey completed (i.e., To what extent have you felt hopeless
about your relationships in the past three hours?). The item was
rated on a 5-point scale from 1 (not at all) to 5 (extremely).
Measure of Loneliness
Daily loneliness was assessed using items from the UCLA Lone-
liness Scale (Russell, 1996). Six items were selected based on a
previous daily diary study (Hawkley, Preacher, & Cacioppo,
2012). The wording of the stem was changed to capture per-
ceived isolation since the last survey completed (i.e., In the past
three hours, to what extent did you feel…). The loneliness scale
was reliable (ω = .886).
PERFECTIONISM AND LONELINESS 125
PROCEDURE
Participants completed informed consent and were provided
an electronic copy of the consent form. Participants then regis-
tered their phone number in SurveySignal (http://surveysignal.
com/) and were routed to Qualtrics (https://www.qualtrics.
com/) where they completed the initial measures, including
demographics, MPS, and UCLA. Beginning the next day par-
ticipants received Qualtrics links for the survey via text message
through SurveySignal three times per day for seven days on a
fixed scheduled—at 10:00 am, 4:00 pm, and 9:00 pm. Every par-
ticipant received their first survey at 10:00 am. Of note, approxi-
mately 35% of participants incorrectly selected the Central time
zone (instead of the Eastern time zone) while registering their
phone number in SurveySignal; therefore, they received their
surveys an hour later, at 11:00 am, 5:00 pm, and 10:00 pm Eastern
Standard Time (EST). The links to the surveys were active for
two hours to ensure that participants completed them within a
specified time period. To increase compliance, participants were
assigned research credit for the initial survey and separate credit
for completing approximately 75% of the surveys (i.e., 15 of the
21 surveys). The study had a good response rate (approximately
84.76%).
ANALYTIC PLAN
Multilevel structural equation modeling (MSEM) was used to
test the hypotheses in the current study. Multilevel structural
equation modeling is appropriate for testing multilevel media-
tional effects that involve both within- and between-person vari-
ation (Asparouhov & Muthén, 2007). The models were specified
at both the within and between level where the ESM variables
(i.e., expectations, social hopelessness, within-day loneliness)
were Level 1 (within-person) variables and perfectionism was
a Level 2 (between-person) variable. This allows for examining
how variations in perfectionism are associated with time-vary-
ing changes in expectations and loneliness using latent variable
random intercepts (Ludtke et al., 2008). For example, in Model
1, at Level 1, loneliness was regressed on expectations and social
hopelessness. At Level 2, the latent variable random intercept for
126 HARPER ET AL.
expectations and the latent variable random intercept for social
hopelessness was regressed on SPP and SOP, and the latent vari-
able random intercept of loneliness was regressed on expecta-
tions and social hopelessness. Of note, analyses captured within
day variation only. The data were coded so that analyses never
regressed loneliness from a given day on expectations or social
hopelessness from the previous day. Moreover, the time period
between expectations (and social hopelessness) and loneliness
was always consistent. For example, if a participant missed the
middle of the day survey, their data from the beginning of the
day would not be used to predict data at the end of the day.
All within-person variables were group-mean (i.e., within-
person) centered, and between-person variables (MPS subscales
and UCLA) were standardized in the models. Socially prescribed
perfectionism and SOP were both included in models to control
for the overlap between the dimensions (Stoeber, Feast, & Hay-
ward, 2009).
The models were run in Mplus 8 with maximum likelihood
estimation with robust standard errors (Muthen & Muthen,
1998–2017) or maximum likelihood estimation with standard er-
rors approximated by the first-order derivatives as recommend-
ed (Asparouhov & Muthén; 2012). Although the bootstrapping
method is typically recommended for examining indirect effects
(Hayes, 2013), two-level analyses in Mplus 8 are not compat-
ible with bootstrapping. Therefore, indirect effects for the path
model were estimated using Sobel test (Sobel, 1982). Research
suggests that standard SEM fit indices (e.g., RMSEA, CFI) can be
biased with MSEM (Ryu & West, 2009). The current study used
Akaike information criterion (AIC), Bayesian information crite-
rion (BIC), and log-likelihood values to evaluate and compare
fit between models (Preacher, Zyphur, & Zhang, 2010). Several
previous studies utilizing MSEM also used AIC and BIC to com-
pare model fit (Dunkley et al., 2003; Mushquash & Sherry, 2012;
Wardell, Ramchandani, & Hendershot, 2015). Akaike informa-
tion criterion and BIC difference scores greater than 10 indicate
strong support for one model over the other, with smaller values
indicating better fit (Burnham & Anderson, 2002; Raftery, 1995).
PERFECTIONISM AND LONELINESS 127
TABLE 1. Means, Standard Deviations, and Bivariate Correlations for Between-Level Variables
N M SD 1. 2. 3. 4. 5. 6. 7.
1. SPP 143 60.20 13.66 —
2. SOP 143 72.52 14.01 0.53*** —
3. UCLA 142 45.56 11.97 0.44*** 0.05 —
4. Social
hopelessness
2,498 1.47 0.81 0.41** 0.01 0.45*** —
5. Expectations 2,500 2.25 0.82 0.40*** 0.02 0.56*** 0.35*** —
6. Loneliness 2,501 1.94 0.82 0.37*** -0.03 0.58*** 0.69*** 0.59*** —
7. LonelinessT2 2,500 1.95 0.85 0.38*** -0.02 0.58*** 0.44*** 0.54*** 0.56*** —
Notes. SPP = socially prescribed perfectionism; SOP = self-oriented perfectionism; UCLA = UCLA Loneliness Scale; LonelinessT2 = loneliness measured at the next
time point, M = Mean; SD = standard deviation, and bivariate correlations are for whole sample (N = 145). For social hopelessness, expectations, loneliness, and
lonelinessT2, N = number of observations. **p < 0.01; ***p < 0.001.
128 HARPER ET AL.
RESULTS
DESCRIPTIVE STATISTICS
Table 1 presents the means, standard deviations, and bivariate
correlations for the study variables. Socially prescribed perfec-
tionism was moderately positively correlated with SOP and trait
loneliness, and weakly positively correlated with expectations
and loneliness (within-person). Intraclass correlation coefficients
(ICC’s) for the within-person variables suggest multilevel mod-
eling was appropriate (ICC for expectations = .55; ICC for loneli-
ness = .51; ICC for social hopelessness = .40; ICC for time-lagged
loneliness = .48).
Model 1: Negative Expectations and Loneliness at the
Concurrent Time Point
To first examine the association between expectations about fu-
ture social interactions, social hopelessness, and loneliness con-
currently, Model 1 examined whether expectations and social
hopelessness mediated between perfectionism (SPP and SOP)
and loneliness where expectations, social hopelessness, and
loneliness were measured at the concurrent time point. In this
model, at the within level, concurrent loneliness was regressed
onto expectations and onto social hopelessness. At the between
level, expectations and social hopelessness were regressed onto
SPP and SOP, while controlling for UCLA scores, and concurrent
loneliness was regressed onto expectations and social hopeless-
ness (see Figure 1). The model fit was AIC = 13,476.93, BIC =
13,638.92, log-likelihood = −6,711.46.
At the within level, expectations significantly predicted lone-
liness (β = 0.22, SE = 0.03, p < 0.001, 95% CI [0.15, 0.30]). Social
hopelessness also significantly predicted loneliness (β = 0.47, SE
= 0.03, p < 0.001, 95% CI [0.37, 0.49]). At the between level, SPP
positively predicted expectations (β = 0.53, SE = 0.07, p < 0.001,
95% CI [0.22, 0.41]) and SOP negatively predicted expectations
(β = −0.29, SE = 0.09, p = 0.01, 95% CI [−0.26, −0.05]), which in
turn significantly predicted loneliness (β = 0.24, SE = 0.08, p =
0.001, 95% CI [0.08, 0.35]). Socially prescribed perfectionism also
PERFECTIONISM AND LONELINESS 129
positively predicted social hopelessness (β = 0.56, SE = 0.07, p
< 0.001, 95% CI [0.20, 0.37]) and SOP negatively predicted so-
cial hopelessness (β = −0.29, SE = 0.09, p = 0.001, 95% CI [−0.24,
−0.05]), which in turn significantly predicted loneliness (β = 0.69,
SE = 0.07, p < 0.001, 95% CI [0.57, 0.89]).
The indirect effects in the model were significant. Expectations
mediated the relation between SPP and loneliness (B = 0.07, SE
= 0.02, p < 0.01, 95% CI [0.02, 0.12]) and the relation between
SOP and loneliness (B = −0.03, SE = 0.02, p = 0.04, 95% CI [−0.07,
−0.001]). Social hopelessness mediated the relation between SPP
and loneliness (B = 0.21, SE = 0.04, p < 0.001, 95% CI [0.14, 0.28])
and the relation between SOP and loneliness (B = −0.11, SE =
0.04, p = 0.004, 95% CI [−0.18, −0.04]). Results suggest that people
high on SPP have higher negative expectations for future social
interactions and higher levels of social hopelessness, which both
in turn explain higher levels of loneliness. People high on SOP,
in contrast, have lower negative expectations and lower social
hopelessness, and in turn have lower levels of loneliness.
FIGURE 1. Between Level Model for Model 1
with Standardized Regression Coefficients
Note. SPP = Socially prescribed perfectionism;
SOP = Self-oriented perfectionism; LonelinessT2 =
loneliness measured at the next time point;
N = 142, *p < .05; **< p .01; ***p< .001.
130 HARPER ET AL.
Model 2: Temporal Effect of Negative Expectation on Loneliness
A second model was tested to examine the temporal relation be-
tween negative expectations, social hopelessness, and loneliness.
Specifically, Model 2 examined whether negative expectations
about future social interactions and social hopelessness pre-
dicted loneliness at the following time point, above and beyond
loneliness at the concurrent time point (controlling for trait lone-
liness; see Figure 2). Model 2 had poorer fit than Model 1 based
on fit indices: AIC = 18,904.19, BIC = 19,126.18, log-likelihood =
−9,415.10.
At the within-person level of the multilevel model, the path
from negative expectations to loneliness at the following time
point was significant when controlling for loneliness at the con-
current time point (β = 0.06, SE = 0.02, p = 0.004, 95% CI [0.02,
0.10]). However, social hopelessness did not significantly predict
loneliness at the following time point when controlling for lone-
liness at the current time point (β = 0.02, SE = 0.03, p = 0.60, 95%
CI [−0.03, 0.08]).
At the between-person level, the paths from SPP and SOP to
expectations were both significant (SPP: β = 0.53, SE = 0.11, p <
0.001, 95% CI [0.32, 0.72]; SOP: β = −0.26 SE = 0.12 p = 0.03, 95%
CI [−0.47, −0.05]). Additionally, the paths form SPP and SOP to
social hopelessness were significant (SPP: β = 0.56, SE = 0.12, p <
0.001, 95% CI [0.25, 0.78]; SOP: β = −0.29, SE = 0.14, p = 0.04, 95%
CI [−0.53, −0.04]). However, the path from expectations to loneli-
ness at the next time point, while controlling for loneliness at the
concurrent time point, was not significant β = −0.01, SE = 0.28, p
= 0.97, 95% CI [−0.20, 0.26]). The path from social hopelessness
to loneliness at the next time point was also not significant (β =
−0.04, SE = 0.37, p = 0.91, 95% CI [−0.24, 0.41]). Loneliness at the
concurrent time point significantly predicted loneliness at the
next time point β = 1.04, SE = 0.37, p < 0.001, 95% CI [0.87, 1.10]).
Additionally, the indirect effects of SPP and SOP predicting
loneliness at the following time point through expectations (SPP:
B = −0.003, SE = 0.08, p = 0.97, 95% CI [−0.15, 0.15]; SOP: B =
0.002, SE = 0.04, p = 0.97, 95% CI [−0.07, 0.08]) were not signifi-
cant. The indirect effects of SPP and SOP predicting loneliness
through social hopelessness (SPP: −B = 0.01, SE = 0.11, p = 0.91,
PERFECTIONISM AND LONELINESS 131
95% CI [−0.23, 0.20]; SOP: B = 0.01, SE = 0.06, p = 0.91, 95% CI
[−0.10, 0.11]) were also not significant.
Model 3: Expectations Predicting Loneliness at the Following
Time Point
In the last model, we tested the same paths as Model 2 without
controlling for loneliness at the concurrent time point as expec-
tations and social hopelessness (see Figure 3). Model 3 had bet-
ter fit than Model 2 but poorer fit compared to Model 1: AIC =
14,594.87, BIC = 14,756.86, Log-likelihood = −7,270.44.
At the within level, expectations and social hopelessness pre-
dicted loneliness at the following time point (expectations: β =
0.08, SE = 0.02, p < 0.001, 95% CI [0.04, 0.12]; social hopelessness:
β = 0.07, SE = 0.02, p = 0.001, 95% CI [0.02, 0.11]). At the between
level, similar to the previous models, SPP and SOP differently
predicted expectations (SPP: β = 0.53, SE = 0.11, p < 0.001, 95%
CI [0.32, 0.73]; SOP: β = −0.26, SE = 0.10, p = 0.01, 95% CI [−0.46,
FIGURE 2. Between Level Model for Model 2 with
Standardized Regression Coefficients
Note. SPP = Socially prescribed perfectionism; SOP =
Self-oriented perfectionism; LonelinessT1 = loneliness
measured at the concurrent time point as Expectations
and Social hopelessness; LonelinessT2 = loneliness
measured at the next time point; N = 142.
p< .05; **p< .01; ***p< .001
132 HARPER ET AL.
−0.06]) and social hopelessness (SPP: β = 0.56, SE = 0.11, p <
0.001, 95% CI [0.34, 0.79]; SOP: β = −0.29, SE = 0.12, p = 0.02, 95%
CI [−0.52, −0.05]). Additionally, expectations and social hope-
lessness predicted loneliness at the following time point (expec-
tations: β = 0.23, SE = 0.08, p = 0.01, CI 95% [0.07, 0.39]; social
hopelessness: β = 0.73, SE = 0.07, p < 0.001, CI 95% [0.59, 0.88]).
Similar to Model 1, the indirect effect from SPP to loneliness
through expectations was significant (B = 0.06, SE = 0.03, p =
0.02, 95% CI [0.01, 0.12]); however, the indirect effect from SOP
to loneliness through expectations was not significant (B = −0.03,
SE = 0.02, p = 0.07, 95% CI [−0.06, 0.00]). Also similar to Model
1, indirect effects from SPP and SOP to loneliness through social
hopeless were significant (SPP: B = 0.21, SE = 0.06, p < 0.001,
95% CI [0.01, 0.12]; SOP: B = −0.11, SE = 0.05, p = 0.03, 95% CI
[−0.21, −0.01]). People high on SPP had higher negative expecta-
tions about future social interactions and higher levels of social
hopelessness, which in turn both predicted higher levels of lone-
liness later in the day. In contrast, people high on SOP had lower
negative expectations and lower social hopelessness; however,
FIGURE 3. Between Level Model for Model 3
with Standardized Regression Coefficients
Notes. SPP = Socially prescribed
perfectionism; SOP = Self-oriented
perfectionism; LonelinessT2 = loneliness
measured at the next time point; N = 142.
*p < .05; **p < .01; ***p < .001.
PERFECTIONISM AND LONELINESS 133
only social hopelessness explained the relation between SOP and
loneliness later in the day.
DISCUSSION
The purpose of the current study was to examine whether nega-
tive expectations for future social interactions and social hope-
lessness in daily life explained variations in loneliness in perfec-
tionists. Negative expectations and social hopelessness explained
the relation between SPP and loneliness both at the concurrent
time point and the following time point. At the concurrent time
point, both negative expectations and social hopelessness ex-
plained the relation between SOP and loneliness; however, when
loneliness was measured at the following time point, only social
hopelessness mediated between SOP and loneliness. Although
negative expectations and social hopelessness did not hold up to
a more rigorous test of the temporal relationship with loneliness
at the between-person level, negative expectations predicted
loneliness at the following time point (controlling for concurrent
loneliness) at the within-person level. This suggests that expecta-
tions about social interactions is an important factor in explain-
ing variations in loneliness in daily life.
The two dimensions of perfectionism differently predicted
expectations about future social interactions. Whereas SPP had
a positive association with negative expectations for future so-
cial interactions, SOP had a negative association with negative
expectations. Although little research exists on perfectionists’
expectations for social interactions, previous studies suggest
that people high on SPP have lower social self-efficacy (Martin,
Flett, Hewitt, Krames, & Szanto, 1996), which may explain the
higher levels of negative expectations for future social interac-
tions. Additionally, these findings are consistent with research
that suggests that people high on SPP have more negative fu-
ture thinking (O’Connor et al., 2004). Regarding SOP, research
has not focused on self-oriented perfectionists’ expectations for
social events (only achievement events). Research on SOP sug-
gests that they have positive achievement expectations (Brown
et al., 1999) and positive future thinking (O’Connor et al., 2004;
Stoeber & Corr, 2017). To our knowledge, this is the first study
134 HARPER ET AL.
to examine perfectionism and social expectations in daily life;
therefore, the literature would benefit from future replication of
these associations.
Additionally, SPP and SOP differently predicted social hope-
lessness. Consistent with previous research, SPP was positively
associated with social hopelessness (Roxborough et al., 2012).
Although a previous study found no relation between SOP and
social hopelessness (Roxborough et al., 2012), the current study
found that people high on SOP had lower levels of social hope-
lessness. One possible explanation for the difference between the
current finding and previous finding is the different methods
of measuring social hopelessness. Roxborough and colleagues
(2012) had adolescent participants rate how often they felt hope-
less about their social relationships. The current study examined
perceptions of social hopelessness across the day, for multiple
days suggesting that SOP may be associated with within-person
variations of social hopelessness.
Although the mediating effect of expectations and social hope-
lessness (between SPP and loneliness) was nonsignificant when
controlling for concurrent loneliness, both expectations and so-
cial hopelessness predicted loneliness concurrently and later day
loneliness. Moreover, having negative expectations about future
social interactions and social hopelessness explained variability
in loneliness during the day above trait levels of loneliness. These
findings suggest an important relation between socio-cognitive
variables (i.e., expectations for social interactions and social
hopelessness) and loneliness for perfectionists. People high on
SPP appear to have more negative cognitions about their social
interactions and relationships, which in turn is associated with
higher levels of loneliness. Because socially prescribed perfec-
tionists are at an increased risk for depression (Shafran & Man-
sell, 2001) and have higher suicidality (Hewitt et al., 2006), and
loneliness is associated with both depression and suicidality (see
Heinrich & Gullone, 2006 for a review), future research on the
role of negative social cognitions and loneliness in perfectionists
is warranted.
In contrast to people high on SPP, expectations about future
social interactions and social hopelessness emerged as adaptive
cognitive processes for people high on SOP. People high on SOP
PERFECTIONISM AND LONELINESS 135
tended to have less negative thinking regarding their future in-
teractions and social relationships, which in turn explained lower
levels of loneliness. Interestingly, it appears that lower levels of
social hopelessness are especially important in explaining lone-
liness for people high on SOP. Self-oriented perfectionism was
indirectly associated with loneliness through social hopelessness
when loneliness was measured at the concurrent time point and
at the following time point; however, negative expectations only
explained the relation between SOP and loneliness at the concur-
rent time point. Overall, the findings with SOP, social cognitions,
and loneliness are similar to Dunkley et al.’s (2014) findings with
high personal standards and positive reinterpretations. Self-ori-
ented perfectionism was not directly associated with daily lone-
liness; however, through cognitions about social interactions and
relationships, SOP was indirectly associated with loneliness. Fu-
ture research would benefit from continuing to examine wheth-
er cognitions in daily life explain indirect associations between
SOP and affect.
A limitation of the current finding is the correlational nature of
the models. Although expectations explained variation in loneli-
ness at the following time point above loneliness at the concur-
rent time point in the within person level, this did not hold up at
the between person level. Additionally, social hopelessness did
not predict future loneliness (while controlling for concurrent
loneliness). Therefore, the causal direction of the relationships
between these socio-cognitive factors and loneliness is not fully
clear. The hopelessness theory of depression suggests that socio-
cognitive factors (expectations and hopelessness) would lead to
feeling disconnected (Abramson, Metalsky, & Alloy, 1989). How-
ever, the loneliness theory posits that lonely people tend to have
negative perceptions of others and negative social expectations,
which in turn leads to behaviors that sustain loneliness (Ca-
cioppo & Hawkley, 2014; Cacioppo & Hawkley, 2009). Empirical
studies suggest that lonely people have higher negative percep-
tions of others (Wittenberg & Reis, 1986) and perceive relation-
ship and communication quality as more negative (Duck, Pond,
& Learham, 1994) compared to non-lonely people. More nega-
tive perceptions of others likely lead to more negative expecta-
tions of others. Therefore, more longitudinal studies are needed
136 HARPER ET AL.
to tease apart the relationships between negative social expecta-
tions, social hopelessness, and loneliness.
Despite this limitation, the findings have implications for the
treatment of perfectionism. Results from the current study sug-
gest that socially prescribed perfectionists had higher negative
expectations and higher levels of social hopelessness, which may
explain higher levels of loneliness. Based on the loneliness theory
(Cacioppo & Hawkley, 2014) it is also possible that this is a cycli-
cal relation where people who are lonely, such as people high on
SPP, also have negative expectations of others and feel hopeless
about their social relationships. Therefore, targeting processes
that lead to (or result from) loneliness, may be effective in reduc-
ing loneliness and negative affect. Cognitive behavioral therapy
(CBT) is the most prominent treatment for perfectionism. Cogni-
tive behavioral therapy for perfectionism uses cognitive restruc-
turing techniques to identify more adaptive/useful cognitions.
If perfectionists are reporting high levels of loneliness, it may
be useful to examine social expectations and feelings of social
hopelessness to potentially target cognitions using cognitive re-
structuring techniques.
The purpose of the current study was to examine whether ex-
pectations for future social interactions and social hopelessness
explained the relation between two dimensions of perfectionism
and loneliness. Although the findings should be replicated using
longitudinal designs, they provide support that people high on
SPP have higher negative expectations about social interactions
and higher levels of social hopelessness in their everyday lives,
which is related to higher levels of loneliness. Therefore, cog-
nitions related to social expectations and relationships may be
an effective treatment target to decrease loneliness and improve
mood in perfectionists. Findings also suggest that cognitions
related to social expectations and relationships may be factors
that explain an indirect association between SOP and loneliness.
Future research on the role that socio-cognitive factors play in
explaining the relation between different dimensions of perfec-
tionism and mood in daily life is needed.
PERFECTIONISM AND LONELINESS 137
REFERENCES
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A the-
ory-based subtype of depression. Psychological Review, 96, 358.
Asparouhov, T., & Muthén, B. (2007). Computationally efficient estimation of multilev-
el high dimensional latent variable models. Proceedings of the Joint Statistical
Meeting in Salt Lake City, August 2007. ASA section on Biometrics.
Asparouhov, T., & Muthén, B. (2012). Saddle points. Retrieved from: https://www.
statmodel.com/download/Saddle%20point.pdf
Brown, E. J., Heimberg, R. G., Frost, R. O., Makris, G. S., Juster, H. R., & Leung, A.
W. (1999). Relationship of perfectionism to affect, expectations, attributions and
performance in the classroom. Journal of Social and Clinical Psychology, 18, 98–120.
Burnham, K. P., & Anderson, D. R. (2002). Model selection and multimodel inference. A
practical information-theoretic approach (2nd ed.). New York, NY: Springer.
Cacioppo, J. T., & Hawkley, L. C. (2009). Perceived social isolation and cognition. Trends
in Cognitive Science, 13, 447–454. https://doi.org/10.1016/j.tics.2009.06.005
Cacioppo, J. T., & Hawkley, L. C. (2014). People thinking about people: The vicious
cycle of being a social outcast in one’s own mind. In K. D. Williams, J. P. Forgas,
& W. Hippel (Eds.), The Social Outcast: Ostracism, Social Exclusion, Rejection, and
Bullying. (pp. 91–108). New York: Psychology Press.
Chang, E. C. (2013). Perfectionism and loneliness as predictors of depressive and
anxious symptoms in Asian and European Americans: Do self-construal sche-
mas also matter? Cognitive Therapy and Research, 37, 1179–1188. https://doi.
org/10.1007/s10608-013-9549-9
Chang, E. C., Sanna, L. J., Chang, R., & Bodem, M. R. (2008). A preliminary look at
loneliness as a moderator of the link between perfectionism and depressive and
anxious symptoms in college students: Does being lonely make perfectionistic
strivings more distressing? Behaviour Research and Therapy, 46, 877–886. https://
doi.org/10.1016/j.brat.2008.03.012
Conner, T. S., Tennen, H., Fleeson, W., & Barrett, L. F. (2009). Experience sampling
methods: A modern idiographic approach to personality research. Social and
Personality Psychology Compass, 3, 292–313. https://doi.org/10.1111/j.1751-
9004.2009.00170.x
Duck, S., Pond, K., & Leatham, G. (1994). Loneliness and the evaluation of relation-
al events. Journal of Social and Personal Relationships, 11, 253-276. https://doi.
org/10.1177/0265407594112006
Dunkley, D. M., Ma, D., Lee, I. A., Preacher, K. J., & Zuroff, D. C. (2014). Advancing
complex explanatory conceptualizations of daily negative and positive affect:
Trigger and maintenance coping action patterns. Journal of Counseling Psychol-
ogy, 61, 93–109. https://doi.org/10.1037/a0034673
Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self-critical perfectionism
and daily affect: Dispositional and situational influences on stress and cop-
ing. Journal of Personality and Social Psychology, 84, 234–252. https://doi.
org/10.1037/0022-3514.84.1.234
Flett, G. L., Druckman, T., Hewitt, P. L., & Wekerle, C. (2012). Perfectionism, coping,
social support, and depression in maltreated adolescents. Journal of Rational —
Emotive & Cognitive — Behavior Therapy, 30, 118–131. http://.doi.org/10.1007/
s10942-011-0132-6
138 HARPER ET AL.
Hawkley, L. C., Preacher, K. J., & Cacioppo, J. T. (2010). Loneliness impairs daytime
functioning but not sleep duration. Health Psychology, 29, 124–129. https://doi.
org/10.1037/a0018646
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analy-
sis. New York: Guilford.
Heinrich, L. M., & Gullone, E. (2006). The clinical significance of loneliness: A litera-
ture review. Clinical Psychology Review, 26, 695–718. https://doi.org/10.1016/j.
cpr.2006.04.002
Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: Con-
ceptualization, assessment, and association with psychopathology. Journal of
Personality and Social Psychology, 60, 456–470. https://doi.org/10.1037/0022-
3514.60.3.456
Hewitt, P. L., Flett, G. L., Sherry, S. B., & Caelian, C. (2006). Trait perfectionism dimensions
and suicidal behavior. In T. E. Ellis (Ed), Cognition and suicide: Theory, research,
and therapy. (pp. 215–235). Washington, DC: American Psychological Associa-
tion.
Joiner, T. E., Jr., & Schmidt, N. B. (1995). Dimensions of perfectionism, life stress, and
depressed and anxious symptoms: Prospective support for diathesis-stress but
not specific vulnerability among male undergraduates. Journal of Social and Clin-
ical Psychology, 14, 165–183. http://doi.org/10.1521/jscp.1995.14.2.165
Laurenti, H. J., Bruch, M. A., & Haase, R. F. (2008). Social anxiety and SPP: Unique
and interactive relationships with maladaptive appraisal of interpersonal situa-
tions. Personality and Individual Differences, 45, 55–61. https://doi.org/10.1016/j.
paid.2008.02.018
Ludtke, O., Marsh, H. W., Robitzsch, A., Trautwein, U., Asparouhov, T., & Muthén, B.
(2008). The multilevel latent covariate model: A new, more reliable approach
to group-level effects in contextual studies. Psychological Methods, 13, 203–229.
MacLeod, A. K., & Byrne, A. (1996). Anxiety, depression, and the anticipation of future
positive and negative experiences. Journal of Abnormal Psychology, 105, 286–289.
https://doi.org/10.1037/0021-843X.105.2.286
Maniaci, M. R., & Rogge, R. D. (2014). Caring about carelessness: Participant inatten-
tion and its effects on research. Journal of Research in Personality, 48, 61–83.
Martin, T. R., Flett, G. L., Hewitt, P. L., Krames, L., & Szanto, G. (1996). Personal-
ity correlates of depression and health symptoms: A test of a self-regulation
model. Journal of Research in Personality, 30, 264–277. https://doi.org/10.1006/
jrpe.1996.0017
Mushquash, A. R., & Sherry, S. B. (2012). Understanding the socially prescribed per-
fectionist’s cycle of self-defeat: A 7-day, 14-occasion daily diary study. Journal
of Research in Personality, 46, 700–709. https://doi.org/10.1016/j.jrp.2012.08.006
Muthen, L. K., & Muthen, B. O. (1998–2017). Mplus User’s Guide. Eighth Edition. Los
Angeles, CA: Muthen & Muthen.
O’Connor, R. (2007). The relations between perfectionism and suicidality: A systematic
review. Suicide and Life-Threatening Behavior, 37, 698–714.
O’Connor, R., O’Connor, D., O’Connor, S., Smallwood, J., & Miles, J. (2004). Hopeless-
ness, stress, and perfectionism: The moderating effects of future thinking. Cogni-
tion & Emotion, 18, 1099–1120.
PERFECTIONISM AND LONELINESS 139
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework
for assessing multilevel mediation. Psychological Methods, 15, 209–233. https://
doi.org/10.1037/a0020141
Raftery, A. E. (1995). Bayesian model selection in social research. Sociological Methodol-
ogy, 25, 111–163.
Rice, K. G., Leever, B. A., Christopher, J., & Porter, J. D. (2006). Perfectionism, stress,
and social (dis)connection: A short-term study of hopelessness, depression, and
academic adjustment among honors students. Journal of Counseling Psychology,
53, 524–534. https://doi.org/10.1037/0022-0167.53.4.524
Roxborough, H. M., Hewitt, P. L., Kaldas, J., Flett, G. L., Caelian, C. M., Sherry, S., &
Sherry, D. L. (2012). Perfectionistic self-presentation, socially prescribed per-
fectionism, and suicide in youth: A test of the Perfectionism Social Discon-
nection Model. Suicide and Life-Threatening Behavior, 42, 217–233. https://doi.
org/10.1111/j.1943-278X.2012.00084.x
Russell, D. W. (1996). UCLA Loneliness Scale (Version 3): reliability, validity, and factor
structure. Journal of Personality Assessment, 66, 20–40.
Ryu, E., & West, S. G. (2009). Level-specific evaluation of model fit in multilevel struc-
tural equation modeling. Structural Equation Model, 16, 583–601. http://doi.
org/10.1080/10705510903203466
Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathology: A review of
research and treatment. Clinical Psychology Review, 21, 879–906. http://doi.
org/10.1016/S0272-7358(00)00072-6
Smith M. M., Sherry S. B., Chen S., Saklofske, D. H., Mushquash, C., Flett, G. L., &
Hewitt, P. L. (2017) The perniciousness of perfectionism: A meta-analytic review
of the perfectionism–suicide relationship. Journal of Personality, 1–21. https://
doi.org/10.1111/jopy.12333
Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural
equation models. In S. Leinhardt (Ed.) Sociological Methodology 1982. (pp. 290–
312). Washington DC: American Sociological Association.
Stoeber, J., & Corr, P. J. (2017). Perfectionism, personality, and future-directed thinking:
Further insights from revised Reinforcement Sensitivity Theory. Personality and
Individual Differences, 105, 78–83. https://doi.org/10.1016/j.paid.2016.09.041
Stoeber, J., Feast, A. R., & Hayward, J. A. (2009). Self-oriented and socially prescribed
perfectionism: Differential relationships with intrinsic and extrinsic motivation
and test anxiety. Personality and Individual Differences, 47, 423–428. https://doi.
org/10.1016/j.paid.2009.04.014
Wardell, J. D., Ramchandani, V. A., & Hendershot, C. S. (2015). A multilevel structural
equation model of within- and between-person associations among subjective
responses to alcohol, craving, and laboratory alcohol self-administration. Jour-
nal of abnormal psychology, 124, 1050–1063. https://doi.org/10.1037/abn0000121
Wittenberg, M.m & Reis, H. (1986). Loneliness, social skills and social percep-
tion. Personality and Social Psychology Bulletin, 12 121-130. https://doi.
org/10.1177/0146167286121012
