Personality and Individual Differences 52 (2012) 828–832

Contents lists available at SciVerse ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier .com/locate /paid

Perfectionism, mattering, and depression: A mediational analysis

Gordon L. Flett a, Ingrid Galfi-Pechenkov a,⇑, Danielle S. Molnar a, Paul L. Hewitt b, Abby L. Goldstein c

a York University, 4700 Keele Street, Toronto, ON, Canada M3J 1P3b University of British Columbia, 3404-2136 West Mall, Vancouver, BC, Canada V6T 1Z4c University of Toronto, 252 Bloor Street, Toronto, ON, Canada M5S 1V5

a r t i c l e i n f o a b s t r a c t

Article history:Received 27 August 2011Received in revised form 17 December 2011Accepted 29 December 2011Available online 9 February 2012

Keywords:PerfectionismMatteringDistressSocial disconnectionMediation

0191-8869/$ - see front matter � 2012 Elsevier Ltd. Adoi:10.1016/j.paid.2011.12.041

⇑ Corresponding author. Address: Department of4700 Keele Street, Toronto, ON, Canada M3J 1P3. Tel.:+1 416 736 5814.

E-mail address: ingridg@yorku.ca (I. Galfi-Pechenk

While past research has examined the association between perfectionism and low self-esteem, researchhas not explored the possible link between perfectionism and feelings of not mattering to others. Themattering construct and feelings of insignificance to others are relevant to an understanding of interper-sonal pressures to be perfect. In the current study, a sample of 246 university students completed theMultidimensional Perfectionism Scale, the Perfectionistic Self-Presentation Scale, the PerfectionismCognitions Inventory, and Scales tapping mattering and depressive symptoms. Correlational analysesconfirmed that negative appraisals of mattering were associated with elevated levels of depressivesymptoms, socially prescribed perfectionism, and perfectionistic self-presentation. Additional analysesfound that mattering was a partial mediator of the link between interpersonal perfectionism and depres-sion. Our analyses suggest that certain perfectionistic students may be particularly at risk as a result ofthe association between low perceived mattering and chronic interpersonal pressures to be perfect.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction are neurotic perfectionists with low self-esteem, narcissistic per-

Research has established that perfectionism plays a vitalcontributing role in the vulnerability to and the persistence ofdepression (e.g., Hewitt & Flett, 1991; Kawamura, Hunt, Frost, &DiBartolo, 2001). Other research indicates that distressed perfec-tionists are less responsive to treatment than people with lowerlevels of perfectionism, and longer interventions are needed toachieve positive outcomes (see Blatt & Zuroff, 2002). Given thelinks with such important outcomes, researchers have sought toidentify factors that influence this relation. One mechanism thathas not been considered thus far is the importance of perceivedmattering – that is, feeling that one has social significance andmakes a difference in the lives of others. Accordingly, the currentstudy explored the associations among trait dimensions ofperfectionism, perceived mattering, and depression in universitystudents.

While perfectionism is believed to reflect feelings of inferiorityand negative feelings about the self, this may not always be thecase. Classic self-regulation models (e.g., Kanfer & Hagerman,1981) distinguish between the pursuit of impossible standardsand the self-evaluative reactions to the attainment or non-attainment of these standards. Indeed, while most perfectionists

ll rights reserved.

Psychology, York University,+1 416 736 2100x33727; fax:

ov).

fectionists would have an inflated sense of themselves and theirperceived significance to others (see Sorotzkin, 1985).

Rosenberg and McCullough (1981) defined mattering as ‘‘thefeeling that others depend on us, are interested in us, are con-cerned with our fate, or experience us as an ego-extension.’’ (ascited in Taylor & Turner, 2001). This sense of being significantand important to other people should be instrumental as a key pro-tective resource that should help buffer life stressors and setbacks.Taylor and Turner (2001) described the four sources of feeling thatone matters as: (1) dependence (i.e., obligations arising from socialbonds and the perception that one’s actions toward others willhave an effect on them); (2) importance (i.e., the perception thatwe are of interest and concern to others); (3) attention (i.e.,perceiving one’s actions as being noticed and acknowledged byimportant others); and (4) ego-extension (i.e., the perception thatone could bring about joy or disappointment through personalsuccesses or failures).

It is not very surprising that mattering is conceptualized asessential to overall well-being. Several researchers have found thatamong adults, a stronger perceived sense of mattering predicts lessdepression and greater self-esteem (e.g., Elliott, Kao, & Grant, 2004;Marshall, 2001; Schieman & Taylor, 2001; Taylor & Turner, 2001).Mattering has been shown to come about from social exchanges(Schieman & Taylor, 2001) with previous studies finding associa-tions between forms of social integration and mattering (e.g.,Bakan, 1966; Baumeister & Leary, 1995). These studies reflect thepremise that well-being is related fundamentally to rewardingsocial relationships (Taylor & Turner, 2001).

Perfectionism Depression Mattering

Respondents' Sex

Fig. 1. Hypothesized model of perfectionism, mattering, and depression.

G.L. Flett et al. / Personality and Individual Differences 52 (2012) 828–832 829

1.1. Dimensions of perfectionism and the social disconnection model

To our knowledge, no study thus far has investigated the possi-bility that perfectionists are particularly vulnerable to depressionand associated feelings of distress as a result of feeling that theydo not matter to others. How can a focus on mattering contributeto our understanding of perfectionism? Perfectionism has beenconceptualized as a highly interpersonal construct with implica-tions for problems in interpersonal relationships (see Habke &Flynn, 2002). The need to matter should be particularly pertinentfor perfectionists with a heightened sense of interpersonal sensi-tivity. Accordingly, perfectionism was conceptualized and assessedin the current study according to the model outlined by Hewitt andFlett (1991). Several perfectionism measures were included tobroadly evaluate possible links between mattering and perfection-ism but with a particular emphasis on how perfectionism dimen-sions that reflect interpersonal needs and sensitivity (i.e., sociallyprescribed perfectionism, perfectionistic self-presentation) wouldbe linked with negative perceptions of mattering.

A link between interpersonal perfectionism and low matteringwould be in keeping with the perfectionism social disconnectionmodel (see Hewitt, Flett, Sherry, & Caelian, 2006). This model isbased on the premise that a distancing from the social environ-ment caused by interpersonal dimensions of perfectionismcontributes to depressive symptoms. In response to growing evi-dence for interpersonal conflict as a consequence of perfectionism,Sherry, Law, Hewitt, Flett, and Besser (2008) investigated the roleof conflicted interpersonal relationships in influencing depressivesymptoms. Their study found support for the perfectionism socialdisconnection model, showing that interpersonal perfectionism isassociated with low perceived social support and low perceivedsocial support acts as a mediator in the interpersonal perfectionism-depression link. Our current emphasis on perfectionism andmattering seems quite relevant to a social disconnection perspec-tive in that perfectionists who need attention and value approvalwill be prone to depression after having determined, rightly orwrongly, that they are disconnected and do not matter to otherpeople. The specific predictions derived from the perfectionismsocial disconnection model account for our focus on mattering asit relates to interpersonal perfectionism per se rather than relatedconstructs such as self-esteem. This emphasis reflects the model’semphasis on the distress that ensues when perfectionists feeldisconnected and perceive that they do not matter to otherpeople.

1.2. Goals and hypotheses of the current study

The initial goal of this study was to examine how various mea-sures and associated conceptualizations of perfectionism relate toperceived mattering. As noted earlier, several perfectionism mea-sures were included in the current study, including measures oftrait perfectionism, perfectionistic self-presentation, and perfec-tionistic thoughts. The three trait perfectionism dimensionsassessed were self-oriented perfectionism (i.e., unrealistic stan-dards imposed on oneself); other-oriented perfectionism (i.e., thedemand for others to be perfect); and socially prescribed perfec-tionism (i.e., the perception that others demand perfection fromus) (see Hewitt & Flett, 1991). Perfectionistic self-presentation isan expressive feature of perfectionism that involves the need toappear perfect (i.e., perfectionistic self-promotion) and concealmistakes (i.e., the need to avoid displaying and disclosing imper-fections to others) (see Hewitt et al., 2003). Finally, perfectionismalso involves automatic and intrusive thoughts revolving aroundperfectionistic themes (Flett, Hewitt, Blankstein, & Gray, 1998).Work on perfectionistic automatic thoughts indicates that thesethoughts predict unique variance in indices of distress that is not

accounted for by existing trait measures of perfectionism (Flettet al., 1998).

In addition to examining the link between perfectionism andmattering, an overarching goal of the present study was to investi-gate the mediating role of mattering to significant others in the linkbetween perfectionism and depression. This follows generally fromresearch indicating that self-esteem mediates the link betweenperfectionism and depression (Preusser, Rice, & Ashby, 1994; Rice,Ashby, & Slaney, 1998) and low conditional self-acceptance is apartial mediator of the link between socially prescribed perfection-ism and depression (Flett, Besser, Davis, & Hewitt, 2003). As can beseen in Fig. 1, we expected perfectionism to be associated nega-tively with depression and that this association would be partiallymediated by mattering, such that greater perfectionism would beassociated with reduced levels of mattering, which, in turn wouldbe associated with greater depression. This possibility is consistentwith research that indicates that perceived social support mediatesthe link between socially prescribed perfectionism and depression,and represents a unique contribution as no study to date has con-ceptualized the perfectionism-depression link as related to theconstruct of mattering.

2. Method

2.1. Participants

The participants were 246 first year university students at YorkUniversity (91 men, 155 women). The mean age of the participantswas 20.3 years (SD = 3.7) who volunteered for a study examiningtransition to university and collegiate health. Each student re-ceived course credit in their introductory psychology class for par-ticipation. Information on cultural background was not obtained inaccordance with provincial policy. Study measures administeredall showed good internal consistency with alpha coefficients rang-ing from 0.73 to 0.93 (see Table 1). Study measure descriptionsfollow.

2.2. Measures

The Multidimensional Perfectionism Scale (MPS; Hewitt & Flett,1991). The MPS is a 45-item measure tapping self-oriented perfec-tionism (e.g., ‘‘One of my goals is to be perfect in everything I do’’),other-oriented perfectionism (e.g., ‘‘I have high expectations forthe people who are important to me’’), and socially prescribedperfectionism (e.g., ‘‘My family expects me to be perfect’’). Exten-sive evidence attests to this instrument’s psychometric properties(Hewitt & Flett, 2004; Hewitt, Flett, Turnbull-Donovan, & Mikail,1991).

The Perfectionistic Self-Presentation Scale (PSPS; Hewitt et al.,2003). The PSPS is a 27-item questionnaire that measures perfec-tionistic self-promotion (e.g., ‘‘I always try to present a picture ofperfection’’), non-display of imperfection (e.g., ‘‘I will do almosteverything to cover up a mistake.’’), and non-disclosure of imper-fection (e.g., ‘‘I should always keep my problems to myself.’’). ThePSPS subscales have shown strong test–retest reliability andconvergent and discriminant validity (Hewitt et al., 2003)

Table 1Means, standard deviations, coefficient alphas, and first-order correlations.

1 2 3 4 5 6 7 8 9

1. SOP –2. SPP 0.36⁄⁄ –3. OOP 0.54⁄⁄ 0.38⁄⁄ –4. Promote 0.57⁄⁄ 0.56⁄⁄ 0.42⁄⁄ –5. Nondisplay 0.37⁄⁄ 0.48⁄⁄ 0.30⁄⁄ 0.79⁄⁄ –6. Nondisclosure 0.37⁄⁄ 0.56⁄⁄ 0.36⁄⁄ 0.59⁄⁄ 0.59⁄⁄ –7. PCI 0.50⁄⁄ 0.55⁄⁄ 0.34⁄⁄ 0.58⁄⁄ 0.50⁄⁄ 0.54⁄⁄ –8. Mattering �0.01 �0.33⁄⁄ 0.00 �0.18⁄⁄ �0.29⁄⁄ �0.34⁄⁄ �0.18⁄⁄ –9. Depression 0.11 0.45⁄⁄ 0.06 0.34⁄⁄ 0.37⁄⁄ 0.39⁄⁄ 0.45⁄⁄ �0.44⁄⁄ –M 69.88 57.02 58.17 42.36 42.87 23.54 50.40 3.21 19.43SD 15.66 12.60 10.56 11.15 10.84 7.43 19.20 0.55 10.47a 0.90 0.83 0.73 0.88 0.88 0.80 0.93 0.81 0.90

Note: ⁄⁄p < 0.01, p < 0.05, n = 246.Note: SOP = self-oriented perfectionism; SPP = socially prescribed perfectionism; OOP = other-oriented perfectionism; PCI = perfectionistic cognitions.

830 G.L. Flett et al. / Personality and Individual Differences 52 (2012) 828–832

Perfectionism Cognitions Inventory (PCI; Flett et al., 1998). ThePCI is a 25-item measure of the frequency of automatic thoughtsabout perfectionistic themes (e.g., ‘‘Why can’t I be perfect?’’).Respondents indicate how frequently each thought had occurredover a 1-week period. This measure is unidimensional and has highand incremental validity (Flett, Hewitt, Whelan, & Martin, 2007;Flett et al., 1998).

The Rosenberg Mattering Scale (RMS; Rosenberg & McCullough,1981). The RMS is a five-item measure of how much one perceivesthey matter to others. A sample item is, ‘‘How important are you toothers?’’ Factor analysis confirmed this measure is unidimensional(Taylor & Turner, 2001), and it is valid in a variety of contexts (Tay-lor & Turner, 2001).

The Center for Epidemiologic Studies Depression Scale (CES-D;Radloff, 1977). The CES-D scale is a 20-item measure of the levelof depressive symptoms within the past week. The CES-D hasshown adequate test–retest reliability and construct validity inboth clinical and nonclinical samples (Radloff, 1977).

3. Results

3.1. Descriptive information

Means, standard deviations, and bivariate correlations betweenall model variables are presented in Table 1. Results from indepen-dent t-tests found that women, relative to men, had higher levels ofmattering (t244 = 2.08, p = 0.04, d = 0.27), and depressive symptom-atology (t244 = 2.28, p = 0.02, d = 0.29). There were no other signif-icant gender differences.

First-order correlations confirmed that socially prescribedperfectionism, perfectionistic self-presentation, and perfectionisticcognitions were associated negatively with mattering and theywere associated positively with depressive symptomatology. Therewas a moderate negative correlation between mattering anddepression.

3.2. Statistical analysis

Using the statistical analysis package, AMOS, version 17(Arbuckle, 2008), structural equation modeling (SEM) with maxi-mum likelihood estimation was conducted to test our hypothe-sized model in which mattering mediates the associationbetween perfectionism and depression. The effects of respondents’sex were accounted for in the model by allowing a correlationbetween respondents’ sex and perfectionism and by allowing pathsfrom respondents’ sex to mattering and to depression. Raw dataserved as the input, and the latent variable for perfectionism wasscaled by fixing one of its indicator paths to 1.0.

Several indices were used to evaluate the fit of our model: thecomparative fit index (CFI; Bentler, 1990), the root mean-squareerror of approximation (RMSEA; Browne & Cudeck, 1993) and thestandardized root mean squared residual (SRMR). The CFI variesbetween 0 and 1, where values of 0.95 or greater indicate adequatefit (Tabachnick & Fidell, 2001). The RMSEA is bounded by 0 and at-tains that value when a model exactly reproduces a set of observeddata. A value of 0.05–0.06 signifies close fit, a value of 0.08 signifiesmarginal fit, and values greater than 0.08 signify poor fit (Browne &Cudeck, 1993). The SRMR represents the average differencesbetween the sample variances and covariances and the estimatedpopulation variances and covariances. Values of the SRMR rangefrom 0 to 1.0 and values of 0.08 or less denote a well-fitting model(Tabachnick & Fidell, 2001).

Indirect effects were tested using the biased-corrected boot-strap method. This method has been found to offer a more accuratebalance between Type 1 and Type 2 errors compared to othermethods used to test indirect effects (MacKinnon, Lockwood, &Williams, 2004). As recommended, 1000 bootstrap samples andthe 95% bias-corrected confidence intervals (CIs) were utilized toassess the significance of indirect effects (Shrout & Bolger, 2002).When necessary, theory and modification indices guided modelchanges to achieve adequate fit, with v2 analysis utilized to deter-mine significant improvements at each step.

3.2.1. Treatment of missing data and outliersOne participant did not indicate his or her sex and was thus re-

moved from further analysis. Overall, 86% of the participants hadcomplete data with participants missing information on a maxi-mum of two variables (i.e., 1.6% of participants) and informationon less than one variable on average (SD = 0.40). Results of inde-pendent sample t-tests indicated that students with complete datadid not differ from those with incomplete data on any of the modelvariables. All missing values were imputed using the expectationmaximum (EM) algorithm in SPSS based on all available data foreach individual, as empirical research has clearly demonstratedthat this method is preferential to more conventional methodsincluding listwise deletion, pairwise deletion, and mean substitu-tion (see Schafer & Graham, 2002).

3.2.2. Structural equation modelingOur model hypothesized that mattering would mediate the

association between perfectionism and depression. The customaryway of establishing mediation is to first verify a direct associationbetween the independent variable (perfectionism) and thedependent variable (depression). Second, the fully mediated modelmust be shown to provide an adequate fit to the data with all pathsstatistically significant and in the correct direction. Finally, thefully mediated model should be compared with a model that also

G.L. Flett et al. / Personality and Individual Differences 52 (2012) 828–832 831

includes the direct effect, with partial mediation being demon-strated if the direct effect is significantly lowered, but significantlydifferent from zero, and full mediation being demonstrated if thedirect effect is nullified (Baron & Kenny, 1986).

3.2.3. Model 1: perfectionism and depressionThe first model tested the direct link between perfectionism

and depression. Given the strong associations among socially pre-scribed perfectionism, perfectionistic self-presentation, and perfec-tionistic cognitions, perfectionism was modeled as a latent factorusing each of the aforementioned measures as a separate indicator.The direct effects model fit the data very well (v2

ð4Þ ¼ 2:90, p = 0.57;CFI = 1.00; RMSEA = 0.00, pclose = 0.80; SRMR = 0.01). The standard-ized path coefficient between perfectionism and depression wassignificant and positive (b = 0.59, CI = 0.49–0.67) while the esti-mate for the path between respondents’ sex and depression wassignificant and negative (b = �0.20, CI = �0.30 to �0.09). Thus, wo-men and individuals with higher levels of perfectionism reportedhigher levels of depression. The association between respondents’sex and perfectionism was not statistically significant. The latentfactor loadings were all statistically significant (p < 0.001) and allof appropriate magnitude (b P 0.75), with the direct effects modelaccounting for 36% of the variability in depression.

3.2.4. Model 2: mediated model of perfectionism, mattering, anddepression

According to Baron and Kenny (1986), a necessary criterion forestablishing a mediational model is that the association betweenthe independent variable (perfectionism) and the dependent vari-able (depression) should be substantially reduced when the influ-ence of the mediating variable (mattering) is included. To test thisrequirement, a mediated model, which included a path from per-fectionism to mattering and a path from mattering to depression,was estimated. Respondents’ sex was statistically controlled forin the model. Results indicated that this model was a poor fit tothe data (v2

ð7Þ ¼ 71:06, p < 0.001; CFI = 0.84; RMSEA = 0.20, pclose <0.001; SRMR = 0.13). Inspection of the standardized residualcovariance matrix revealed that the standardized residuals be-tween each indicator of perfectionism and depression exceededan absolute value of 2, suggesting that an added direct path be-tween the latent factor for perfectionism and depression wouldimprove model fit. Inclusion of this path greatly improved modelfit (v2

differenceð1Þ ¼ 56:85, p < 0.05) and resulted in a well-fitting mod-el (v2

ð6Þ ¼ 14:21, p = 0.03; CFI = 0.98; RMSEA = 0.08, pclose = 0.18;SRMR = 0.03).

As shown in the final model displayed in Fig. 2, perfectionismwas associated negatively with mattering, which, in turn, wasassociated negatively with depression. The implied indirect effectof perfectionism to depression via mattering was tested using

SPP PSPS PCI

Depression Mattering

Respondents' Sex

.75 .79 .75

ns

-.30-.35

.48

ns -.23

Perfectionism

Fig. 2. Final mediated model of perfectionism, mattering, and depression. Errorterms not included in model for ease of presentation. Abbreviations used: SPP,socially prescribed perfectionism; PSPS, perfectionistic self-presentation; PCI,perfectionism cognitions inventory.

1000 bootstrap samples and the 95% bias-corrected confidenceintervals (CIs). Results indicated that the indirect effect of perfec-tionism to depression through mattering was statistically signifi-cant (b = 0.10; 95% CI = 0.06–0.17). However, a significant directassociation between perfectionism and depression was also ob-served (b = 0.48; 95% CI = 0.38–0.59); thus, mattering only partiallymediated the relationship between perfectionism and depression.Specifically, the indirect effect of perfectionism to depressionthrough mattering accounted for 17% of the variance in the link be-tween perfectionism and depression. The final model depicted inFig. 2 accounted for 14% of the variance in mattering and 44% ofthe variance in depression.1

4. Discussion

The current study examined perfectionism and mattering inuniversity students. Perfectionism was examined as a multidimen-sional construct with both personal and interpersonal componentsgiven our interest in focusing on mattering as feelings ofinsignificance to the lives of others. Correlational analysis providedsupport for the relevance of these interpersonal perfectionismdimensions. Lower mattering was associated with socially pre-scribed perfectionism and all three perfectionistic self-presentationfacets. These data are in keeping with the perfectionism socialdisconnection model and the general notion that interpersonalperfectionism involves a sense of detachment and a lack ofmeaningful interpersonal relationships. Hewitt and Flett (1991)suggested originally that extreme socially prescribed perfectionismcould reflect feelings of helplessness and hopelessness and theassociation with hopelessness has been supported in subsequentresearch (see Hewitt et al., 2006). The current results point to thepossibility that extreme socially prescribed perfectionism may in-clude a bleak sense of not ever mattering to other people in thefuture because these other people can never be pleased. Similarly,the obtained associations between perfectionistic self-presentationand reduced mattering further highlight the fragility of the self thatunderscores the need to either narcissistically engage in perfection-istic self-promotion or the need to defensively avoid revealing ordisplaying mistakes in public. To some extent, trying to seem per-fect in public may reflect an attempt to get attention and garneran enhanced sense of mattering among perfectionists who feelinterpersonally insignificant.

As expected, mattering was also associated negatively withdepression. The finding that mattering is linked with depressionand with certain perfectionism dimensions enabled us to test theproposed mediational model that ties together interpersonalperfectionism, low mattering, and depression. We found that mat-tering did indeed mediate the link between interpersonal perfec-tionism and depression but this association was only partiallymediated; there was still a direct link between interpersonalperfectionism and depression after taking mattering into consider-ation. Evidence that perfectionism still predicted depression can beregarded as evidence of the value of distinguishing at the empiricaland conceptual levels between perfectionism and self-evaluativeresponses since an association between perfectionism and depres-sion was still detected once mattering was taken into account. Infuture research, it will be illuminating to test self-esteem as amediator along with mattering to establish their relativeimportance as mediators of the link between perfectionism andpsychological distress. Similarly, examining mattering along withother constructs reflecting a sense of social disconnection (e.g.,loneliness, low social support) is indicated in light of the growing

1 It should be noted that the model results held when the model was testedwithout imputed data.

832 G.L. Flett et al. / Personality and Individual Differences 52 (2012) 828–832

emphasis on the role of social disconnection in perfectionism anddistress.

Certain limitations of the current study should be noted. First, thisstudywascross-sectionalinnature,sonocausalassumptionsarewar-ranted. Future research must explore the associations among perfec-tionism, mattering, and depression in longitudinal research designsandotherpossiblemodelsshouldbetestedtoconfirmthetemporalor-der of the variables in our model. For instance, it may be the case thatmattering influences distress, which, in turn, influences subsequentperceptions of mattering. Another plausible sequence that deservesconsiderationinlongitudinalresearchisthatperfectionismmediatesthelinkbetweenmatteringanddistress. Indeed,wetestedthismodelwithourcross-sectionaldataandalsofoundevidencethatperfection-ismcouldact asa partialmediatorof theassociationbetweenmatter-inganddepression.However,themainsequencewefocusedonismorein keeping withpast researchand conceptualizations suggesting thatnegative conceptualizations of the self mediate the perfectionism-depression link (see Flett et al., 2003; Preusser et al., 1994; Rice et al.,1998).Implicitinthesemodelsisthenotionthatfeelingsofself-esteemand mattering, relative to perfectionism, may be more state-like andmore reactive to daily interpersonal feedback and life setbacksthan perfectionism that was established early in life (also seeDunkley, Berg, & Zuroff, 2011).

Second, this study was based solely on self-report measures.Future research should re-examine these issues with informantratings. Third, we limited our focus to mattering as a mediatorand there are several other related factors (e.g., the frequency ofnegative social interactions, social support, etc.) that also deserveconsideration in future research. As suggested earlier, given thatmattering and self-esteem are related, future research shouldinvolve comparisons of these related constructs, with self-esteemwith both general measures and in specific forms (e.g., self-worthcontingent on approval and acceptance). Here it is important tonote, however, that mattering and self-esteem should still besomewhat distinguishable, both at the empirical and conceptuallevels. Indeed, in their classic paper, Rosenberg and McCullough(1981) showed that mattering to parents predicted a number ofimportant outcomes independent of self-esteem.

In summary, the current study confirmed that interpersonal as-pects of perfectionism (i.e., socially prescribed perfectionism andperfectionistic self-presentation) were associated with deficits inperceived mattering to others. In addition, mattering partiallymediated the link that these same perfectionism dimensions hadwith depressive symptoms. Presumably, an important goal forcounselors and therapists treating distressed perfectionists is to as-sess levels of interpersonal perfectionism and mattering and thenexplore and address the widespread feelings of insignificance thatare clearly evident among certain perfectionists. Feelings of notmattering could lead to a form of alienation and disconnection thatcould underscore the extreme acts of self-injury and suicideattempts that are far too common among troubled perfectionists.

References

Arbuckle, J. L. (2008). AMOS™ 17.0 user’s guide. Chicago, IL: SPSS.Bakan, D. (1966). The duality of existence. Isolation and communion in western man.

New York: Boston Beacon Press.Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in

social psychological research: Conceptual, strategic, and statisticalconsiderations. Personality and Social Psychology, 51, 1173–1182.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonalattachment as a fundamental human motivation. Psychological Bulletin, 117,497–529.

Bentler, P. M. (1990). Comparative fit indexes in structural equation models.Psychological Bulletin, 107, 238–246.

Blatt, S. J., & Zuroff, D. C. (2002). Perfectionism in the therapeutic process. In G. L.Flett & P. L. Hewitt (Eds.), Perfectionism: Theory, research, and treatment(pp. 393–406). Washington, DC: American Psychological Association.

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In I. K.A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–152).Newbury Park, CA: Sage.

Dunkley, D. M., Berg, J.-L., & Zuroff, D. C. (2011). The role of perfectionism in dailyself-esteem, attachment, and negative affect. Journal of Personality.

Elliott, G. C., Kao, S., & Grant, A. (2004). Mattering: Empirical validation of a social-psychological concept. Self and Identity, 3, 339–354.

Flett, G. L., Besser, A., Davis, R. A., & Hewitt, P. L. (2003). Dimensions ofperfectionism, unconditional self-acceptance, and depression. Journal ofRational-Emotive and Cognitive-Behavior Therapy, 21, 119–138.

Flett, G. L., Hewitt, P. L., Blankstein, K. R., & Gray, L. (1998). Psychological distressand the frequency of perfectionistic thinking. Personality and Social Psychology,75, 1363–1381.

Flett, G. L., Hewitt, P. L., Whelan, T., & Martin, T. R. (2007). The perfectionismcognitions inventory: Psychometric properties and associations with distressand deficits in cognitive self-management. Journal of Rational-Emotive andCognitive-Behavior Therapy, 25, 255–277.

Habke, A. M., & Flynn, C. M. (2002). Interpersonal aspects of trait perfectionism. InG. L. Flett & P. L. Hewitt (Eds.), Perfectionism: Theory, research, and treatment(pp. 151–180). Washington, DC: American Psychological Association.

Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts:conceptualization, assessment, and association with psychopathology.Personality and Social Psychology, 60, 456–470.

Hewitt, P. L., & Flett, G. L. (2004). The multidimensional perfectionism scale: Technicalmanual. Toronto: Multi-Health Systems Inc.

Hewitt, P. L., Flett, G. L., Sherry, S., & Caelian, C. (2006). Trait perfectionismdimensions and suicidal behavior. In T. Ellis (Ed.), Cognition and suicide: Theory,research, and therapy (pp. 215–235). Washington, DC: American PsychologicalAssociation.

Hewitt, P. L., Flett, G. L., Sherry, S., Habke, M., Parkin, M., Lam, R. W., et al. (2003). Theinterpersonal expression of perfectionism: Perfectionistic self-presentation andpsychological distress. Personality and Social Psychology, 84, 1303–1325.

Hewitt, P. L., Flett, G. L., Turnbull-Donovan, W., & Mikail, S. F. (1991). Themultidimensional perfectionism scale: Reliability, validity, and psychometricproperties in psychiatric samples. A Journal of Consulting and Clinical Psychology,3, 464–468.

Kanfer, F. H., & Hagerman, S. (1981). The role of self-regulation. In L. P. Rehm (Ed.),Behavior therapy for depression (pp. 143–179). New York: Academic Press.

Kawamura, K. Y., Hunt, S. L., Frost, R. O., & DiBartolo, P. M. (2001). Perfectionism,anxiety, and depression: Are the relationships independent? Cognitive Therapyand Research, 25, 291–301.

MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for theindirect effect: Distribution of the product and resampling methods.Multivariate Behavioral Research, 39, 99–128.

Marshall, S. K. (2001). Do I matter? Construct validation of adolescents’ perceivedmattering to parents and friends. Journal of Adolescence, 24, 473–490.

Preusser, K. J., Rice, K. G., & Ashby, J. S. (1994). The role of self-esteem in mediatingthe perfectionism-depression connection. Journal of College StudentDevelopment, 35, 88–93.

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research inthe general population. Applied Psychological Measurement, 1, 385–401.

Rice, K. G., Ashby, J. S., & Slaney, R. B. (1998). Self-esteem as a mediator betweenperfectionism and depression: A structural equation analysis. Journal ofCounseling Psychology, 45, 303–314.

Rosenberg, M., & McCullough, B. C. (1981). Mattering: Inferred significance andmental health. Research in Community and Mental Health, 2, 163–182.

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art.Psychological Methods, 7, 147–177.

Schieman, S., & Taylor, J. (2001). Statuses, roles, and the sense of mattering.Sociological Perspectives, 44, 469–484.

Sherry, S., Law, A., Hewitt, P. L., Flett, G. L., & Besser, A. (2008). Social support as amediator of the relationship between perfectionism and depression: Apreliminary test of the social disconnection model. Personality and IndividualDifferences, 45, 339–344.

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimentalstudies: New procedures and recommendations. Psychological Methods, 7,422–455.

Sorotzkin, B. (1985). The quest for perfection: Avoiding guilt or avoiding shame?Psychotherapy, 22, 564–571.

Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed). Toronto:Allyn & Bacon.

Taylor, R., & Turner, J. (2001). A longitudinal study of the role and significance ofmattering to others for depressive symptoms. Journal of Health and SocialBehavior, 42, 310–325.