PERFECTIONISM AND BIBLIOTHERAPY: EFFICACY OF A MINDFULNESS BASED INTERVENTION By TESSA WIMBERLEY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
Â© 2014 Tessa Wimberley
To my loving and supportive parents
4 ACKNOWLEDGMENTS I would like to thank my family, friends , and cohort for their warm and consistent encouragement as well as my advisor , Laurie Mintz, for her unwavering mentorship and compassion.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ............................ 4 LIST OF TABLES ................................ ................................ ................................ ...... 6 LIST OF FIGURES ................................ ................................ ................................ .... 7 ABSTRACT ................................ ................................ ................................ ............... 8 CHAPTER 1 INTRODUCTION ................................ ................................ .............................. 10 2 METH OD ................................ ................................ ................................ .......... 17 Participants ................................ ................................ ................................ ....... 17 Measures ................................ ................................ ................................ .......... 17 Procedure ................................ ................................ ................................ ......... 2 1 Intervention ................................ ................................ ................................ ....... 23 3 RESULTS ................................ ................................ ................................ ......... 28 Preliminary Analyses ................................ ................................ ........................ 28 Short Term Efficacy of Bibliotherapy Intervention ................................ ............. 29 Longer Term Efficacy of Bibliotherapy Intervention ................................ .......... 33 4 DISCUSSION ................................ ................................ ................................ ... 43 REFERENCES ................................ ................................ ................................ ........ 50 BIOGRAPHICAL SKETCH ................................ ................................ ...................... 54
6 LIST OF TABLES Table page 2 1 Table of Participant Demographics ................................ ............................... 26 3 1 Table of Mean s , Standard Deviations, and Repeated Measures ANOVAs for the Effects of Group X Time on Dependent Variables ................................ .. 35 3 2 A Summary Table for the Effects of Time on Dependent Variables in the Delayed Treatment Group ................................ ................................ ............ 36 3 3 A Summary Table for the Effects of Time on Dependent Variables in the Intervention Group ................................ ................................ ........................ 37
7 LIST OF FIGURES Figure page 2 1 Flow Chart of Participant Re tention ................................ .............................. 25 3 1 Figure depicting the Group X Time Interaction Effect for APS R Total Scores ................................ ................................ ................................ .......... 38 3 2 Figure depicting the Group X Time Interaction Effect for APS R Discrepancy Scores ................................ ................................ ................................ .......... 38 3 3 Figure depicting the Group X Time Interaction Effect for APS R High Standards Scores ................................ ................................ ......................... 39 3 4 Figure depicting the Group X Time Interaction Effect for PSS Total Scores . 39 3 5 Figure depicting the Group X Time Interaction Effect for PSS Distress Scores ................................ ................................ ................................ .......... 40 3 6 Figure depicting the Group X Time Interaction Effect for FFMQ Total Scores ................................ ................................ ................................ .......... 40 3 7 Figure depicting the Intervention Group s Time Effects for APS R Total Scores ................................ ................................ ................................ .......... 41 3 8 Figure depicting the Intervention Group s Time Effects for PANAS Negative Scores ................................ ................................ ................................ .......... 41 3 9 Figure depicting the Intervention Group s Time Effects for PSS Total Scores ................................ ................................ ................................ .......... 42 3 10 Figure depicting the Intervention Group s Time Effects for FFMQ Total Scores ................................ ................................ ................................ .......... 42
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PERFECTIONISM AND BIBLIOTHERAPY: EFFICACY OF A MINDFULNESS BASED INTERVENTION By Tessa Wimberley August 2014 Chair: Laurie Mintz Major: Psychology This study examined the efficacy of a mindfulness based bibliotherapy intervention to reduce perfectionism and assoc iated psychological distress. Fifty six individuals re sp onded to an advertisement for participation in a study on perfectionism and distress, and were assigned to either the intervention or the wait list control group . The intervention group completed the Almost Perfect Scale Revised (APS R; Slaney et al., 2001 ), the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), the Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983), and the Five Facet Mindfulness Questionnaire (FFMQ ; Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006 ), read the self help book under study in 6 weeks, and completed the measures a second ti me. The control group completed the same measures 6 weeks apart . Results demonstrated that the interv ention group made statistically greater gains over time as com pared with the control group on measures of perfectionism , perceived stress , and mindfulness. Participants in the intervention group completed a 6 week follow up study and were shown to m aintain their gains on these outcomes over time. Additionally, partic ipants in a delayed treatment group were those
9 in the wait list control group who were given the intervention following the a six week waiting period, and six weeks later were shown to mirror gains found in the intervention group. The results of this study have important implications for the treatment of perfectionism and the continued investigation of efficacious self help approaches in general and for perfectionism specifically.
10 CHAPTER 1 INTRODUCTION Widely regarde maladaptive aspects have been the focus of several different research lines over the past decade . In general, adaptive perfectionism pertains to an individual having high standards for ones elf and it has been associated with positive affect, academic achievement, academic adjustment, high self esteem, and social adjustment (Rice, Leever, Christopher, & Porter, 2006) . Maladaptive perfectionism is assumed to be present when high standards comb ine with feelings of disproportionate, self critical perceptions of inadequacy; this is referred to in the perfectionism literature as discrepancy (Slaney, Rice, Mobley, Trippi, & Ashby, 2001) . In other words, discrepancy is the disparity between an indivi perception of their performance . Maladaptive perfectionism has been associated with varying forms of psychological distress, such as depression, anxiety, suicidality, eating disorders, substance related disorde rs, relationship difficulties, and low self esteem (Rice, Neimeyer, & Taylor, 2010) . Additionally, maladaptive perfectionism has been found to reduce the effectiveness of therapy for depression, suggesting that perfectionism is not only a contributing fac tor to a variety of presenting problems but also potentially a barrier in treatment for psychological distress (Blatt & Zuroff, 2005) . Maladaptive p erfectionism is common in university settings, with one study reporting that approximately one fourth to one third of student populations being classified as such (Rice & Ashby, 2007) . Given both the high prevalence of maladaptive
11 perfectionism 1 and its negative psychological effects, finding effective treatment is of great importance. Because of its accessibility for those who would otherwise not seek treatment, and findings of its efficacy in treating a multitude of other psychological concerns, including depression, anxiety, and sexual dysfunctions (Norcross, 2006), the utility of sel f help in treating maladaptive perfectionism is of particular interest . To date, however, the efficacy of cognitive behavioral therapy (CBT) self help techniques in reducing perfectionism has received mixed empirical support . Pleva and Wade (2007) tested t he effectiveness of a cognitive behavioral self help book ( ; Anthony & Swinson, 1998 ) for the treatment of perfectionism in a nonclinical sample of participants whose levels of perfectionism matched or surpassed levels previou sly reported in a study conducted with a clinical population . The study compared pure self help (PSH) and guided self help (GSH) conditions . Participants randomized to GSH received eight weekly 50 minute therapy sessions to complement their use of the self help material, whereas those randomized to PSH were provided only initial written instructions for using the book independently . Findings demonstrated that participants in both conditions decreased their levels of perfectionism, as well as obsessive compu lsive and depressive symptomatology . Those in the GSH condition experienced greater improvement than did those in the PSH condition, al though improvements in both groups were maintained after 3 months . Alarmingly, however, Pleva and Wade (2007) found that 20% of PSH participants experienced significant increases in 1 Unless otherwise noted, hereafter the term perfectionism will be used to refer to maladaptive perfectionism.
12 depressive symptoms over the treatment and follow up period in comparison to those in the GSH condition . Also alarming, the authors reported that one of the PSH participants committed suicide dur ing the course of the study . Overall, these findings indicate that utilizing cognitive behavioral self help methods for perfectionism can be effective in reducing OC and depressive symptomatology, with two rather critical caveats . First, CBT based biblioth erapy for perfectionism seems to be more beneficial for individuals when supplemented with in person, therapeutic guidance and second, there may be some individuals for whom CBT based self help is deleterious. Rather than eliminating pure self help as an option for those suffering from perfectionism, the identification of a pure self help modality that does not result in reduced functioning would be quite useful . Despite the popularity of CBT and its efficacy in treating perfectionism in a guided self help format, it may be that self help based in a different theoretical framework would be more effective . A second randomized controlled trial sheds light on this possibility . Rice, Neimeyer, and Taylor (2011) targeted both perfectionism and procrastination wi th two intervention conditions: CBT oriented bibliotherapy for procrastination (utilizing the book, The Now Habit; Fiore, 2007 ) and face to face Coherence Therapy . Findings from this study demonstrated that while there was no difference in the effectivenes s between the two treatments immediately following their conclusion, Coherence Therapy was more effective in reducing perfectionism when assessed at the six month follow up . Even more specifically, it was n perfectionism increased at six month follow up, whereas the effects of the self help condition weakened . While at first glance this finding seems to speak for the inability of self help to assist with
13 perfectionism, a more nuanced analysis indicates that it may have been the orientation of the self help material that was problematic. In discussing the results of their study and the difference in efficacy between CBT based self help and Coherence therapy, Rice et al. (2011) compare strategies adopted by th e two treatments utilized . They note that CBT based self help books, including the one they utilized, typically propose counteracting strategies based on cognitive behavioral principles, whereas Coherence Therapy acknowledges and integrates, rather than di sputes or counteracts, the underlying aspects of presenting problems . In other words, CBT based self help instructs the reader to dispute or counteract their underlying beliefs, whereas Coherence Therapy seeks to integrate the cluding their perfectionistic thoughts and associated make up . To quote Rice and colleagues , t his aspect of Coherence Therapy would support the likelihood of enduring change in a way that may differentiate it from approaches designed primarily around counteractive strategies that oppose the expression of problematic behaviors without embracing or fulfilling the underlying 132) . Given the weight that these authors place in the potential impact of fundamental differences in therapeutic approaches for perfectionism, continued investigation is clearly warranted . In fact, Rice, easonable extension of this preliminary study would be to pit two conceptually opposed, bona fide treatment
14 One approach that contrasts with the correctional nature of cognitive behavioral methods and that, of late, has received much empirical support in treating a variety of psychological issues is mindfulness . As described by Williams (2008), mindfulness by moment events as they un fold in the internal and external world; noticing habitual reactions to such events, often characterized by aversion or attachment; and, cultivating the ability to respond to events, and reactions to them, with an attitude of open curiosity (p. 721) . The underlying idea of acceptance which is inherent in this framework and which appears to correspond with Coherence Therapy, clearly differs from cognitive behavioral theories which generally directly challenge and attempt to change presenting problems (Rice, Neimeyer, & Taylor, 2011) . Baer (2003) reports a number of studies, seven of which were randomized control trials (RCTs), with findings suggesting that training in mindfulness practice is associated with decreased depression and anxiety (Ka bat Zinn, Massion, Kristeller, Peterson, Fletcher, Pbert, et al., 1992; Teasdale, Williams, Soulsby, Segal, Ridgeway, & Lau, 2000; Williams, Teasdale, Segal, & Soulsby, 2000), decreased levels of disordered eating (Kristeller & Hallett, 1999), decreased ch ronic pain (Kabat Zinn, 1982; Kabat Zinn, Lipworth, & Burney, 1985; Randolph, Caldera, Tacone, & Greak, 1999), physical and psychological improvements in patients with medical disorders (Kabat Zinn, Wheeler, Light, Skillings, Scharf, Cropley, et al., 1998; Speca, Carlson, Goodey, & Angen, 2000), and symptom improvements in patients with personality disorders (Kutz, Leserman, Dorrington, Morrison, Borysenko, & Benson, 1985) . Nevertheless, despite its proven effectiveness across a range of psychological disor ders, no RCT could be located which examined
15 the efficacy of mindfulness for alleviating distress associated directly with perfectionism . However, one correlational study conducted in a clinical sample of depressed participants may help to shed light on the relationship between mindfulness and perfectionism . Argus and Thompson (2008) found that perfectionism was associated with a lack of mindful awareness, and that this association was related to severity of depression symptoms . Thus, it appears that mind fulness may have potential application in the treatment of perfectionism and related distress . Yet, again, there have been no studies of either therapist administered or self administered mindfulness based interventions for perfectionism and it is not yet known if mindfulness based self help would be efficacious in reducing perfectionistic tendencies and associated psychological distress . Given the high prevalence of perfectionism and the accessibility of self help to those who otherwise may not seek treatm ent, this is a question worthy of investigation. The purpose of this study was to examine the efficacy of a mindfulness based self help book for treating perfectionism . Although, as noted earlier, Rice and colleagues (2011) called for the comparison of div erse therapeutic orientations in the treatment of perfectionism, given the lack of long term efficacy found for CBT based self help in one study (Rice et al., 2011) and the deterioration effect found for CBT based self help for a significant number of part icipants in another study (Pleva & Wade, 2007), it was deemed inadvisable to continue to examine the efficacy of this approach . Instead, this study aim ed to better understand how a mindfulness self help approach of maladapti ve perfectionism, as well as their perceived distress and negati ve affect, both of which have previously been associated with higher levels of maladaptive perfectionism ( Dunkley, Zuroff, & Blankstein, 2003 l Molnar, Reker,
16 Culp, Sadava, & DeCourville, 2006 ) . Additionally, the efficacy of the book to increase was also examined. In short , this study sought to determine if participants reading a mindfulness based self help book for perfectionism would ma ke greater gains over time in perfectionism, psychological distress and mindfulness comparison to a wait list control group and whether this improvement would be maintained at follow up . Additionally examined was the efficacy of the book among a delayed tr eatment group.
17 CHAPTER 2 METHOD Participants Sixty three adults completed pre test measures and were randomly assigned to the either the intervention or wait list control (WLC) group, with 30 being assigned to former and 33 being assigned to the latter . Twenty six of the 30 participants in the intervention group completed the post test measure (attrition rate = 13.3 %) and 30 of the 33 participants in the control group competed the post test measures (attrition rate = 9.1 %) . Thus, there were a total of 56 participants who completed both pre and post test assessments and who were included in the final sample. Additionally, of the 26 participants in the intervention group who completed the post test measure, 24 completed the follow up measure six weeks later and composed the intervention follow up sample. Finally, the 30 participants in the wait list control group who completed the post test were sent the intervention book; of these participants, 21 completed the series of mea sures again six weeks later and composed the delayed treatment sample. See Figure 2 1 for the flow of participants through the study. Table 2 1 presents demographic characteristics for the final sample, the intervention follow up sample, and the delayed tr eatment sample, respectively. Measures Perfectionism. Overall levels of perfectionisms, as well as three specific dimensions of perfectionism, were assessed using the total score and the three subscales scores of the 23 item Almost Perfect Scale Revised (A PS R; Slaney et al., 2001). Specifically, twelve items of the APS R assess discrepancy (experienced difference between expectation and reality of meeting standards), seven items assess
18 high standards (holding oneself to a high standard of performance and h aving high expectations of oneself), and four items assess order (preference for neatness and organization). Discrepancy represents the maladaptive dimension of perfectionism; item my best Participants indicated the extent to which they agreed or disagreed with each item on a 7 point Likert scale (1 = strongly disagree , 7 = strongly agree ). Scores on the Discrepancy subscale can range from 12 84 , scores on the High Standards subscale can range from 7 49, scores on the Order su bscale can range from 4 28, and T otal scores can range from 23 161 . In all cases, higher scores indicate higher levels of that aspect of perfectionism. Psychometric analyses on the APS R have supported the reliability and validity of the total score and the three subscales . Structure coefficients of items have been reported to range from .42 to .88 and internal reliability has been reported as follows: High Standards = .85, Discrepancy = .91, and Order = .82. (Slaney, Rice, & Ashby, 2002; Slaney et al., 2001). In this study, in t he final samp le ( N = 56), was .90 at pre test and .91 at post test for Total scores, .95 at pre test and . 95 at post test for Discrepancy, .66 at both pre test and post test for High Standards, and .59 at pre test and . 87 at post test for Order . Affect. Mood (or affect) was assessed using the 20 item Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). Ten items measure Positive Affect (subjective satisfaction and enjoyable engagement) and the remaining 10 items measure Negative Affect (subjective distress and un enjoyable engagement).
19 indicate how much they experienced feeling each item during a time period specified by researchers on a 5 p oint Likert scale (1 = not at all , 5 = extremely ). The chosen time both range from 10 50 , and a total score is not calculated. On both the Negative and Positive sca les, higher scores indicate higher levels of positive or negative affect, respectively been independently confirmed (Crocker, 1997) and the PANAS has been reported to have accepta 489) . Split half reliability for Positive Affect is reported at r = .87 and for Negative Affect as r = .78 (Watson et al., 1988). In the final sample ( N = 56) , fo r the Positive subscale was .87 for both pre test and post test ; the Negative subscale had s of .86 at pre test and .87 at post test . Perceived Stress. Perceived stress, distress, and coping were assessed using the 14 item Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983). Seven items compose the subscale measuring Perceived Distress (general distress capturing feelings of lack of control and negative affective reactions) and seven items measure the subscale of Perceived Coping (a positive perception of coping and self efficacy). Total scores on the PSS represent Perceived Stress via the combination of scores on the two subscales (i.e. perceived distress in addition to lack of perceived coping). Participants indicated the frequency with which they felt each item over the last month on a 5 point Likert scale (0 = Never , 4 = Very Often ). Item examples include:
20 abilities to handle your personal problems? Scores on both the Coping and Di stress subscale can range from 0 28 ; scores on the T otal scal e can range from 0 56. Due to reverse scoring for Coping, higher scores indicate lower levels of this construct. For Total scores and Distress, higher score s indicate higher levels of the construct. The .86; Cohen et al. , 1983) and good concurrent validity ( r = .62) in correlation with the Daily Stress Inventory (Lavoie & Douglas, 2011). For the purpose of this study, both the Total Scores and the two subscale scores were of interest. In the final sample ( N = 56) was .86 at pre test and .90 at post test for Total scores, .83 at pre test and .89 at post test for Perceived Coping, and .80 at pre test and .82 at post test for Perceived Distress . Mindfulness. Mindfulness was assessed using the 39 item Five Facet Mindfulness Questionnaire (FFMQ; Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006 ). The FFMQ is a questionnaire that assess es multiple components of mindfulness: (a) Nonreactivity to Inner Experience; (b) Observing; (c) Describing; (d) Nonjudging of Inner Experience; and (e) Acting with Awareness. Items are rated on a 5 point Likert scale (1 = Rarely or Very Rarely True , 5 = Very Often or Always True ). Item examples Scores on the Observing, Describing, Acting, and Nonju dging subscales can range from 8 40; scores on the Nonreactivity subscale can range from 7 35; an d Total scores can range from 39 19 5 . In all cases, higher scores indicate higher levels of the construct. In this study, only the total scores on the me asure were of interest. The FFMQ has demonstrated adequate to good internal consistencies for all facets in samples of
21 individuals who identify both as meditators and non .92; Baer et al., 2006; Baer et al., 2008). Construct validity of the facets has been demonstrated by their strong correlation with a variety of other constructs (e.g. openness to experience, emotional intelligence, and self compassion) that match predictions of the scale creators (Baer et al., 2006). In the final sam ple ( N = 56), was .88 at pre test and .89 at post test . Procedure An a priori power analysis determined that a minimum of 40 participants would be necessary to find a medium effect size at a power of .80. Because of this, as well as projected intervention group attrition rates of 30% 35% found in other bibliotherapy studies (e.g., Floyd, Scogin, McKendree Smith, Floyd, & Rokke, 2004; Malouff, Noble, Schutte, & Bhullar, 2010), a minimum sample size of 60 was sought. Specific ally, following the receipt of Campus Institutional Review Board (IRB) approval, recruitment advertisements were placed in local newspapers and radio, social media outlets, university wide email announcements, and locally posted flyers. The advertisements specifically sought individuals who identified as perfectionists, were currently experiencing distress associated with their perfectionism, and had internet access. Interested individuals responded via phone or email to these advertisements, and were provi ded with additional information regarding the study including being sent the informed consent to examine. Participants then indicated willingness to partake in the study and 67 participants indicated such willingness. These 67 participants were then sent a link to the informed consent and the pre test survey. Sixty three participants completed the pre test survey and were randomly assigned to either the intervention group or the WLC group. Intervention participants were then mailed a copy of the book
22 and in structed to read it as well as complete its exercises within 6 weeks. WLC participants were sent an email notifying them that they would receive the book in 6 weeks, after completing a second set of measures. Three weeks after the estimated arrival of the books to the Intervention group, emails were sent to both groups with a reminder that they had three more weeks to complete the book or three weeks remaining before receiving the book, for the intervention and WLC group respectively. Three weeks later, pos t test measures were emailed to both groups. In addition to the pre test measures, this assessment also included questions pertaining to whether participants had taken any action outside of the current study to address their perfectionism and, for the inte rvention group, a question about what page of the book they had read to and a question about what percentage of the exercises they had completed. Fifty percent of intervention group participants reported completing the book in full and the average page num ber reached for those who did not complete the book was 112 of 208 pages. On average, participants in the Intervention group reported completing 42% of the exercises in the book. Following completion of the post test surveys, WLC participants were mailed a copy of Present Perfect and provided identical reading instructions given to the Intervention group. They were also sent the same three week reminder that was sent to those in the Intervention group, reminding them they had three weeks to complete the bo ok. Six weeks following the estimated arrival of the book, they were sent the measures again, as well as the questions described above pertaining to book completion. Six weeks following the completion of the post test measures, participants in the Interven tion group were emailed a final link to follow up measures which, again,
23 included questions about actions taken outside of the current study to address their perfectionism. Following completion of the final study measures, all participants were fully debr iefed and provided with additional resources for mental health concerns. Across all survey administrations, participants not responding within five days were e mailed up to three reminders, each spaced 5 days apart. Incentives provided for study completion included both the book, given to the Intervention group after completing the pre test measures and to the WLC group following the post test measures, and the selection of a $5 gift card to either Amazon.com or Starbucks for those participants who complete d measures at all three time points. Intervention The current study examined bibliotherapy as an intervention strategy for reducing maladaptive perfectionism and associated distress. The self help book used was Present Perfect (2010) written by Pavel Somo v , a licensed clinical psychologist with expertise in the areas of both mindfulness based and acceptance and commitment therapies. The book introduces and explores fundamental manifestations of perfectionism that lead to distress. Within the context of a m indfulness approach, Present Perfect then identifies common examples and repercussions of maladaptive perfectionism in daily life and provides over 100 unique exercises. These exercises vary from reflective journaling and visualizations to concrete tasks (e.g . thoughts. Each and every time you have a new thought event, tap your finger. Remain a dispassionate observer of whatever pops into your head, as if you were sitting on the inevitable process of help
24 book is 208 pages. It has 15 chapters, which are contained within six parts: (a) Introduction to Perfectionism in General and to Your Perfectionism in Particular (one c hapter); (b) Perfectly Imperfect, Completely Incomplete, and Just So ( three c hapters); (c) Overcoming Mindfulness, Guilt, Shame, and Motivational Apathy (four c hapters); (d) Rehabilitation of Self View: Self Esteem, Self Acceptance, Am Ness (two c hapters); (e) Time, Performance, Uncertainty (four c hapters); and (f) Coexistence, Compassion, Connection (one chapter ).
25 Figure 2 1. Participant Flow Chart
26 Table 2 1. Participant Demographics Final Sample ( N =56) Delayed Treatment Sample ( N = 21) Intervention Follow Up Sample ( N = 24) M (or N ) SD (or %) M (or N ) SD (or %) M (or N ) SD (or %) Age 29.7 11.22 29.33 11.9 29 10.9 Gender Male 3 5.4% 1 4.8% 2 8.3% Female 53 94.6% 20 95.2% 22 91.7% Race/Ethnicity White/European American 35 62.5% 15 71.4% 13 54.2% Black/African American 1 1.8% 0 0 Biracial/Multiracial 1 1.8% 1 4.8% 0 Latino(a)/Hispanic 6 10.7% 3 14.3% 2 8.3% Asian/Pacific Islander 7 12.5% 1 4.8% 5 20.8% Native American 1 1.8% 0 1 4.2% Middle Eastern 4 7.1% 1 4.8% 2 8.3% Other 1 1.8% 0 1 4.2% Education High School Degree/GED 3 5.4% 1 4.8% 2 8.3% Some College 14 25 .0 % 5 23.8% 6 25 .0 % 6 10.7% 6 28.6% 0 5 8.9% 2 9.5% 2 8.3% Some Graduate/ Professional Training 10 17.9% 4 19 .0 % 3 12.5% 15 26.8% 2 9.5% 10 41.7% Doctoral Degree 3 5.4% 1 4.8% 1 4.2%
27 Table 2 1. Continued Final Sample ( N =56) Delayed Treatment Sample ( N = 21) Intervention Follow Up Sample ( N = 24) M (or N ) SD (or %) M (or N ) SD (or %) M (or N ) SD (or %) Income < $15,000 14 25 .0 % 5 23.8% 5 20.8% $15,000 $25,000 6 10.7% 2 9.5% 3 12.5% $25,000 $50,000 14 25 .0 % 5 23.8% 8 33.3% $50,000 $75,000 7 12.5% 4 19 .0 % 2 8.3% $75,000 $100,000 10 17.9% 2 9.5% 4 16.7% > $100,000 5 8.9% 3 14.3% 2 8.3% Religion Judaism 2 3.6% 0 1 4.2% Christianity 26 46.4% 10 47.6% 12 50 .0 % Islam 3 5.4% 1 4.8% 1 4.2% Non Religious 7 12.5% 3 14.3% 3 12.5% Atheist 5 8.9% 1 4.8% 2 8.3% Agnostic 9 16.1% 4 19% 4 16.7% Other 4 7.1% 2 9.5% 1 4.2% Primary Language English 49 87.5% 20 95.2% 19 79.2% Spanish 2 3.6% 0 2 8.3% Other 5 8.9% 1 4.8% 3 12.5% Sexual Orientation Heterosexual 45 80.4% 17 81 .0 % 19 79.2% Bisexual 7 12.5% 3 14.3% 2 8.3% Lesbian 2 3.6% 1 4.8% 1 4.2% Questioning 1 1.8% 0 1 4.2% Other 1 1.8% 0 1 4.2% Note. Of the 56 participants in the final sample, 26 (46%) were in the intervention condition and 30 (54%) were in the wait list co ntrol condition.
28 CHAPTER 3 RESULTS Preliminary Analyses Data was screened prior to main analyses to identify missing data points, identify and remove outliers in the outcome variables (Â±3 SD), and to affirm that skewness and kurtosis were within an acceptable range. Data was found to meet assumptions of normali ty following the use imputed expectation maximization to replace missing values (one point in PANAS post test and two points in FFMQ follow up) and the removal of outliers ( one in APS R Total, two in APS R High Standards, four in APS R Order, one in PANAS Negative, one in PSS Total, and three in PSS Coping) . A one way ANOVA was conducted to determine if significant pre test differences existed in demographics and outcome variables between participants who completed both the pre test and post test and those who completed the pre test only. No differences were found. Similarly, a one way ANOVA was performed to determine whether significant pre test differences existed between the intervention and WLC groups in terms of both demographic and outcome va riables. No differences were found. A one way ANOVA was also conducted to ascertain differences between post test outcome variables among those in the intervention group who completed the book and those who did not. Again, no differences were found. Lastly , final one way ANOVAs were performed within each group to determine differences between participants who completed measures at the third time point and those who completed at the first and second time points only. Again, no differences were found.
29 Short Term Efficacy of Bibliotherapy Intervention Final Sample. Results were analyzed via an omnibus multivariate analysis of variance (MANOVA) followed by separate univariate analyses of variance (ANOVAs) . While t his approach is widely consider ed a common and accepted analytic strategy , it does receive criticism for its assumed protection against Type I error inflation in running multiple subsequent univariate analyses ( Enders, 2003 ; Huberty & Morris, 1989 ) . Thus, in the current study, Bonferron i corrections were employed in subsequent univariate analyses to further protect against inflated Type I error. Because comparative changes over time (pre to posttest) and across groups (intervention vs. control) were under investi gation, the ANOVA resul t s below pertain to the Group X Time interaction . Additionally, significant main effects for time found in the absence of Group X Time Interactions are also reported. Also, due to the problems associated with null hy pothesis significance testing when interp reting social science data (see Ferguson, 2009 , for further discussion), effect sizes are presented for all variables, regardless of ANOVA significance level. B oth the d ) and percentage of variance explained ( p 2 ) are reported (Ferguson, 2009; Sink & Stroh, 2006), along d , .2 = small, .5 = medium, and .8 = large; for p 2 , .01 = small, .06 = medium, and .14 = large; Sink & Stroh, 2006). A one way MANOVA was conducted to deter mine the omnibus effect of group membership on the PANAS Negative, the PANAS Positive, and the remained of the outcome variables which yielded Total scores (i.e. APS R Total, PSS Total, and FFMQ Total ) at pre and post test. Differences between the interve ntion group and WLC group were statistically significant, F (1, 48) = 3.84 , p = .001, p 2 = .47 (large), observed power = .99. Separate follow up repeated measures ANOVAs were then performed for each
30 individual outcome variable, encompassing both total and relevant subscale scores at pre and post test to more specifically determine Group X Time interaction effects. The results of these ANOVAS are described below and presented in Table 3 1 . ANOVA results for perfectionism, as measured by the APS R, showed s ignificant interaction effects. As depicted in Figure 3 1 , the intervention group made statistically significant gains in Total APS R scores as compared to the WLC group, F (1, 53) = 20.21, p < .000, p 2 = .28 (large). Total APS R score means decreased in the intervention group from 138.71 at pre test to 115.21 at post test, whereas the WLC group total score means decreased negligibly (pre test M = 133.55, post test M = 128.42). The post test effect size ( d ) was .78 (medium); the observed power was .99. In examining the subscales for the APS R, similar interactions effects were found for both Discrepancy and High Standards. As depicted in Figure 3 2 , the intervention group means for Discrepancy decr eased from 66.40 at pre test to 50.24 at post test while the WLC means remained relatively unchanged (pre test M = 64.94, post test M = 59.39), F (1, 54) = 11.07, p < .01, p 2 d was .58 (medium) and the observed power was .90. Simila rly as depicted in Figure 3 3 , the High Standards means for the intervention group decreased from 46.21 at pre test to 42.24 at post test whereas the WLC group did not experience a decrease (pre test M = 45.74, post test M = 45.35), F (1, 53) = 14.84, p < .000, p 2 = .22 (large). Post test effect size d ) was .86 (large) and the observed power was .97. No interaction or time effect was found for the Order subscale of the APS d = .65, p 2 = .03, observed power = .26).
31 ANOVA results for the Negative Affect subscale of the PANAS revealed no d = .66, p 2 = .01, observed power = .12). However, there was a significant time effect for Negative Affect, F (1, 54) = 25.63, p < .000, p 2 = .32 (large), observed power = 1.0. Negative Affect means decreased from 31.84 at pre test to 27.36 at post test in the intervention group and from 35.13 at pre test to 31.84 at post test in the WLC group . Effect sizes were in the medium range for the intervention group ( d = . 63 ) and th e small range for the WLC group ( d = . 46 ). No interaction or time d = .41, p 2 = .06, observed power = .46). ANOVA results for the PSS revealed a significant interaction effect for the Total score. As depicted in Figure 3 4 , the PSS Total score mean decreased from 36.20 at pre test to 31.49 at post test, whereas the WLC group total score means remained basically unchanged (pre test M = 37.8, post test M = 36.31), F (1, 53) = 4.97, p < .05, p 2 = .09 (medium). Th d ) was .81 (large); the observed power was .59. Similar findings were seen for the Perceived Distress subscale. As depicted in Figure 3 5 pre test to 17.10 at post test while the WLC group remained basically the same (pre test M = 22.23, post test M = 21.27), F (1, 54) = 8.94, p < .01, p 2 = .14 (large), observed powe d was 1.03 (large). No interaction or time effect was found for d = .32, p 2 = .02, observed power = .06). The intervention group also made statistically significant gains in mindfulness as compared to the WLC gro up, F (1, 54) = 21.88, p < .000, p 2 = .29 (large). FFMQ total score means increased from 115.6 at pre test to 128.2 at post test for the intervention
32 group while the WLC group experienced no change (pre test M = 112.26, post test M = d was .89 (large) and the observed power was 1.0. See Figure 3 6 . Delayed Treatment Sample. A delayed treatment sample was utilized in this study to further investigate and potentially corroborate the short term effects of the intervention. WLC participants wh o completed the second set of surveys (i.e., given after six weeks of waiting) were then given the intervention, and six weeks later, completed a third set of surveys. One way Repeated measures ANOVAs with Bonferroni corrections were conducted to compare t dependent measures at pre test (i.e., those taken six weeks after waiting but prior to the intervention) and post test (i.e., those given six weeks later, following the intervention). These results are presented below and in Table 3 2 . On the APS R Total, significant time effects were found, demonstrating decreases in scores on this measure from pre test to post test ( M = 125.62 pre test and 115.24 post test); F (1, 20) = 6.19, p < .05, p 2 d = .52 (medium), observed power = .66. Similar time effects were found for both the APS R High Standards scores ( M = 44 .76 pre test and 41.86 post test), and APS R Order scores ( M = 24 .76 pre test and 23.48 post test): F (1, 20) = 9.85, p < .01, p 2 = .33 (large) d = .74 (medium), observed power = .85; F (1, 20) = 6.08, p < .05, p 2 = .23 (large), d = .51 (medium), observed power = .65, respectively. Additionally, significant time effects were found for the PSS Total scores ( M = 35.0 pre test and 29.9 post test), PSS Perceived Coping scores ( M = 14 .75 pre test and 13.55 post test), and PSS Perceived Distress scores ( M = 20.62 pre test and 16.62 post test): F (1, 19) = 19.39, p < .000, p 2 d = .97 (large), observed power = .99; F (1, 19) = 5.41,
33 p < .05, p 2 d = .44 (small), observed power = .60; F (1, 20) = 22.70, p < .000, p 2 d = . 93 (large), observed power = 1.0, respectively. Lastly, FFMQ Tota l scores also showed a significant effect over time with means increasing from 112.86 at pre test to 121.90 at post test, F (1, 20) = 8.23, p = .01, p 2 d = . 54 , observed power = .78. Scores for PANAS Positive Affect, PANAS Negative Affect, and APS R Discrepancy did not show significant effects over time ; p 2 d = . 23, .33, and .35 (respectively), observed power = .36, .43, and .41 (respectively) . Longer Term Efficacy of Bibliotherapy Interv ention Longer term efficacy of the book was assessed by examining the follow up responses given by the intervention group at six weeks post intervention. Repeated scores on the dependent variables at pre test, post test, and follow up. Significant time effects were found for APS R Total scores, APS R Discrepancy scores, APS R High Standards scores, and APS R Order scores: F (2, 44) = 31.83, p < .000, p 2 = .59 (large), observed power = 1.0; F (2, 46) = 25.16, p < .000, p 2 = .52 (large), observed power = 1.0; F (2, 42) = 12.23, p < .000, p 2 = .37 (large), observed power = .99; F (2, 42) = 4.74, p < .05, p 2 = .18 (large), observed power = .76, respectivel y. Additionally, a significant time effect emerged for PANAS Negative Affect scores, F (2, 44) = 13.32, p < .000, p 2 = .38 (large), observed power = 1.0. Time effects for PSS Total scores and Perceived Distress scores were also found to be significant: F (2, 46) = 14.44, p < .000, p 2 = .39 (large), observed power = 1.0; F (2, 46) = 19.26, p < .000, p 2 = .46 (large), observed power = 1.0, respectively. Lastly, FFMQ Total scores also showed a significant effect over time: F (2, 46) = 21.20, p < .000, p 2 = .48 (large),
34 observed power = 1.0. Scores for Positive Affect on the PANAS and Perceived Coping on the PSS did not show significant effects over time ; p 2 = .20 and .11 (respectively), observed power = .52 and .37 (respectively). Post hoc analyses indic ated that, in all aforementioned significant findings except for PANAS Negative Affect, pre and post test scores significantly differed, pre test and follow up scores significantly differed, but post test and follow up scores did not significantly differ. This demonstrates that, at the six week follow up, intervention group participants maintained gains made on measures of perfectionism, mindfulness behaviors, and perceived stress. In terms of scores on Negative Affect, in addition to showing significant d ifferences from pre test to post test and from pre test to follow up, there were also significant changes from post test to follow up. Of note, these scores decreased significantly from post test to follow up indicating that gains continued after the inter vention continued. See Table 3 3 for these results. Additionally, see Figures 3 7 to 3 10 for a depiction of change over time for APS R Total, PANAS Negative, PSS Total, FFMQ Total scores.
35 Table 3 1 . Mean, Standard Deviations, and Repeated Measur es ANOVAs for the Effects of Group X Time on Dependent Variables Intervention Group Wait List Control Group Pre Test Post Test Pre Test Post Test Measure M SD M SD M SD M SD F(1, 53) p d APS R Total *** 138.71 14.51 115.21 14.90 133.55 17.39 128.42 19.00 20.21 0.000 0.78 APS R Di sc ** 66.40 16.53 50.24 13.37 63.94 15.41 59.39 18.15 11.07 0.002 0.58 APS R Stand *** 46.21 3.65 42.25 4.21 45.74 3.28 45.35 3.04 14.84 0.000 0.86 APS R Order 23.40 4.40 21.96 4.09 24.50 3.51 24.23 2.87 1.78 0.190 0.65 PANAS Neg 31.84 8.11 27.36 6.18 35.13 6.91 31.84 7.48 0.60 0.440 0.66 PANAS Pos 34.92 7.98 37.08 6.71 35.10 5.83 34.58 5.60 3.62 0.060 0.41 PSS Total * 36.20 6.45 31.49 5.36 37.80 5.68 36.31 6.50 4.97 0.030 0.81 PSS Coping 15.12 2.88 14.40 2.41 15.77 2.47 15.26 2.96 0.11 0.750 0.32 PSS Distress ** 21.08 4.43 17.10 3.67 22.23 4.01 21.27 4.43 8.94 0.004 1.03 FFMQ Total *** 115.60 17.84 128.20 15.64 112.23 18.75 113.00 18.56 21.88 0.000 0.89 Note. N = 56 for analyses. Degrees of freedom for APS R Disc, FFMQ Total, PANAS Neg, PANAS Pos, and PSS Distress analyses were 1, 54. For all measures except PSS Coping, higher scores represent higher levels of the construct. Higher PSS Coping scores represen t lower levels of the construct. APS R Total = Almost Perfect Scale Revised Total Score (range = 23 161); APS R Disc = Almost Perfect Scale Revised Discrepancy subscale (range = 12 84); APS R Stand = Almost Perfect Scale Revised High Standards subscale (range = 7 49); APS R Order = Almost Perfect Scale Revised Order subscale (range = 4 28); P ANAS Neg = Positive and Negative Affect Schedule Negative subscale (range = 10 50); PANAS Pos = Positive and Negative Affect Schedule Positive subscale (range = 10 50); PSS Total = Perceived Stress Scale Total Score (range = 0 56); PSS Coping = Perceived S tress Scale Perceived Coping subscale (range = 0 28); PSS Distress = Perceived Stress Scale Perceived Distress subscale (range = 0 28); FFMQ Total = Five Facet Mindfulness Questionnaire Total Score (range = 39 195).
36 Table 3 2 . Summary Table for the Effects of Time on Dependent Variables in the Delayed Treatment Group Pre Test Post Test 95% Confidence Intervals Measure M SD M SD Pre to Post APS R Total * 125.62 18.09 115.24 22.10 [1.67, 19.09] * APS R Disc 56.10 17.11 49.90 18.05 [ 0.87, 13.25] APS R Stand ** 44.76 3.13 41.86 4.76 [0.97, 4.84] ** APS R Order * 24.76 2.21 23.48 2.80 [0.20, 2.37] * PANAS Neg 31.52 7.81 29.14 6.51 [ 0.28, 5.05] PANAS Pos 35.10 6.06 36.52 6.27 [ 3.21, 0.35] PSS Total *** 35.00 5.82 29.90 4.71 [2.68, 7.52] *** PSS Coping * 14.75 2.63 13.55 2.11 [0.12, 2.28] * PSS Distress *** 20.62 4.47 16.62 4.12 [2.25, 5.75] *** FFMQ Total ** 112.86 16.04 121.90 17.38 [ 15.63, 2.47] ** Note. N = 21 for analyses. For all measures except PSS Coping, higher scores represent higher levels of the construct. Higher PSS Coping scores represent lo wer levels of the construct. APS R Total = Almost Perfect Scale Revised Total Score (range = 23 161); APS R Disc = Almost Perfect S cale Revised Discrepancy subscale (range = 12 84); APS R Stand = Almost Perfect Scale Revised High Standards subscale (range = 7 49); APS R Order = Almost Perfect Scale Revised Order subscale (range = 4 28); FFMQ Total = Five Facet Mindfulness Questionnair e Total Score (range = 39 195); PANAS Neg = Positive and Negative Affect Schedule Negative subscale (range = 10 50); PANAS Pos = Positive and Negative Affect Schedule Positive subscale (range = 10 50); PSS Total = Perceived Stress Scale Total Score (range = 0 56); PSS Coping = Perceived Stress Scale Perceived Coping subscale (range = 0 28); PSS Distress = Perceived Stress Scale Perceived Distress subscale (range = 0 28).
37 Table 3 3 . Summary Table for the Effects of Time on Dependent Variables in the Intervention Group Pre Test Post Test 6 Week Follow Up 95% Confidence Intervals Measure M SD M SD M SD Pre to Post Post to Follow Pre to Follow APS R Total *** 139.87 13.65 114.91 15.16 112.57 17.25 [15.78, 34.14] *** [ 7.18, 11.88] [16.40, 38.21] *** APS R Disc *** 67.54 15.85 50.25 13.66 49.46 15.40 [10.29, 24.30] *** [ 6.32, 7.90] [9.93, 26.24] *** APS R Stand *** 46.23 3.80 42.50 3.65 41.91 4.37 [1.36, 6.10] ** [ 2.03, 3.21] [1.93, 6.71] *** APS R Order * 24.14 3.52 22.64 3.19 21.91 3.88 [ 0.63, 3.63] [ 0.81, 2.26] [.19, 4.27] * PANAS Neg *** 31.09 7.77 27.00 6.24 24.04 4.30 [0.75, 7.43] * [0.15, 5.76] * [2.70, 11.38] *** PANAS Pos 34.83 8.14 37.29 6.77 37.29 5.68 [ 5.31, 0.39] [ 3.12, 3.12] [ 5.85, 0.93] PSS Total *** 36.38 6.53 31.31 5.39 29.83 5.29 [1.77, 8.37] ** [ 1.92, 4.86] [3.35, 9.74] *** PSS Coping 15.25 2.86 14.29 2.40 14.46 2.19 [ 0.55, 2.47] [ 1.26, 0.93] [ 0.68, 2.27] PSS Distress *** 21.13 4.52 17.02 3.73 15.38 4.07 [1.94, 6.28] *** [ 1.06, 4.34] [3.26, 8.24] *** FFMQ Total *** 114.42 17.19 127.63 15.70 127.73 15.85 [ 19.00, 7.42] *** [ 7.17, 6.95] [ 18.53, 8.10] *** Note. N = 24 for analyses. For all measures except PSS Coping, higher scores represent higher levels of the construct. Higher PSS Cop ing scores represent lower levels of the construct. APS R Total = Almost Perfect Scale Revised Total Score (range = 23 161); APS R Disc = Almost Perfect Scale Revised Discrepancy subscale (range = 12 84); APS R Stand = Almost Perfect Scale Revised High Standards subscale (range = 7 49); APS R Order = Almost Perfect Scale Revised Order subscale (range = 4 28); PANAS Neg = Positive and Negative Affect Schedule Negative subscale (range = 10 50); PANAS Pos = Positive and Negative Affect Schedule Positive subscale (range = 10 50); PSS Total = Perceived Stress Scale Total Score (range = 0 56); PSS Coping = Perceived Stress Scale Perceived C oping subscale (range = 0 28); PSS Distress = Perceived Stress Scale Perceived Distress subscale (range = 0 28); FFMQ Total = Five Facet Mindfulness Questionnaire Total Score (range = 39 195). * p p p
38 Figure 3 1 . Group X Time Interaction Effect for APS R Total Scores Figure 3 2 . Group X Time Interaction Effect for APS R Discrepancy Scores 110 115 120 125 130 135 140 145 Pre-Test Post-Test Intervention Wait-List Control 45 50 55 60 65 70 Pre-Test Post-Test Intervention Wait-List Control
39 Figure 3 3 . Group X Time Interaction Effect for APS R High Standards Scores Figure 3 4 . Group X Time Interaction E ffect for PSS Total Scores 40 41 42 43 44 45 46 47 48 49 50 Pre-Test Post-Test Intervention Wait-List Control 30 31 32 33 34 35 36 37 38 39 40 Pre-Test Post-Test Intervention Wait-List Control
40 Figure 3 5 . Group X Time Interaction Effect for PSS Distress Scores Figure 3 6 . Group X Time Interaction Effect for FFMQ Total Scores 15 16 17 18 19 20 21 22 23 24 25 Pre-Test Post-Test Intervention Wait-List Control 100 105 110 115 120 125 130 Pre-Test Post-Test Intervention Wait-List Control
41 Figure 3 7 . Intervention Group Time Effects for APS R Total Scores Figure 3 8 . Intervention Group Time Effects for PANAS Negative Scores 100 105 110 115 120 125 130 135 140 145 150 Pre-Test Post-Test Follow-Up Intervention 20 22 24 26 28 30 32 34 Pre-Test Post-Test Follow-Up Intervention
42 Figure 3 9 . Intervention Group Time Effects for PSS Total Scores Figure 3 10 . Intervention Group Time Effects for FFMQ Total Scores 25 27 29 31 33 35 37 39 Pre-Test Post-Test Follow-Up Intervention 110 115 120 125 130 135 Pre-Test Post-Test Follow-Up Intervention
43 CHAPTER 4 DISCUSSION As far as the authors could determine, this was the first study to evaluate the efficacy of a mindfulness based bibliotherapy intervention for individuals struggling with perfectionism. Those who read the book under study made greater gains in reducing bot h overall levels of perfectionism as measured by the APS R (Slaney et al., 2001), as They also made greater gains in reduction of scores on a general measure of perceived stress, as well as the subscale measuring perceived distress. Finally, they also made greater gains in increasing levels of mindfulness than did participants in a wait list control group. Conversely, both those in the intervention group and those in the W LC group decreased their level of negative affect, with no significant differences in terms of this decrease. In short, these initial results indicated that those who read the book made significant improvements in perfectionism (overall, discrepancy, and h igh standards), perceived stress (perceived distress combined with perceived lack of coping), and mindfulness and that all participants (regardless of receipt of intervention) experienced a decrease in negative affect. Those who read the book were found to maintain all gains at a six week follow up. Specifically, gains were maintained in overall level of perfectionism, discrepancy, high standards, negative affect, perceived stress, perceived distress, and mindfulness. Results obtained with a delayed treatm ent sample confirmed the short term efficacy of the book, although there was not an identical correspondence of results found with the original intervention group. Specifically, while for both the original intervention sample and the delayed treatment samp le, there were decreases in overall
44 perfectionism, high standards, overall distress and coping, perceived distress, and increases in mindfulness, in the delayed treatment sample there were no changes in discrepancy evidenced after reading the book, nor wer e there changes in negative affect. There were changes, however, in the Order subscale of the ASP R (Slaney et al., 2001) and the Perceived Coping subscale of the PSS (Cohen, Kamarck, & Mermelstein, 1983) not shown by the original intervention sample. Desp ite these slight differences between the original intervention sample and the delayed treatment sample, it can still be concluded that the book under study was efficacious in decreasing overall levels of perfectionism and distress, as well as in increasing mindfulness among those who read it. Results pertaining to the construct of discrepancy deserve more detailed examination. Specifically, as noted previously, discrepancy is the aspect purported to be most associated with the maladaptive perfectionism and the one targeted by this, and other pe rfectionism interventions. In this study, the intervention group evidenced greater changes in discrepancy than the WLC group, and these changes were maintained at follow up; conversely, the delayed treatment group did not evidence changes in this important aspect of perfectionism. It appears that this was an issue of power for the delayed treatment sample. Nevertheless, given the importance of this construct, additional studies should be conducted to verify the change in discrepancy found in the original in tervention sample. The results pertaining to negative affect also deserve additional attention. Initially, both the intervention and the WLC groups evidenced a decrease in negative affect as measured by the PANAS (Watson, Clark, & Tellegen, 1988). At foll ow up,
45 those in the intervention group continued to evidence this decrease. On the other hand, those in the delayed treatment sample did not evidence any gains in negative affect, although they did evidence gains in perceived coping not seen with the inter vention sample. Stated simply, those receiving and reading the book without a waiting period showed a decrease in negative affect immediately after reading the book, with additional decreases six weeks later. Conversely, those waiting to read the book decr eased their affect during this waiting period, yet among those who subsequently read the book, there were no additional decreases in negative affect. They did, however, evidence increases in terms of perceived coping, a result not found in the original int ervention sample. Again, the issue of power in the delayed treatment sample may account for aspects of these variations. These differences may also have been the result of participants in the delayed treatment group experiencing the six week waiting interv al before receiving the intervention. Perhaps participants who spent time anticipating the receipt of the intervention had higher expectations for their potential increase in coping than did the intervention group. In regard to the lack of change observed in negative affect, it is possible that this finding is related to the similarly unchanged levels of discrepancy in the delayed treatment sample, since the two constructs have been shown to be related (Dunkley, Zuroff, & Blankstein, 2003). Future investiga tions could explore these results by further clarifying the effect of the intervention on both discrepancy and negative affect, as well as exploring the relationship between anticipation of the intervention and expected increases in coping among WLC partic ipants.
46 The results found in this study contribute to both the body of research supporting the efficacy of mindfulness interventions as well as the use of bibliotherapy as an effective mode of treatment . Continued empirical exploration along each of these research lines and their intersections is recommended . A future study should compare the effectiveness of this book with an in person mindfulness based or acceptance and commitment therapy treatment . One specific study of interest would be comparing the ef ficacy of this bibliotherapy intervention to the in person Coherence therapy found by Rice and colleagues (2011) to be efficacious in decreasing perfectionism. Past studies comparing bibliotherapy and face to face treatment have led to the conclusion that to face counseling is certainly van Lankveld, 2009, p. 150) . Whether this is true for the treatment of perfectionism is an empirical question warranting investigation . Another timely question worthy investigation is the comparat ive efficacy of this book and an internet based mindfulness intervention for perfectionism, perhaps even one created based on the content of this book if a pre existing comparable internet intervention is not available. Several other questions deserve inve stigation . Another f ruitful avenue would entail comparing the efficacy of this book with anot her self help book on the topic , including either another mindfulness based book or a book written from a different perspective. If the latter were undertaken, cau tion would need to be exercised if choosing a CBT book given the complex, and at times, deleterious results found for this type of intervention in the past ( Pleva & Wade, 2007 ). A lso worthy of investigation would be comparing the effectiveness of reading this book with or without accompanying minimal therapist support . It would be expected that supplementing this book with
47 therapist support would result in higher outcomes than simply reading the book alone. A dismantling study would help identify which cha pters, or treat ment approaches, contributed most to . Similarly , collecting a broader range of qualitative data regarding reactions to the book could assist in directing future interventions. Despite the many contributions that this stud y offers the existing literature surrounding mindfulness and bibliotherapy, methodological shortcomings were present, including those pertaining to sample size and demographic diversity. Although small sample sizes are common in bi bliotherapy studies (van Lankveld, 2009) and despite the fact that significant results and large effect sizes were still found, the generalizability of these findings remains limited due to the sample size of the study. Additionally, the primary recruitment from two large, public university settings, both pose large limitations in terms of the generalizability of these results. Future studies would benefit from seeking larger and more demographi cally diverse samples. A related limitation that has particular relevance to the field of counseling psychology pertains to the somewhat privileged nature of bibliotherapy as a treatment modality. While generally more accessible and affordable to most ind ividuals than face to face counseling, bibliotherapy remains a viable treatment option only to those who have the means and ability to purchase the book, the ability to easily read and synthesize the information that the book provides, as well as the avail able time to complete these former tasks. Future lines of research could explore the use and effectiveness of bibilotherapy (and/or other self help approaches) among individuals
48 who have lower income levels, fewer years of education, and perhaps more deman ding time schedules. Such studies could help us to understand the applicability of bibliotherapy to more diverse populations and could then assist in guiding the creation of such interventions with broader applicability across educational and income levels . . Despite the need for additional examination , clinicians can now feel comfortable recommending this intervention to those struggling with perfectionism. In fact, Norcross (2006) noted that in less s evere forms of distress, bibliotherapy may be particularly useful during initial stages of treatment and also be used in later stages to complement other counseling interventions. Indeed, as noted by Norcross (2006) and echoed by Mintz et al. (2012), books with evidence of their efficacy can be used in multiple ways by clinicians, including recommending during waiting periods p rior to face to face treatment , as an adjunct to face to face treatment, or after treatment concludes to bolster and maintain gains . Clinicians could implement Present Perfect in any of these ways. Colleges and universities might also benefit from including Present Perfect in orientation materials or freshman seminars. Given the high percentage of college students struggling with perfe ctionism (Rice & Ashby, 2007), this book could be used to intervene with college students early in their college career, hopefully preventing later more substantive struggles. While the aforementioned implications pertain to Present Perfect specifically, there are also broader implications that can be gleaned from this study. This study adds to the growing body of literature on the efficacy of self help, a modality that has the potential to reduce the burden placed on mental health practitioners, as well a s to aid in
49 diminishing client barriers to seeking treatment by pro viding more easily obtainable and effectiv e self . Even more specifically, this study has added to the growing body of literature on the efficacy of bibliotherapy, one form of self help. Nevertheless, few self help books are examined for efficacy and as stated by R osen, Glasgow, and More (2003): The only way to know the effectiveness of well intentioned instructional materials, when they are entirely self administered, is to test those specific materials in the specific c ontext of their intended usage. Psychologists who write self help materials based on methods they find effective in office settings have no assurance that the public can successfully apply these procedures on their own. (p. 410) The results of this study indicate that one particular book, Present Perfect , contains methods that individual s can effectively apply on their own to diminish perfectionism and related distress. The discovery of effective methods for reducing perfectionism and psychological distress in non clinical, high achieving populations might also have beneficial social imp lications . Specifically, such interventions have the potential to help high achieving individuals gain increased psychological a djustment and life satisfaction which, in turn, could allow them to attain a greater capacity for creativity and positive impact in their respective area s of expertise and personal influence. Such outcomes could contribute not only to the well being of the individuals accessing the intervention, but also to their personal and professional circles or to society as a whole . It is hop ed that this study will encourage counseling psychologists to conduct additional research regarding effective interventions for perfectionism and the efficacy of bibliotherapy for this and other commonplace, yet disruptive, concerns that are faced by the c lients and students we serve.
50 REFERENCES Antony, M. M., & Swinson, R. P. (1998). coping with perfectionism . Oakland, CA: New Harbinger Publications. Argus, G., & Thompson, M. (2008). Perceived Social Problem Solving, Perfectionism, and Mindful Awareness in Clinical Depression: An Exploratory Study. Cognitive Therapy and Research, 32 (6), 745 757. Baer, R. A. (2003). Mindfulness Training as a Clinical Intervention: A Conceptual and Empirical Review. Clinical Psy chology: Science and Practice, 10 , 125 143. Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self report assessment methods to explore facets of mindfulness. Assessment, 13 , 27 45. Blatt, S. J., & Zuroff, D. C. (2005). Emp irical evaluation of the assumptions in identifying evidence based treatments in mental health. Clinical Psychology Review, 25, 459 486. Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Be havior, 24 , 385 396. Crocker, P. R. E. (1997). A confirmatory factor analysis of the Positive and Negative Affect Schedule (PANAS) with a youth sport sample. Journal of Sport and Exercise Psychology, 19 , 91 97. Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self critical perfectionism and daily affect: Dispositional and situational influences on stress and coping. Journal of Personality and Social Psychology, 84 (1), 234 252. Enders, C. K. (2003). Performing multivariate group comparisons followi ng a statistically significant MANOVA . Measurement and Evaluation in Counseling and Development, 36 , 40 56 Fiore, N. (2007). The now habit: A strategic program for overcoming procrastination and enjoying guilt free play . New York, NY: J. P. Tarcher/Penguin . Ferguson, C. J. (2009). An effect size primer: A guide for clinicians and researchers. Professional Psychology: Research and Practice, 40 , 532 538. doi:0.1037/a0015808 Floyd, M., Scogin, F., McKendree Smith, N. L., Floyd, D. L., & Rokke, P. D. (2004). C ognitive therapy for depression: A comparison of individual psychotherapy and bibliotherapy for depressed older adults. Behavior Modification, 28 , 297 318. doi:10.1177/0145445503259284
51 77). Self Help in Mental Health . New York, NY: Springer New York. Huberty, C. J., & Morris, J. D. (1989). Multivariate analysis versus multiple univariate analyses. Psychological Bulletin, 105 , 302 308. Kabat Zinn, J. (1982). An outpatient program in behavioral med icine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. General Hospital Psychiatry, 4 , 33 47. Kabat Zinn, J., Lipworth, L, & Burney, R. (1985). The clinical use of mindfulness me ditation for the self regulation of chronic pain. Journal of Behavioral Medicine, 8 , 163 190. Kabat Zinn, J., Massion, M. D., Kristeller, J., Peterson, L. G., Fletcher, K. E., Pbert, L., et al. (1992). Effectiveness of a meditation based stress reduction p rogram in the treatment of anxiety disorders. American Journal of Psychiatry, 149 , 936 943. Kabat Zinn, J., Wheeler, E., Light, T., Skillings, Z., Scharf, M. J., Cropley, T. G., et al. (1998). Influence of a mindfulness meditation based stress reduction in tervention on rates of skin clearing in patients with moderate to severe psoriasis undergoing phototherapy (UVB) and photochemotherapy (PUVA). Psychosomatic Medicine, 50 , 625 632. Kristeller, J. L., & Hallett, C. B. (1999). An exploratory study of a medita tion based intervention for binge eating disorder. Journal of Health Psychology , 4 , 357 363. Kutz, I., Leserman, J., Dorrington, C., Morrison, C., Borysenko, J., & Benson, H. (1985). Meditation as an adjunct to psychotherapy. Psychotherapy and Psychosomati cs, 43 , 209 218. Lavoie, J. A. A., & Douglas, K. S. (2011). The Perceived Stress Scale: Evaluating Configural, Metric and Scalar Invariance across Mental Health Status and Gender. Journal of Psychopathology and Behavioral Assessment, 34 (1), 48 57. Malouff, J. M., Noble, W., Schutte, N. S., & Bhullar, N. (2010). The effectiveness of bibliotherapy in alleviating tinnitus related distress. Journal of Psychosomatic Research, 68, 245 251. doi:10.1016/ j.jpsychores.2009.07.023 Mintz, L. B., Balzer, A. M., Zhao, X ., & Bush, H. E. (2012). Bibliotherapy for low sexual desire: evidence for effectiveness. Journal of Counseling Psychology, 59 (3), 471 478. Molnar, D. S., Reker, D. L., Culp, N. A., Sadava, S. W., & DeCourville, N. H. (2006). A mediated model of perfectio nism, affect, and physical health. Journal of Research in Personality, 40 , 482 500.
52 Norcross, J. C. (2006). Integrating self help into psychotherapy: 16 practical suggestions. Professional Psychology: Research and Practice, 37 (6), 683 693. Pleva, J., & Wad e, T. D. (2007). Guided self help versus pure self help for perfectionism: a randomised controlled trial. Behaviour Research and Therapy, 45 (5), 849 861. Randolph, P. D., Caldera, Y. M., Tacone, A. M., & Greak, M. L. (1999). The long term combined effects of medical treatment and a mindfulness based behavioral program for the multidisciplinary management of chronic pain in West Texas. Pain Digest, 9 , 103 112. Rice, K. G., & Ashby, J. S. (2007). An efficient method for classifying perfectionists. Journal of Counseling Psychology, 54 , 72 85. Rice, K. G., Neimeyer, G. J., & Taylor, J. M. (2011). Efficacy of Coherence Therapy in the Treatment of Procrastination and Perfectionism. Counseling Outcome Research and Evaluation, 2 (2), 126 136. Rice, K. G., Leever, B. A, Christopher, J., & Porter, J. D. (2006). Perfectionism, stress, and social (dis)connection: A short term study of hopelessness, depression, and academic adjustment among honors students. Journal of Counseling Psychology, 53 (4), 524 534. Rosen, G. M., Gl asgow, R. E., & Moore, T. E. (2003). Self help therapy: The science and business of giving psychology away. In S. O. Lilienfeld, S. J. Lynn, & J. M. Lohr (Eds.), Science and pseudoscience in clinical psychology (pp. 399 424). New York: Guilford. Slaney, R. B., Rice, K. G., & Ashby, J. S. (2002). A programmatic approach to measuring perfectionism: The Almost Perfect Scales. In G. L. Flett & P. L. Hewitt (Eds.), Perfectionism: Theory, research, and treatment (pp. 63 88). Washington, DC: American Psychological Association. Slaney, R. B., Rice, K. G., Mobley, M., Trippi, J., & Ashby, J. S. (2001). The Revised Almost Perfect Scale. Measurement and Evaluation in Counseling and Development, 34 (3), 130 145. Speca, M., Carlson, L. E., Goodey, E., & Angen, M. (2000). A randomized, wait list controlled clinical trial: The effect of a mindfulness meditation based stress reduction program on mood and symptoms of stress in cancer outpatients. Psychosomatic Medicine, 62 , 613 622. Somov, P. (2010). Present perfect: A mindful ness approach to letting go of perfectionism and the need for control. Oakland, CA: New Harbinger Publications.
53 Teasdale, J. D., Williams, J. M., Soulsby, J. M., Segal, Z. V., Ridgeway, V. A., & Lau, M. A. (2000). Prevention of relapse/recurrence in major depression by mindfulness based cognitive therapy. Journal of Consulting and Clinical Psychology , 68 , 615 623. van Lankveld, J. (2009). Self help for sexual dysfunction. Journal of Sex Research, 46 , 143 155. doi:10.1080/00224490902747776 Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology , 54 , 1063 1070. Williams, J. M. G. (2008). Mindfulness, Depression and Modes of Mind . Cognitive Therapy and Research , 32 (6), 721 733. Williams, J. M. G., Teasdale, J. D., Segal, Z. V., & Soulsby, J. (2000). Mindfulness based cognitive therapy reduces overgeneral autobiographical memory in formerly depressed patients. Journal of Abnormal Psychology , 109 , 150 155.
54 BIOGRAPHICAL SKETCH Tessa E. Wimberley was born in Fairfax, Virginia in 1989. She graduated from Barron Collier High School in 2007. Tessa earned her Bachelor of Arts degree in Psychology from the Universi ty of Tampa, graduating Magna Cum Laude in spring of 2011 . In fall of 2012 Tessa joined the doctoral program in Counseling Psychology at the University of Florida. Her primary research interests pertain to the construct of perfectionism as well as effective interventions for psychological distress.