Literature DB >> 31929572

The shield of self-compassion: A buffer against disordered eating risk from physical appearance perfectionism.

Luisa Bergunde1, Barbara Dritschel1.   

Abstract

General perfectionistic tendencies as well as perfectionism focussed specifically on one's physical appearance have been implicated as risk factors for disordered eating. This study extends previous research on protective factors by investigating self-compassion as a moderator of the relationship between both general and physical-appearance-perfectionism and disordered eating. A cross-sectional online survey assessed general perfectionism, physical-appearance-perfectionism, disordered eating symptoms, self-compassion and negative affect in female UK university students (N = 421). Results showed physical-appearance-perfectionism explained variance (15%) in disordered eating symptoms above general perfectionism and negative affect. Both perfectionistic concerns about and strivings for appearance perfection were significant unique predictors of disordered eating. Self-compassion moderated the relationship between both perfectionistic concerns and strivings of physical-appearance-perfectionism, but not general perfectionism, and disordered eating. This study suggests both perfectionistic concerns about and strivings for appearance perfection represent potential risk factors for disordered eating among female university students and that self-compassion may reduce their impact.

Entities:  

Year:  2020        PMID： 31929572      PMCID： PMC6957174          DOI： 10.1371/journal.pone.0227564

Source DB:  PubMed          Journal:  PLoS One        ISSN： 1932-6203            Impact factor:   3.240


Introduction

Disordered eating symptoms, such as dieting, binge eating and unhealthy weight control practices, are highly prevalent in the general population with 6.4% [1] and university students with 20.4% [2]. Engaging in disordered eating increases risk for problematic outcomes, such as full-blown eating disorders [3], obesity [4] and psychological distress [5]. Evidence suggests perfectionism is an important contributor to the development and maintenance of disordered eating symptoms [6], with perfectionism regarding one’s physical appearance potentially being particularly salient [7].

General perfectionism as a risk factor for disordered eating

Perfectionism is a general personality disposition characterized by striving for perfection and setting high personal standards as well as by being overly critical and fearing negative evaluations [8]. It is considered a multidimensional construct [9]. One of the most prominent models of multidimensional perfectionism emphasises both intrapersonal and interpersonal components of perfectionism [10]. This model suggests high standards of perfectionists may be experienced as self-imposed (self-oriented perfectionism) or as imposed by others (socially-prescribed perfectionism). Both self-oriented perfectionism, which involves having disproportionately high personal standards and a motivation to achieve them, and socially-prescribed perfectionism, characterised by beliefs that others hold excessively high standards for oneself, have been linked to disordered eating symptoms, namely dieting and bulimic symptoms respectively, in a non-clinical sample [11]. Other research has demonstrated two broad dimensions of perfectionism, namely perfectionistic concerns and perfectionistic strivings [12], which map onto these two components. Perfectionistic concerns captures worries associated with perfectionism, includes socially-prescribed-perfectionism as well as other perfectionism components, such as concern over mistakes, and has been associated with psychopathology. In contrast, perfectionistic strivings captures high standards and striving, includes self-oriented-perfectionism as well as other perfectionism components, such as personal standards, and has been linked to wellbeing [10, 13]. However, perfectionistic strivings can also become unhealthy and result in clinically-relevant perfectionism when perfectionistic strivings are over-valued and all-consuming [14]. With regards to eating pathology, research has found evidence suggesting both perfectionistic concerns and perfectionistic strivings to be uniquely related to increased eating pathology in non-clinical and clinical populations both cross-sectionally and longitudinally (see [15] for a review). Recent experimental studies further support both perfectionistic concerns and strivings perfectionism as potential causal risk factors of disordered eating symptoms [16, 17]. Findings showed that participants who received an experimental induction of perfectionistic concerns or perfectionistic strivings reported higher disordered eating symptoms 24 hours after the manipulation than those who received no induction [17]. It may be that the induction of perfectionistic concerns and perfectionistic strivings lead to the use of maladaptive eating as an emotion regulation coping strategy for negative emotions resulting from the induction. Individuals high in trait general perfectionism frequently experience feelings of incompetence and failure due to a tendency to set, pursue and fail to achieve unattainable goals and perceived expectations [18]. Furthermore, perfectionists frequently employ emotion regulation strategies such as rumination and avoidance in response to failure, which are considered maladaptive in this context [19]. These responses may increase the likelihood of engaging in compensatory behaviours [20]. This could involve losing control by for instance binge eating to avoid negative feelings [21] or an attempt to restore control of one’s goals by restricting food intake [20]. This is supported by Rivière and Douilliez’ [22] finding that use of these emotion regulation strategies partially explains the relationship between general perfectionism and disordered eating symptoms in female participants.

Physical-appearance-perfectionism and its relation to eating pathology

Research suggests perfectionists have particularly high standards in highly valued life domains [23]. One such domain may be physical appearance. Qualitative interviews with eating disorder patients showed that in this group, perfectionistic tendencies were often directed towards one’s physical appearance [24]. In addition, positive correlations have been found between a measure assessing physical-appearance perfectionism, the physical appearance perfectionism scale (PAPS), and self-reported weight control behaviours [25]. Furthermore, a cross-sectional survey study with university students demonstrated that physical-appearance-perfectionism explained variance in disordered eating symptoms above general perfectionism [7]. These preliminary findings suggest physical-appearance-perfectionism may represent a salient risk factor for disordered eating symptoms. However, research has not directly tested whether physical-appearance-perfectionism encompasses all the risk for disordered eating symptoms associated with perfectionism. Since other domain-specific perfectionism measures have been found to do so regarding domain-related outcomes [26], it is pertinent to evaluate if general perfectionism still explains variance in disordered eating symptoms above physical-appearance-perfectionism. Similar to general perfectionism, the PAPS assesses two components of physical appearance perfectionism, worries-about-appearance-imperfections and hope-for-appearance-perfection, intended to represent the general perfectionism subcomponents of perfectionistic concerns and perfectionistic strivings respectively [7]. Research has supported this conceptualization by demonstrating that worries-about-appearance-imperfection correlated more strongly with perfectionistic concerns than perfectionistic strivings of general perfectionism. However, contrary to expectations, hope-for-appearance-perfectionism has not shown a consistently stronger relationship to perfectionistic strivings than perfectionistic concerns [7]. Thus, validity of this subscale may be questionable. To hope for something signifies wanting something to happen, whereas to strive implies actively making great efforts to achieve something [27]. These definitions highlight that “hope” may not adequately capture the active component of perfectionistic strivings which is key to distinguishing perfectionistic tendencies from general wishes [8]. Moreover, hope-for-appearance-perfection has consistently been shown to be either unrelated to or a non-significant predictor of eating pathology [7, 25], despite perfectionistic strivings of general perfectionism conferring unique risk in eating pathology [11]. This highlights a potential pitfall of hope-for-appearance-perfection’s miswording, as researchers have concluded perfectionistic strivings regarding physical appearance is unrelated to negative outcomes, such as disordered eating symptoms [7]. To clarify the role of perfectionistic strivings regarding physical appearance in disordered eating symptoms, this study introduced a modified version of the hope-for-appearance-perfection subscale where the word “hope” was replaced with “strive” with the aim of better reflecting perfectionistic strivings in the domain of physical appearance.

Self-compassion as a protective factor against disordered eating symptoms

With regards to disordered eating, a recent review concluded self-compassion, a healthy way of relating to one’s self in distress [28], is negatively related to eating pathology [29]. In a seminal experimental study of female participants high in restrictive eating, Adams and Leary [30] demonstrated that after consumption of a doughnut, individuals who received a self-compassion induction experienced fewer negative emotions and less disinhibited eating afterwards than those who received no induction. Self-compassion encompasses three aspects: 1. responding to one’s suffering with self-kindness rather than self-judgment, 2. being mindfully aware of rather than overidentifying with one’s feelings and thoughts of suffering, and 3. seeing one’s suffering as part of a common human experience rather than feeling isolated [31]. In their review, Inwood and Ferrari [32] found self-compassion to be related to mental health partially through emotion regulation, suggesting self-compassion may act as a facilitator of emotion regulation. Additional studies have reported a moderate positive relationship between self-compassion and psychological flexibility, the ability to be consciously present and adapt one’s behaviour to the current situation [33, 34]. It may therefore be that self-compassion exerts its protective effect through increased psychological and emotional flexibility, such that context-appropriate responses and emotion regulation strategies are more easily accessible [35, 36]. Findings examining physiological measures corroborate this by demonstrating that highly self-compassionate individuals exhibit higher resting levels of vagally mediated heart rate variability, an indicator of flexible emotion regulation capacity [37]. This implies self-compassion may enable better adaptation of emotional responses to environmental demands, both physiologically and psychologically. With regards to disordered eating symptoms, self-compassion may be particularly important since emotion regulation difficulties have been found to be associated with disordered eating symptoms [38]. Based on the above, this study proposes a model for self-compassion as a protective factor against perfectionism’s effect on disordered eating symptoms (Fig 1). As outlined earlier, perfectionists may be prone to experiences of incompetence [18], which in turn may constitute a situation of suffering [28]. How individuals cope with feelings and thoughts related to their suffering may depend on individual differences in self-compassion. If low on self-compassion, one may be more likely to respond in a context of judgement, thereby potentially limiting the ability to respond with situation-appropriate coping strategies and consequently increasing the likelihood of employing compensatory behaviours, including disordered eating symptoms [21]. In contrast, if high on trait self-compassion, thoughts can be responded to in a context of self-compassion, such that one treats oneself with kindness and sees one’s suffering as part of common humanity. This may enable a more flexible and situation-appropriate choice of emotion regulation strategy, thereby reducing the impact of perfectionistic failure on disordered eating symptoms [32].
Fig 1

Theoretical model of the moderating role of self-compassion on the relationship between perfectionism and disordered eating.

XYZ refers to a domain in which an individual is highly perfectionistic.

Theoretical model of the moderating role of self-compassion on the relationship between perfectionism and disordered eating.

XYZ refers to a domain in which an individual is highly perfectionistic.

Present study

This study aimed to increase understanding of physical-appearance-perfectionism facets as risk factors of disordered eating symptoms and investigate trait self-compassion as a protective factor against the impact of perfectionism, both general and physical-appearance-perfectionism, on eating pathology. Given challenges in the conceptualization and valid assessment of disordered eating symptoms in males, this study limited recruitment to female university students [2, 39]. BMI and negative affect were included in our model as both have demonstrated significant associations with disordered eating symptoms and perfectionism’s unique relationship with disordered eating symptoms above these variables remains to be investigated [40]. General perfectionism was expected to explain variance in disordered eating symptoms beyond physical-appearance-perfectionism. We also hypothesised that physical-appearance-perfectionism would predict variance in disordered eating symptoms above perfectionistic concerns and perfectionistic strivings of general perfectionism and negative affect and that both the perfectionistic concerns and a modified perfectionistic strivings subscale of physical-appearance-perfectionism would predict unique variance in disordered eating symptoms. Finally, this study investigated whether the relationship between perfectionism facets and disordered eating symptoms would be moderated by the level of trait self-compassion. Negative affect was included as a covariate to control for the effect of differences in current mood on perfectionism and disordered eating symptoms [41]. The central hypothesis was that the relationship between perfectionism facets and disordered eating symptoms would be weaker the higher the level of trait self-compassion regardless of current negative affect.

Methods

Participants

From 691 people who opened the survey, 270 participants were excluded from further analysis. Inclusion criteria were: aged 18 years or over, female and a university student. To ensure high data quality [42], exclusion criteria included: completion of survey in less than 60% of projected time (≤9min) and omission of more than 20% of questionnaire questions. The final sample comprised 421 female university students (M = 20.95; SD = 3.30). Participants self-described their ethnicity as White (86.9%), Asian (6.0%), Mixed (4.0%), Black (1.4%) and Other (1.2%). Participants were recruited using opportunity sampling with the survey being advertised online for 12 weeks.

Measures

Body Mass Index (BMI) calculated from self-reported height and weight (kg/m2) was normal (18.5≥BMI≤24.99) in 73.5% of students, whereas 9.7% were underweight (BMI<18.5), 13.2% were overweight (25≥BMI≤29.99) and 3.6% were obese (BMI≥30). The 15-item short form of the 45-item Multidimensional Perfectionism Scale (MPS; [10]; short form: [43]) has shown good construct, convergent and discriminant validity [44]). It was used to measure the two components of general perfectionism shown to be related to disordered eating symptoms [15]: self-oriented-perfectionism (5 items) and socially-prescribed-perfectionism (5 items). Items were responded to on a 7-point Likert scale (1 = disagree to 7 = agree). The Physical Appearance Perfectionism Scale (PAPS; [25]) is a 12-item self-report questionnaire capturing worries-about-appearance-imperfections (7 items) and hope-for-appearance-perfection (5 items). This study used a modified version of the second subscale, replacing the word “hope” with “strive”. As assumptions of exploratory factor analysis (EFA) were met, an EFA using principal axis factoring with direct quartimin oblimin rotation was conducted on the 12 PAPS items. Two factors had eigenvalues above 1, were positively correlated (r = .41,p<.001) and together explained 73.2% of the variance. Factor 1 represented worries-about-appearance-imperfection and factor 2 represented striving-for-appearance-perfection. Items were assessed via a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). PAPS has demonstrated good 4-week test-retest reliability and good preliminary construct validity [25]. The 26-item self-report Eating Attitudes Test (EAT-26; [3]) measured disordered eating symptoms. The EAT-26 has been shown to be highly reliable and valid [45]. Three subscales are captured: Dieting (13 items), Bulimia and Food Preoccupation (6 items) and Oral Control (7 items). Participants reported the frequency with which behaviours outlined in items were experienced by responding on a 6-point Likert scale (1 = never to 6 = always). As per Garner et al. [3], 20% of participants were at risk of an eating disorder, reflected by a score of 20 or above. The 26-item Self-Compassion-Scale (SCS; [31]) is a self-report questionnaire which has demonstrated good construct, convergent, discriminant and test-retest validity [31]. It assesses how individuals treat themselves in difficult times and comprises six subscales measuring three facets of self-compassion. (1) Self-kindness versus self-judgement (10 items), (2) Common humanity versus isolation (8 items) and (3) Mindfulness versus over-identification (8 items). Items were rated on a 5-point Likert scale (1 = almost never to 5 = almost always). The 21-item Depression, Anxiety and Stress Scale (DASS-21; [46]) is a highly reliable and valid measure of the frequency and severity of negative affect over the past week [47]. Although the DASS-21 was conceptualised to respectively assess depression, anxiety and stress, recent research suggests a general negative affect factor explains the majority of the variance in the DASS-21 [48]. Hence this study employed the total DASS-21 score as a measure of negative affect. Items were rated on a 4-point Likert scale (0 = did not apply to me at all to 3 = applied to me very much of the time).

Procedure

Ethical approval was granted by St Andrews University Research Ethics Committee. The survey was completed online using the platform Qualtrics®. It included demographic questions (age, gender, ethnicity, height, weight, current university student), followed by the questionnaires presented in a randomized order. Participation was rewarded with the chance to win a £20 Amazon voucher.

Data screening and analysis

As missing data accounted for only 0.25% of all values, missing items were replaced with the item mean [49]. BMI and disordered eating symptoms were strongly positively skewed, likely due to the non-clinical sample. Further, self-oriented perfectionism showed strong negative skew. Inspection of residual scatterplots indicated heteroscedasticity. To adjust for these violations in assumptions, BMI was transformed logarithmically, and square-root-transformations were applied to self-oriented-perfectionism and EAT26 scores (see S1 Table). After applying these transformations, assumptions for regression analysis were better met. Analyses with transformed data showed no differences in significance of results with the exception of one finding. Self-compassion no longer moderated the relationship between socially-prescribed perfectionism and disordered eating symptoms when transformed data was used for analysis. Therefore, p-, t- and F-values for this relationship are reported from analysis using transformed data. All other results are reported from analyses using untransformed data. A cross-sectional correlational design was used. Data were analysed using SPSS 24 [50] and the PROCESS macros [51].

Results

Preliminary analysis

Table 1 shows descriptive statistics and bivariate correlations. Internal reliability was good to excellent for all variables. Self-compassion correlated negatively with all perfectionism components, disordered eating symptoms and negative affect. Negative affect and all perfectionism components were positively correlated with disordered eating symptoms. Striving-for-appearance-perfection correlated more strongly with self-oriented-perfectionism (r(421) = .30,p<.001), than socially-prescribed-perfectionism (r(421) = .22,p<.001). In contrast, worries-about-appearance-imperfection correlated more strongly with socially-prescribed-perfectionism (r(421) = .34,p<.001) than self-oriented-perfectionism (r(421) = .21,p<.001). As BMI was not significantly related to any other variable in this study, it was not included in subsequent analyses.
Table 1

Summary of descriptive statistics.

This includes mean (M), standard deviation (SD), Cronbach’s Alpha (α) and bivariate correlations of study variables (N = 421).

VariableMSD12345678α
1. Self-oriented perfectionism (MPS)26.86.31-.45*.30*.21*.28*-.31*-.08.26*.87
2. Socially-prescribed perfectionism (MPS)18.626.83-.22*.34*.29*-.41*-.04.39*.82
3. Striving for appearance perfection (PAPS)3.671.01-.46*.40*-.29*-.02.20*.91
4. Worries about appearance imperfection (PAPS)3.121.07-.57*-.60*.07.49*.93
5. Disordered eating symptoms (EAT-26)11.4511.74--.41*-.08.48*.91
6. Self-compassion (SCS)2.67.70-.04-.60*.93
7. Body Mass Index (BMI)22.243.20--.06-
8. Negative affect (DASS-21)20.4413.06-.93

* Bonferroni-corrected alpha levels were applied (p < .0017; 2-tailed). MPS (Multidimensional Perfectionism Scale); PAPS (Physical Appearance Perfectionism Scale), EAT-26 (26-item Eating Attitudes Test); SCS (Self-compassion Scale); BMI (Body Mass Index); DASS-21 (21-item Depression, Anxiety and Stress Scale)

Summary of descriptive statistics.

This includes mean (M), standard deviation (SD), Cronbach’s Alpha (α) and bivariate correlations of study variables (N = 421). * Bonferroni-corrected alpha levels were applied (p < .0017; 2-tailed). MPS (Multidimensional Perfectionism Scale); PAPS (Physical Appearance Perfectionism Scale), EAT-26 (26-item Eating Attitudes Test); SCS (Self-compassion Scale); BMI (Body Mass Index); DASS-21 (21-item Depression, Anxiety and Stress Scale)

Perfectionism and disordered eating symptoms

A two-stage hierarchical regression in which physical-appearance-perfectionism facets was entered in Step 1 and general perfectionism facets in Step 2 significantly predicted variance in disordered eating symptoms (R2 = .37, F(4, 416) = 60.97, p < .001; adjusted R2 = .36) (see Table 2). General perfectionism accounted for a small, albeit significant unique amount of variance in disordered eating symptoms ( = .02, ΔF(2, 416) = 6.57, p = .002) above that accounted for by physical-appearance-perfectionism ( = .35, ΔF(2, 418) = 112.39, p < .001).
Table 2

Hierarchical multiple regression predicting disordered eating symptoms, with 95% confidence intervals reported in parenthesis.

Confidence intervals, standard errors and significance levels based on 2000 bootstrap samples (N = 421).

VariableR2ΔR2BSE(B)βp
Step 1.34.34
    Constant-12.65(-16.36, -9.17)1.86< .001
    Worries about appearance imperfection5.44(4.38, 6.48).49.50< .001
    Striving for appearance perfection1.94(.87, 3.00).52.17.001
Step 2.36.02
    Constant-17.94(-22.92, -13.55)2.35< .001
    Worries about appearance imperfection5.17(4.12, 6.20).50.47< .001
    Striving for appearance perfection1.52(.44, 2.58).52.13.010
    Socially-prescribed perfectionism.08(-.08, .24).08.04.320
    Self-oriented perfectionism.23(.07, .39).08.12.007

Note: R2 = r_squared. ΔR2 = change in R2. B = unstandardized regression coefficient. SE(B) = standard error B. β = standardised regression coefficient.

Hierarchical multiple regression predicting disordered eating symptoms, with 95% confidence intervals reported in parenthesis.

Confidence intervals, standard errors and significance levels based on 2000 bootstrap samples (N = 421). Note: R2 = r_squared. ΔR2 = change in R2. B = unstandardized regression coefficient. SE(B) = standard error B. β = standardised regression coefficient. A three-stage hierarchical regression showed that negative affect, general perfectionism and physical-appearance-perfectionism significantly predicted variance in disordered eating symptoms (R2 = .41, F(5, 415) = 57.93, p < .001; adjusted R2 = .40) (see Table 3). Negative affect (Step 1) accounted for 23% of the variance in disordered eating symptoms ( = .23, ΔF(1, 419) = 123.35, p < .001). Adding general perfectionism (Step 2) explained an additional 3% of variance in disordered eating symptoms ( = .03, ΔF(2, 417) = 8.43, p < .001). Further, adding physical-appearance-perfectionism (Step 3) accounted for an additional 15% of variance in eating pathology, ( = .15, ΔF(2, 415) = 54.12, p < .001), with both worries-about-appearance-imperfection (β = .36, p < .001) and striving-for-appearance-perfection (β = .15, p = .001) predicting unique variance.
Table 3

Hierarchical multiple regression predicting disordered eating symptoms, with 95% confidence intervals reported in parenthesis.

Confidence intervals, standard errors and significance levels based on 2000 bootstrap samples (N = 421).

VariableR2ΔR2BSE(B)βp
Step 1.23.23
    Constant2.68(1.01, .52).94.004
    DASS21.43(.33, .52).04.48< .001
Step 2.26.03
    Constant-5.54(-9.46, -1.66)2.21.013
    DASS21.37(.28, .46).04.42< .001
    Socially prescribed perfectionism.10(-.07, .28).09.06.229
    Self-oriented perfectionism.28(.12, .45).09.15.002
Step 3.41.15
    Constant-16.77(-21.57, -12.52)2.29< .001
    DASS21.22(.13, .31).04.25< .001
    Socially prescribed perfectionism-.01(-.18, .14).08-.01.855
    Self-oriented perfectionism.19(.04, .35).08.10.020
    Worries about appearance imperfection3.99(3.01, 4.95).53.36< .001
    Striving for appearance perfection1.76(.74, 2.87).51.15.001

Note: R2 = r_squared. ΔR2 = change in R2. B = unstandardized regression coefficient. SE(B) = standard error B. β = standardised regression coefficient.

Confidence intervals, standard errors and significance levels based on 2000 bootstrap samples (N = 421). Note: R2 = r_squared. ΔR2 = change in R2. B = unstandardized regression coefficient. SE(B) = standard error B. β = standardised regression coefficient.

Moderation analyses

Moderations were analysed using PROCESS with 2000 bootstrap samples and heteroscedasticity-consistent standard errors. Disordered eating symptoms were the outcome variable. In each model, negative affect was entered as a covariate, one of the four perfectionism components was entered as the independent variable and self-compassion was entered as the moderating variable. Including the interaction of self-compassion and self-oriented-perfectionism, ΔR2 = .005,ΔF(1,416) = 2.650,p = .104, and self-compassion and socially-prescribed-perfectionism, ΔR2 = .002,ΔF(1,416) = 1.323,p = .251, did not significantly increase variance accounted for in disordered eating symptoms. This suggests self-compassion did not moderate the relationship between general perfectionism facets and disordered eating symptoms. However, the interaction between self-compassion and worries-about-appearance-imperfections significantly increased variance accounted for in disordered eating symptoms, ΔR2 = .03,ΔF(1,416) = 21.51,p<.001. This indicates self-compassion moderated the relationship between worries-about-appearance-imperfections and disordered eating symptoms. Simple slopes analysis (Fig 2A) demonstrated that worries-about-appearance-imperfection was significantly related to eating pathology at one standard deviation below mean self-compassion score (B = 7.06,t = 10.10,p<.001,95%CI:5.69,8.43) and at mean self-compassion score (B = 5.16,t = 10.30,p<.001,95%CI:4.18,6.15). At one standard deviation above mean self-compassion score the relationship was weaker (B = 3.27,t = 5.53,p<.001,95%CI:2.11,4.43), however it first became non-significant at 1.86 standard deviations above the mean self-compassion score.
Fig 2

Regression lines showing the relationship between disordered eating and worries-about- appearance-imperfection (A) and striving-for-appearance-perfection (B) as a function of self-compassion level after controlling for negative affect. High and low values correspond to ±1SD from the mean.

Regression lines showing the relationship between disordered eating and worries-about- appearance-imperfection (A) and striving-for-appearance-perfection (B) as a function of self-compassion level after controlling for negative affect. High and low values correspond to ±1SD from the mean. Inclusion of the interaction of self-compassion and striving-for-appearance-perfection also resulted in a significant increase of variance accounted for in eating pathology, ΔR2 = .02,ΔF(1,416) = 7.38,p = .007, indicating self-compassion moderated the relationship between striving-for-appearance-perfection and disordered eating symptoms. Simple slopes analysis (Fig 2B) showed that the relationship between striving-for-appearance-perfection and disordered eating symptoms was significant at one standard deviation below the mean self-compassion score (B = 5.01,t = 5.42,p<.001,95%CI:3.19,6.82) and at the mean self-compassion score (B = 3.49,t = 6.65,p<.001,95%CI:2.46,4.52). Whilst attenuated at one standard deviation above the mean self-compassion score (B = 1.98,t = 3.49,p<.001,95%CI:.86,3.09), the relationship first became non-significant at 1.38 standard deviations above the mean self-compassion score.

Discussion

This study aimed to increase understanding of physical-appearance-perfectionism as a risk factor of disordered eating symptoms and investigate whether self-compassion buffers the relationship between perfectionism, both general and physical-appearance-perfectionism, and eating pathology in female students. Overall, means and standard deviations of study variables as well as direction and strength of correlations reflected those reported in prior studies using university students [7]. The only unexpected finding was that BMI was not significantly positively correlated with disordered eating symptoms, as documented in previous studies [52]. This discrepancy may be due to measurement error as BMI findings are not consistent with population statistics in this age group [53]. Based on previous findings on participants showing an underestimation bias in self-reported BMI, it may be that participants in our study failed to accurately report their current height and weight [54].

Perfectionism is linked to eating pathology

Results showed improved differential construct validity for the modified PAPS. Striving-for-appearance-perfection correlated more strongly with self-oriented perfectionism than socially-prescribed perfectionism and vice versa for worries-about-appearance-imperfections. This finding contrasts with previous studies which have not found support for this pattern of correlations with the original PAPS, particularly regarding the hope-for-appearance-perfection subscale [7]. However, findings are consistent with research on other multi-dimensional measures of perfectionism showing similar correlation patterns [55, 26]. Consequently, results suggest the modified striving-for-appearance-perfection subscale may better reflect the active striving of perfectionistic strivings than hope-for-appearance-perfection, thereby aligning the PAPS more closely conceptually with the two-dimensional model of perfectionism [13]. The predictive power of physical-appearance-perfectionism was compared to that of general perfectionism in explaining disordered eating symptoms. As expected, results showed that although physical-appearance-perfectionism explained most of the variance in disordered eating symptoms, general perfectionism explained additional unique variance. On the one hand, this contradicts prior research on other domain-specific forms of perfectionism, which fully encompassed the variance explained by general perfectionism in domain-related outcomes [26]. On the other hand, findings are consistent with research showing that being perfectionistic regarding interpersonal relations also increases risk for disordered eating symptoms [56]. Thus, findings indicate being perfectionistic in life domains other than one’s physical appearance may also confer unique risk for disordered eating symptoms. Further regression analyses showed that physical-appearance-perfectionism explained variance in eating pathology above that explained by general perfectionism and negative affect. This replicates prior research and extends it by including negative affect [7]. Moreover, findings contribute to a growing literature demonstrating the added value of domain-specific measures of perfectionism in explaining domain-related outcomes [26, 57]. Findings imply that being perfectionistic regarding one’s physical appearance has added explanatory power regarding individual differences in disordered eating symptoms. Importantly, as expected, both perfectionistic concerns and perfectionistic strivings of physical-appearance-perfectionism emerged as significant predictors of disordered eating symptoms. This contrasts with studies using the original PAPS, which found only worries-about-appearance-imperfection, and not hope-for-appearance-perfection, to uniquely predict eating pathology [7]. Thus, findings suggest actively striving for, as opposed to hoping for, appearance perfection confers unique risk for disordered eating symptoms. This also fits well with the concept of clinical perfectionism [14] as pursuing high standards in the domain of one’s physical appearance may constitute an important risk factor for pathological outcomes related to eating. These findings highlight two core points. First, hoping for a perfect appearance may not adequately capture the active striving of perfectionistic strivings perfectionism and future research using the PAPS should deliberate this discrepancy. Second, it is important to consider both perfectionistic strivings and perfectionistic concern regarding physical appearance as potential risk factors for eating pathology in female university students.

Self-compassion as a moderator

Contrary to hypotheses, strength of the relationship of general perfectionism facets with disordered eating symptoms did not change significantly with self-compassion level, although the non-significant trend was in the direction we predicted, with the perfectionism-disordered eating symptoms relationship being weaker for those high in trait self-compassion. However, consistent with predictions, self-compassion did significantly moderate the relationship between both perfectionistic concerns (worries-about-appearance-imperfections) and perfectionistic strivings (striving-for-appearance-perfection) of physical-appearance-perfectionism and disordered eating symptoms. Generally, these relationships were weaker for women high compared to low in self-compassion. Interestingly, there was a crossover interaction (see Fig 2A). When worries-about-appearance-imperfections were low those high in self-compassion showed the highest disordered eating. This result may be explained by the effect of a confounding variable not controlled for in this study. For instance, it could be that those individuals reporting low worries-about-appearance-imperfections and high self-compassion were exposed to more weight stigma growing up than the individuals reporting low worries-about-appearance-imperfections and low self-compassion [58]. Overall results indicate that in young women, having a tendency to treat oneself compassionately in times of distress may buffer against the increased disordered eating symptoms associated with higher levels of being perfectionistic regarding one’s physical appearance. Findings corroborate prior research demonstrating negative associations between self-compassion and eating pathology [29]. Moreover, results complement those of Adams and Leary [30] who demonstrated that being more self-compassionate mitigated engagement in disordered eating behaviour. In particular, this study adds to research demonstrating self-compassion as a protective factor against external (media thinness-related pressures [59]), and physical (increased BMI [60]) risk factors of disordered eating symptoms by demonstrating self-compassion may also protect against personality trait risk factors, namely physical-appearance-perfectionism. The finding that self-compassion only moderated the relation between disordered eating symptoms and physical-appearance-perfectionism, but not general perfectionism, suggests self-compassion may be particularly helpful when dealing with suffering related to one’s physical appearance. For instance, self-compassion has shown positive associations with several variables indicative of physical appearance acceptance, such as body image flexibility [60]. Moreover, self-compassion training has been shown to effectively increase body appreciation [61]. It appears that responding kindly and mindfully to worries about or failures in achieving physical-appearance-perfection and viewing associated suffering as common to humanity may facilitate adaptive coping in the eating domain. Consequently, these findings refine our understanding of the relationship between disordered eating, perfectionism and self-compassion as illustrated in the theoretical model in Fig 1. Results suggest self-compassion is a moderator of the relationship between physical-appearance-perfectionism, as opposed to general perfectionism, and disordered eating symptoms. Future research should investigate moderators of the general perfectionism-disordered eating symptoms relationship. Results indicate that perfectionists who experience frequent feelings of inadequacy regarding physical appearance perfection, but who are able to respond to these in a context of self-compassion, rather than self-judgement, may be able to cope using situation-appropriate emotion regulation strategies, thereby reducing the likelihood of engaging in disordered eating. Consequently, findings indirectly support prior research suggesting self-compassion as a facilitator of appropriate emotion regulation [32], potentially due to its link with psychological flexibility [34]. However, the assumption that a self-compassionate response facilitates more flexible and appropriate emotion regulation remains to be tested directly, for example using a moderated mediation model. Finally, results provide preliminary evidence that disordered eating prevention programs may benefit from encouraging self-compassion amongst female university students exhibiting high levels of physical-appearance-perfectionism. Importantly, self-compassion has been shown to be both a trait amenable to change and a skill trainable through daily practice and meditation [62], thereby highlighting its utility as an intervention target.

Limitations and future research

The cross-sectional correlational design impeded ability to determine causality and directionality. Although we suggest that perfectionism leads to disordered eating and this effect may be buffered by self-compassion, this cannot be conclusively determined by our study. Therefore, future research should employ longitudinal as well as experimental designs to more directly test causality of relationships proposed here and solidify self-compassion as a protective factor. Furthermore, use of self-report questionnaires required self-awareness from participants and entailed increased risk for response bias [63], thereby potentially affecting validity of results. Moreover, since opportunity sampling was employed, the possibility that the sample is not representative of the population exists. It could for example be that the individuals who self-selected to participate shared a particular preoccupation with eating. Also, generalisability of findings is restricted to female university students with a mainly white self-reported ethnicity. It would be interesting to study the proposed relationships in individuals from different cultures, as research suggests cultural variations in trait self-compassion. For instance, individuals from Thailand, where Buddhist teachings are central to the way of life, were found to demonstrate elevated levels of self-compassion compared to individuals from Taiwan and the USA [64]. Research may also benefit from comparing self-compassion’s moderating role of the perfectionism-disordered eating symptoms relationship between female and male participants as research suggests trait self-compassion may be higher amongst males [65]. It may also be that elevated trait self-compassion in males may explain why previous studies found no positive association between perfectionism and disordered eating symptoms amongst males [22]. Moreover, this study only examined general disordered eating symptoms, thereby precluding conclusions regarding specific eating disorder symptom groups. Future research should include individual measures of these symptoms to clarify potential differences between symptom groups, as reported elsewhere [11]. Lastly, although a significant moderator, effect sizes were small and self-compassion cannot fully explain existing variance in the relationship between physical-appearance-perfectionism facets and eating pathology. This suggests future research should investigate additional moderators of this relationship and disordered eating symptom prevention efforts should target self-compassion together with other factors.

Conclusion

This study highlighted the added utility of domain-specific perfectionism measures and suggests both concerns about and striving for a perfect appearance may confer unique risk for disordered eating symptoms in female university students. Moreover, this study provided preliminary evidence that self-compassion may protect against increased disordered eating symptoms associated with increased physical-appearance-perfectionism. Although further clarification of findings is needed, results have the potential to inform and enhance the development of disordered eating prevention. A focus on reducing physical-appearance-perfectionism and developing a more compassionate attitude towards oneself appear crucial.

Summary of descriptive statistics for untransformed and transformed values.

This includes mean (M), standard deviation (SD) and range (N = 421). (PDF) Click here for additional data file. 22 Oct 2019 PONE-D-19-24566 The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism PLOS ONE Dear Dr. Dritschel, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. We would appreciate receiving your revised manuscript by November 20, 2019 When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. We look forward to receiving your revised manuscript. Kind regards, Kristin Vickers, Ph.D. Academic Editor PLOS ONE Journal Requirements 1. When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide. Other comments Dear Dr. Dritschel, Thank you for submitting your research to PLOS ONE. I am happy to report that your manuscript entitled "The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism" has received a Minor Revision decision as described below. First, let me apologize the delay in getting feedback to you; a number of experts who were asked to review were unable to do so. Thank you for your patience. Below you will find helpful comments from two reviewers, along with my input. Required revisions are those that need to be tended to; recommended revisions are those where authors may choose to address the reviewer’s concern, or may choose to explain why they are not making the suggested change. With your revision, please include an explanation of how reviewer comments were addressed. Overall, I find this to be a well done study that contributes to the literature. It will be useful to many researchers. I look forward to your revision and would ask that the revision be received by November 20, 2019. Sincerely, Kristin Vickers Kristin Vickers, PhD Comments from Dr. Vickers Required Revisions 1)            It is extremely important that the data be available. Despite using different computers and browsers, I was not able to open the data using the link provided. Attempts to do so resulted in an error message saying: DOI Not Found  10.17630/39426414-eb9b-4b7f-b078-2f88f96cdd76  Reviewer 2 also noted this issue. 2)            Please report the reliability values for the measures as reflected in the current sample (as noted by Reviewer 2). 3)            Please present the DASS-21 as 3 subscales (as noted by Reviewer 2). 4)            Please describe the transformations applied and how the transformations were selected (as noted by both reviewers). 5)            Please clarify whether the authors are referring to the analysis performed by the PROCESS macro (for moderation) as regression. (Reviewer 2 also notes this issue, Results comment 5; it appears that the reviewer thought that the PROCESS macros was not used, given the way in which the analyses were described). Did the authors use the PROCESS macro? If so, please clarify the language. As Hays notes (http://processmacro.org/faq.html) the regression approach may yield the same results as PROCESS but differences are also possible.: “…When the same model is estimated using the same data with the same output options, the results will be the same as what you get with SPSS or SAS's regression procedures.  There are many sources of discrepancies you may notice when discrepancies exist...A common one is requesting heteroscedastity-consistent standard errors in PROCESS, which are different than standard OLS standard errors.  SPSS and SAS won't generate these standard errors, but PROCESS will (as will my RLM and HCREG macros) but only if you ask for them.  When you do, standard errors, t-values, p-values, and confidence intervals are different than what SPSS and SAS's internal regression procedures produce, as they should be.” 6)            Please disregard the comments of Reviewer 2 relating to 25% missing data. This was an error on the reviewer’s part, as only 0.25% of the data are missing. Thus, there is no need to address the comments of Reviewer 2 relating to this point, specifically number 3 and number 4 under Methods. Further, it appears the discrepancy between reviewers (Minor versus Major Revision) is a result of this error. With this in mind, and given my own read of this interesting manuscript, I am assigning a Minor Revision decision. 7)            All other reviewer comments are recommended revisions, and the authors are invited to either address the concerns or explain their rationale for not doing so. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Partly ********** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: No ********** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: No ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ********** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: Comment: Negative Affect (NA) is sometimes presented as a control variable, sometimes as a moderator and sometimes as a predictor. In my experience, it’s customary to make one such prediction rather than casting one variable in all three roles. See the following instances, with line numbers embedded: Abstract: Results showed physical 33 appearance-perfectionism explained variance (15%) in disordered eating symptoms above 34 general perfectionism and negative affect. BMI and negative affect were control variables as both 151 have demonstrated significant associations with DES (31). “general perfectionism and negative affect155 and both the PC and a modified PS subscale of physical-appearance-perfectionism would 156 predict unique variance in DES.” Comment: The weight distribution does not seem to match population stats. Probably because it is self-reported. This idea in the discussion, but only briefly. Please enhance discussion. The way it is currently presented it seems like you are saying BMI doesn’t predict correctly when it really may have been measurement error. See quote: “BMI 240 was unrelated to all variables and hence not included as a control variable subsequently.” Stats: P. 10, You briefly refer to data transformations: give specifics. Reviewer #2: The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism Thank you for the opportunity to review this manuscript. The authors conducted an interesting study exploring the dimensionality of perfectionism as it relates to self-compassion, and modelled both in the context of disordered eating in a sample of female university students. Background and introduction: 1. Literature review for perfectionism is, at times, confusing. From the reviewed articles in the introduction, the consensus appears that various assessments tap into varying dimensions of perfectionism. It is unclear which dimensions are of particular interest to the authors. These sections could benefit from some reorganization to highlight either the existing diversity in the understanding of perfectionism dimensionality, or if the literature is well-established in select dimensions, have a more focused discussion surrounding those specific dimensions of interest. 2. I wonder if the authors could clarify the proposed model and role of self-compassion a little more. In text, the model appears as one of moderation. The graphical representation of this model via Fig. 1 appears more like divergent trajectories marked by adaptive vs. maladaptive coping strategies in response to levels of self-compassion (mediated moderation). Perhaps both in-text description and graphical representation could be better harmonized. 3. Further in response to the discussion of adaptive vs. maladaptive coping strategies (in Fig 1), the emotion regulation literature has also suggested that psychological flexibility is more adaptive overall as strategies like reappraisal may not always be used “adaptively”. For example, some may reappraise the situation to be much more negative than it is. In addition, previously regarded negative coping, such as avoidance, has some basis from a functional contextualism view. Self-compassion may be more closely tied to psychological flexibility, which speaks to the underlying mechanisms of self-compassion, including acceptance, mindful awareness, and non-judgement (DOI: 10.1891/0889-8391.30.1.60). 4. Lines 149-150. With changes in self-perception in the age of social media, a challenge may be that the manifestations of DES is not well conceptualized in males. I suggest the authors rephrase this sentence to the tune of… “Given challenges in the conceptualizations and manifestations of DES in males, we limited the recruitment of only female university students…”. 5. Lines 159-159. Since the authors only recruited females in their study, the statement in relation to women does not apply. Suggest to reword. Methods: 1. Aside from the possibility of receiving a gift card, where participants otherwise compensated for their time? If not, please discuss limitations of opportunity sampling in relation to self-selection bias in the discussion section. 2. Please include reliability statistics of the sampled population (Cronbach’s alpha) for each measure used. 3. 25% missing data is a significant amount to be replaced with mean. Were the mean replacement values the sample mean or individual mean? Were the missing data missing at random, or more concentrated to specific measures? 4. Given the sensitivity of modelling new relationships and the large quantity of missing data, I suggest the authors use an alternative method to deal with missing values that accounts for more error biases in data (e.g., see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3668100/) 5. What data transformation was used? Would be helpful if authors produced a table (supplementary is also okay) documenting the raw means, sd, range and the transformed values. Results: 1. Please note that the DASS-21 is a self-report measure of depression, anxiety, and stress, and should be reported as three separate subscales rather than collapsed into a single variable (Lovibond, S.H. & Lovibond, P.F. (1995). Manual for the Depression Anxiety & Stress Scales. (2nd Ed.)Sydney: Psychology Foundation.). Subsequent analyses should adopt the three-subscale variable structure, and re-analyzed. 2. The rationale for why general perfectionism was entered in step 2 of the hierarchical regression (rather than step 1) was not clearly articulated. 3. Would the authors please include a table of model coefficients for all variables in the hierarchical regressions conducted? 4. Did the authors account for the shared variance of these variables prior to the regression through a single block entry? If not, what was the justification for the order of entry given that correlations at least suggest there may be some overlap or shared variance? 5. Line 228 indicated PROCESS macro would be used for the purpose of conducting moderation. Why did the authors choose a hierarchical regression in lieu of PROCESS? Discussion: 1. A central theme of this paper is the author’s modification of the PAPS from “hope” to “strive”. Could the authors elaborate more on why a change in terminology may have tapped into a different dimension that the previous version was not successful in tapping into? 2. Can the authors speak more to why there was an interaction with worries-about-appearance imperfection when those in low perfectionism and high in self-compassion also were showed highest in disordered eating? What are some of the potential confounds not otherwise controlled for? For example, if it is the existence of high distress, the “stress” subscale of the DASS-21 would at least tap into the self-reported aspect of it (and similarly for self-reported levels of depression and anxiety). 3. Given the interaction that did not support the hypothesized role of self-compassion as a moderator of perfectionism and DES, the extent to which results support self-compassion as a moderator should be more cautiously interpreted. 4. It would be important for the authors to discuss some cultural implications and variations in self-compassion as it relates to disordered eating and coping, as well as discuss how results may or may not be different in a male sample, older/younger samples, etc. Minor: • There are several instances where the paragraph breakdown appears misaligned. This is likely an issue when adapting word format to pdf, or vice versa. May be worth fixing for ease of review. ********** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Karen Shackleford Reviewer #2: No [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step. Submitted filename: Review_Oct17.docx Click here for additional data file. 9 Dec 2019 Dear Dr. Vickers, Thank you for identifying areas in need of improvement in our paper titled “The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism” and thereby giving us the opportunity to strengthen our research prior to publication. We appreciate the time and effort you and the reviewers have dedicated to delivering detailed feedback on our manuscript. We have been able to incorporate changes to reflect most of the suggestions provided by the reviewers. Here is a point-by-point response to the reviewers’ comments and concerns. Data Availability Required Revision 1: It is extremely important that the data be available. Despite using different computers and browsers, I was not able to open the data using the link provided. Attempts to do so resulted in an error message saying: DOI Not Found 10.17630/39426414-eb9b-4b7f-b078-2f88f96cdd76 Reviewer 2 also noted this issue. Response: Thank you for pointing this out. We have been in contact with the University of St Andrews Research Data and Information Services. The DOI (https://doi.org/10.17630/39426414-eb9b-4b7f-b078-2f88f96cdd76) has now been activated and the data supporting our research can be found here. Introduction Reviewer #1 Comment 1: Negative Affect (NA) is sometimes presented as a control variable, sometimes as a moderator and sometimes as a predictor. In my experience, it’s customary to make one such prediction rather than casting one variable in all three roles. See the following instances, with line numbers embedded Response: Thank you for pointing this out. In the hierarchical regressions we included negative affect as a predictor in order to assess the effects of general and physical appearance perfectionism above this known predictor of disordered eating symptoms. Negative affect was used as a control variable in the moderation analysis as prior research suggests it predicts variance both in perfectionism levels and disordered eating symptoms. We have made this distinction clearer in the introduction and results. Reviewer #2 Comment 1: Literature review for perfectionism is, at times, confusing. From the reviewed articles in the introduction, the consensus appears that various assessments tap into varying dimensions of perfectionism. It is unclear which dimensions are of particular interest to the authors. These sections could benefit from some reorganization to highlight either the existing diversity in the understanding of perfectionism dimensionality, or if the literature is well-established in select dimensions, have a more focused discussion surrounding those specific dimensions of interest. Response: Thank you for highlighting this. We have extended our discussion of the dimensions of perfectionism which are of particular interest in our study and have given more detailed definitions of the respective dimensions in our introduction. As we are examining general perfectionism and physical appearance perfectionism we focus on these constructs. We have expanded our discussion of general perfectionism to indicate why we are focussing on its subcomponents of striving and concern. Comment 2: I wonder if the authors could clarify the proposed model and role of self-compassion a little more. In text, the model appears as one of moderation. The graphical representation of this model via Fig. 1 appears more like divergent trajectories marked by adaptive vs. maladaptive coping strategies in response to levels of self-compassion (mediated moderation). Perhaps both in-text description and graphical representation could be better harmonized. Response: Thank you for this helpful comment. We have addressed it by changing the wording in figure 1. The mediating role of emotion regulation is simply a suggestion for a mechanism as to how or why self-compassion may exert its protective effect. However, in our study we were not testing this and instead concentrating on the role of self-compassion as moderator. Comment 3: Further in response to the discussion of adaptive vs. maladaptive coping strategies (in Fig 1), the emotion regulation literature has also suggested that psychological flexibility is more adaptive overall as strategies like reappraisal may not always be used “adaptively”. For example, some may reappraise the situation to be much more negative than it is. In addition, previously regarded negative coping, such as avoidance, has some basis from a functional contextualism view. Self-compassion may be more closely tied to psychological flexibility, which speaks to the underlying mechanisms of self-compassion, including acceptance, mindful awareness, and non-judgement (DOI: 10.1891/0889-8391.30.1.60). Response: Thank you for raising this interesting point. We have included a discussion of the role of psychological flexibility in our introduction and have refrained from labelling emotion regulation strategies as generally adaptive or maladaptive and instead focussed on situation-appropriate strategy selection and use. It is important to point out that although research has found self-compassion and psychological flexibility to be related, self-compassion was found to explain variance in well-being beyond psychological flexibility (Marshall & Brockmann, 2016; DOI: 10.1891/0889-8391.30.1.60). Comment 4: Lines 149-150. With changes in self-perception in the age of social media, a challenge may be that the manifestations of DES is not well conceptualized in males. I suggest the authors rephrase this sentence to the tune of… “Given challenges in the conceptualizations and manifestations of DES in males, we limited the recruitment of only female university students…”. Response: Thank you for this helpful comment. We have rephrased the sentence to better justify our choice of sample and have added a recent review paper as a reference, which highlights current lack of valid assessment tools for disordered eating in males. Comment 5: Lines 159-159. Since the authors only recruited females in their study, the statement in relation to women does not apply. Suggest to reword. Response: Thank you. We agree and have reworded the sentence to reflect this comment. Methods Required Revision 1: Please report the reliability values for the measures as reflected in the current sample (as noted by Reviewer 2). Comment Reviewer #2: Please include reliability statistics of the sampled population (Cronbach’s alpha) for each measure used. Response: Thank you for this suggestion and highlighting the importance of internal reliability for measures used. However, we would like to point out that Cronbach’s Alpha for each measure used in the current sample is reported in our manuscript in the final column of Table 1. Required Revision 2: Please present the DASS-21 as 3 subscales (as noted by Reviewer 2). Comment Reviewer #2: Please note that the DASS-21 is a self-report measure of depression, anxiety, and stress, and should be reported as three separate subscales rather than collapsed into a single variable (Lovibond, S.H. & Lovibond, P.F. (1995). Manual for the Depression Anxiety & Stress Scales. (2nd Ed.)Sydney: Psychology Foundation.). Subsequent analyses should adopt the three-subscale variable structure, and re-analyzed. Response: You have raised an important point here. However, based on recent research findings from Osman et al. (2012) - https://doi-org.wwwdb.dbod.de/10.1002/jclp.21908 - we argue that it is more appropriate to represent the DASS21 as a total score of negative affect or general distress. Osman et al. (2012) investigated the factor structure of the DASS21 in a large sample of nonclinical US university students and found strong evidence for a general distress factor which accounted for the majority of the variance in the DASS21 individual items and the total score. They conclude that the DASS21 is “a theoretically relevant measure of negative emotions that include mixed symptoms of depression, anxiety, and stress”. To enhance our manuscript, we have included this reference and an explanation for using the DASS21 as a measure of negative affect in the methods section of our manuscript. Required Revision 3: Please describe the transformations applied and how the transformations were selected (as noted by both reviewers). Comment Reviewer #1: You briefly refer to data transformations: give specifics. Comment Reviewer #2: What data transformation was used? Would be helpful if authors produced a table (supplementary is also okay) documenting the raw means, sd, range and the transformed values. Response: We agree and have included a more detailed description of the transformations applied. Following Reviewer 2’s suggestion, we have included a table documenting the raw means, sd, range and transformed values for the respective variables (please see supplementary materials for the table). Reviewer #2 Comment 1: Aside from the possibility of receiving a gift card, where participants otherwise compensated for their time? If not, please discuss limitations of opportunity sampling in relation to self-selection bias in the discussion section. Response: Thank you for this comment. Aside from the possibility of receiving the gift card, participants were not otherwise compensated for their time. Hence, we have now included a discussion of the limitations associated with opportunity sampling. Results Required Revision 1: Please clarify whether the authors are referring to the analysis performed by the PROCESS macro (for moderation) as regression. (Reviewer 2 also notes this issue, Results comment 5; it appears that the reviewer thought that the PROCESS macros was not used, given the way in which the analyses were described). Did the authors use the PROCESS macro? If so, please clarify the language. As Hays notes (http://processmacro.org/faq.html) the regression approach may yield the same results as PROCESS but differences are also possible.: “…When the same model is estimated using the same data with the same output options, the results will be the same as what you get with SPSS or SAS's regression procedures. There are many sources of discrepancies you may notice when discrepancies exist...A common one is requesting heteroscedastity-consistent standard errors in PROCESS, which are different than standard OLS standard errors. SPSS and SAS won't generate these standard errors, but PROCESS will (as will my RLM and HCREG macros) but only if you ask for them. When you do, standard errors, t-values, p-values, and confidence intervals are different than what SPSS and SAS's internal regression procedures produce, as they should be.” Comment Reviewer #2: Line 228 indicated PROCESS macro would be used for the purpose of conducting moderation. Why did the authors choose a hierarchical regression in lieu of PROCESS? Response: Thank you for highlighting this very important point. We would like to clarify that PROCESS was used in the moderation analyses. We have adjusted our description of the analyses in the results section to reflect this. We have further used heteroscedasticity-consistent standard errors and adjusted the affected results. Reviewer #2 Comment 1: The rationale for why general perfectionism was entered in step 2 of the hierarchical regression (rather than step 1) was not clearly articulated. Response: Thank you for pointing this out. Please see our response to comment 3 below for a detailed explanation. We have more clearly explained the rationale for this choice in our manuscript introduction. Comment 2: Would the authors please include a table of model coefficients for all variables in the hierarchical regressions conducted? Response: Thank you for your comment. We have included a table of model coefficients for all variables in the two hierarchical regressions conducted. Comment 3: Did the authors account for the shared variance of these variables prior to the regression through a single block entry? If not, what was the justification for the order of entry given that correlations at least suggest there may be some overlap or shared variance? Response: Thank you. We did not use a single block entry to account for the shared variance prior to the regression due to the fact that we had a theoretical rationale for entering our variables in the order that we did. The single block entry method is often considered controversial (Tabachnik & Fidell, 2007) and mainly useful in exploratory research. The rationale for entering physical appearance perfectionism first and then general perfectionism was to assess the unique amount of variance accounted for in disordered eating symptoms by general perfectionism above that accounted for by physical appearance perfectionism. In the second hierarchical regression, negative affect was entered first to assess the variance uniquely accounted for by this known risk factor. Then general perfectionism was entered to again assess the unique variance accounted for in disordered eating above that accounted for by negative affect. Finally, physical appearance perfectionism was entered, our main variable of interest in this case, to accurately assess the unique additional variance explained in disordered eating by physical appearance perfectionism. Where possible, we have attempted to make this rationale clearer in our manuscript in the introduction. Discussion Reviewer #1 Comment 1: The weight distribution does not seem to match population stats. Probably because it is self-reported. This idea in the discussion, but only briefly. Please enhance discussion. The way it is currently presented it seems like you are saying BMI doesn’t predict correctly when it really may have been measurement error. See quote:“BMI 240 was unrelated to all variables and hence not included as a control variable subsequently.” Response: Thank you for pointing this out. We have addressed this both in the results and the discussion by explaining in more detail that this finding was unexpected and that it may be due to measurement error. Reviewer #2: Comment 1: A central theme of this paper is the author’s modification of the PAPS from “hope” to “strive”. Could the authors elaborate more on why a change in terminology may have tapped into a different dimension that the previous version was not successful in tapping into? Response: Thank you for your comment. We have included a more detailed description of the rationale for the modification in the introduction as well as an additional sentence in the discussion. Comment 2: Can the authors speak more to why there was an interaction with worries-about-appearance imperfection when those in low perfectionism and high in self-compassion also were showed highest in disordered eating? What are some of the potential confounds not otherwise controlled for? For example, if it is the existence of high distress, the “stress” subscale of the DASS-21 would at least tap into the self-reported aspect of it (and similarly for self-reported levels of depression and anxiety). Response: Thank you for your comment. We have included your advice by elaborating in the discussion session on potential variables which may explain the crossover interaction but were not included in our study. We already controlled for negative affect in our moderation analyses. We only controlled for negative affect as one variable based on the findings from Osman et al (2012), which suggest that a general distress factor explains more variance in DASS-21 than the separate subscales of depression, anxiety and stress. Comment 3: Given the interaction that did not support the hypothesized role of self-compassion as a moderator of perfectionism and DES, the extent to which results support self-compassion as a moderator should be more cautiously interpreted. Response: Thank you. We see your point and added a cautioning statement in our limitations section. However, considering the large sample and the use of robust methods, we do believe it is valuable to highlight that the findings suggest self-compassion may function as a moderator. Comment 4: It would be important for the authors to discuss some cultural implications and variations in self-compassion as it relates to disordered eating and coping, as well as discuss how results may or may not be different in a male sample, older/younger samples, etc. Response: We agree. To address this comment, we have included a more detailed discussion of our sample in the limitations section of the discussion. We look forward to hearing from you in due time regarding our submission and to respond to any further questions and comments you may have. Sincerely, Barbara Dritschel and Luisa Bergunde Submitted filename: Response to Reviewers PLOS one.docx Click here for additional data file. 23 Dec 2019 The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism PONE-D-19-24566R1 Dear Dr. Dritschel, We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements. Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication. Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. With kind regards, Kristin Vickers, Ph.D. Academic Editor PLOS ONE Additional Editor Comments: The authors did a truly outstanding job in addressing each and every comment made by reviewers in a thoughtful and detailed fashion. The authors presented a thorough revision and a much improved manuscript. Care was taken to make many changes to the manuscript, and the authors also advanced well-explained reasoning (backed up by references) for why a few comments were not actioned. 2 Jan 2020 PONE-D-19-24566R1 The shield of self-compassion: a buffer against disordered eating risk from physical appearance perfectionism Dear Dr. Dritschel: I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. For any other questions or concerns, please email plosone@plos.org. Thank you for submitting your work to PLOS ONE. With kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Kristin Vickers Academic Editor PLOS ONE
  41 in total

1.  The impact of age and BMI on Eating Disorder Examination Questionnaire (EDE-Q) scores in a community sample.

Authors:  Øyvind Rø; Deborah L Reas; Jan Rosenvinge
Journal:  Eat Behav       Date:  2012-01-06

2.  Exploring domain-specific perfectionism.

Authors:  Siobhain McArdle
Journal:  J Pers       Date:  2010-04

Review 3.  Positive conceptions of perfectionism: approaches, evidence, challenges.

Authors:  Joachim Stoeber; Kathleen Otto
Journal:  Pers Soc Psychol Rev       Date:  2006

4.  The eating attitudes test: psychometric features and clinical correlates.

Authors:  D M Garner; M P Olmsted; Y Bohr; P E Garfinkel
Journal:  Psychol Med       Date:  1982-11       Impact factor: 7.723

5.  The Depression Anxiety Stress Scales-21 (DASS-21): further examination of dimensions, scale reliability, and correlates.

Authors:  Augustine Osman; Jane L Wong; Courtney L Bagge; Stacey Freedenthal; Peter M Gutierrez; Gregorio Lozano
Journal:  J Clin Psychol       Date:  2012-08-28

Review 6.  Perfectionism and eating disorders: current status and future directions.

Authors:  Anna M Bardone-Cone; Stephen A Wonderlich; Randy O Frost; Cynthia M Bulik; James E Mitchell; Saritha Uppala; Heather Simonich
Journal:  Clin Psychol Rev       Date:  2006-12-30

7.  The impact of manipulating personal standards on eating attitudes and behaviour.

Authors:  Roz Shafran; Michelle Lee; Elizabeth Payne; Christopher G Fairburn
Journal:  Behav Res Ther       Date:  2005-10-28

8.  Perfectionism traits and perfectionistic self-presentation in eating disorder attitudes, characteristics, and symptoms.

Authors:  P L Hewitt; G L Flett; E Ediger
Journal:  Int J Eat Disord       Date:  1995-12       Impact factor: 4.861

9.  A catalog of biases in questionnaires.

Authors:  Bernard C K Choi; Anita W P Pak
Journal:  Prev Chronic Dis       Date:  2004-12-15       Impact factor: 2.830

Review 10.  Eating disorders in adolescent and young adult males: prevalence, diagnosis, and treatment strategies.

Authors:  Christine A Limbers; L Adelyn Cohen; Bethany A Gray
Journal:  Adolesc Health Med Ther       Date:  2018-08-10
View more
  2 in total

1.  Physical Appearance Perfectionism: Psychometric Properties and Factor Structure of an Assessment Instrument in a Representative Sample of Males.

Authors:  Robin Rica; María Solar; Alba Moreno-Encinas; Sara Foguet; Emilio Juan Compte; Ana Rosa Sepúlveda
Journal:  Front Psychol       Date:  2022-02-18

2.  Parental feeding practices and the relationship with parents in female adolescents and young adults with eating disorders: A case control study.

Authors:  Maria Gruber; Daniel König; Julika Holzhäuser; Deirdre Maria Castillo; Victor Blüml; Rebecca Jahn; Carmen Leser; Sonja Werneck-Rohrer; Harald Werneck
Journal:  PLoS One       Date:  2020-11-19       Impact factor: 3.240
  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.