Development of the Professional Self-Care Scale

Development of the Professional Self-Care Scale

Katherine E. Dorociak, Patricia A. Rupert, Fred B. Bryant, and Evan Zahniser Loyola University Chicago

In recent years, there has been an increased emphasis on the importance of self-care for psychologists and other mental health professionals. With the growth of positive psychology and preventive medicine, self-care is an emerging topic, promulgated as a means of avoiding the adverse effects of stress and promoting professional functioning and well-being. However, the research on self-care is limited because of the lack of an empirically based, psychometrically sound measure of this construct. Thus, the purpose of this project was to develop a measure of professional self-care. Professional psychologists were the focus of study, with the goal being to develop a measure that can be used in this population and similar groups of professionals. Based on expert feedback and a preliminary study of 422 licensed psychologists in Illinois, a 5-factor, 21-item scale was created. Factor analysis identified the following self-care factors: Professional Support, Professional Development, Life Balance, Cognitive Awareness, and Daily Balance. Preliminary analyses provided initial support for the validity of the 5 factors. A follow-up study was conducted with a second sample of clinical psychologists. The 5-factor structure provided a good fit to the data with the second sample. Thus, based on factor analysis and validity data, a 5-factor, 21-item Professional Self-Care Scale was established for further study and use in future research.

Public Significance Statement This article describes the development of the Professional Self-Care Scale (PSCS) for practicing psychologists. The 21-item scale includes subscales assessing 5 dimensions of self-care: professional support, professional development, life balance, cognitive awareness, and daily balance.

Keywords: self-care, professional functioning, prevention, measure development

Practicing psychologists, as well as other mental health profes- sionals, face a myriad of professional and personal demands that they must manage to function in the workplace and in their daily lives. These demands create stress that, if not adequately managed, may have negative implications not only for the professionals themselves, but also for their clients. The process resulting in negative outcomes has been conceptualized as a downward spiral that begins when stress, in the absence of effective coping, leads to distress. Distress, described as the subjective emotional response in reaction to demands and stresses (Barnett, Johnston, & Hillard, 2006), when left unmonitored and unchecked, can then lead to multiple negative outcomes, including poor mental and physical health, professional burnout, and impaired functioning.

With the growth of positive psychology and preventive medi- cine, self-care has been increasingly emphasized as a means of enhancing well-being and preventing these negative outcomes. In

fact, given the potential for impaired functioning, self-care is an ethical responsibility for mental health professionals (e.g., Barnett, Baker, Elman, & Schoener, 2007; Norcross & Guy, 2007; Wise, Hersh, & Gibson, 2012). In psychology, for example, the Ameri- can Psychological Association’s Ethics Code (American Psycho- logical Association, 2002) states that “psychologists strive to be aware of the possible effect of their own physical and mental health on the ability to help those with whom they work” (Prin- ciple A) and directs psychologists to be aware of personal prob- lems and take action to prevent them from interfering with work performance (Standard 2.06). Other codes address self-care even more directly. The Canadian Code of Ethics for Psychologists (Canada Psychological Association, 2000) instructs psychologists to “engage in self-care activities that help to avoid conditions (e.g., burnout, addictions) that could result in impaired judgment and interfere with their ability to benefit and not harm others” (Prin- ciple II.12: Responsible Caring). Similarly, the Code of Ethics for the American Counseling Association (2014) states that “counsel- ors engage in self-care activities to maintain and promote their emotional, physical, and spiritual well-being to best meet their professional responsibilities” (Section C: Professional Responsi- bility).

Despite the attention to self-care in ethics codes and profes- sional literatures, research on self-care is limited. As a result, our understanding of self-care and our ability to offer empirically based recommendations in this important area are also limited. While there is clearly a need for more systematic study, the lack of

This article was published Online First March 9, 2017. Katherine E. Dorociak, Patricia A. Rupert, Fred B. Bryant, and Evan

Zahniser, Department of Psychology, Loyola University Chicago. Certain ideas and data presented in this article were based on the primary

author’s Master’s thesis project. An earlier version of the self-care measure was also discussed as part of a symposium on self-care presented at the American Psychological Association’s Annual Conference in August 2016.

Correspondence concerning this article should be addressed to Katherine E. Dorociak, Department of Psychology, Loyola University Chicago, 1032 W. Sheridan Road, Chicago, IL 60660. E-mail: kdorociak@luc.edu

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Journal of Counseling Psychology © 2017 American Psychological Association 2017, Vol. 64, No. 3, 325–334 0022-0167/17/$12.00 http://dx.doi.org/10.1037/cou0000206

325

 

 

a comprehensive, psychometrically sound measure of self-care poses a significant impediment to research. Defining and opera- tionalizing self-care through the creation of a measurement instru- ment represents a critical step in advancing research aimed at understanding and promoting effective self-care for mental health professionals.

The present research aimed to develop a brief but comprehen- sive self-report measure of self-care for professional psycholo- gists: the Professional Self-Care Scale (PSCS). Toward this end, we followed the guidelines for scale development offered by DeVellis (2012). The first step involved defining and clarifying the key elements of self-care, and for this we turned to the professional literature. The next steps involved item generation, expert evalu- ation of items, and creation of an initial item pool. We then conducted two studies that involved administration of these items to large samples of practicing psychologists. Study 1 was con- ducted to determine the factor structure of the scale, assess item performance, and provide an initial assessment of the validity of the self-care factor scores. Study 2 was conducted to provide a confirmatory evaluation of the factor structure and final scale items, as well as to compare alternative models among a second sample of psychologists.

Defining Self-Care

The growing professional literature on self-care for psycholo- gists has taken a variety of approaches to conceptualizing and discussing self-care. In one of the most comprehensive books regarding psychotherapist self-care, Norcross and Guy (2007) pre- sented a principle-based, flexible approach on how to carry out self-care. These principles, such as refocusing on the rewards of psychotherapeutic work, fostering creativity and growth, and set- ting boundaries between work and family life, are based on a mix of spirituality, mindfulness, and positive psychology values, as well as cognitive–behavioral therapy and physical wellness stan- dards (Wise et al., 2012). Others have approached self-care by focusing on general categories or domains of well-being, such as physical, spiritual, emotional, or social well-being (e.g., Baker, 2003; Carroll, Gilroy, & Murra, 1999). Still others have empha- sized specific behaviors or activities aimed at promoting well- being (e.g., Barnett et al., 2007; Norcross, 2000; Skovholt, Grier, & Hanson, 2001). In this regard, recommended self-care strategies include seeking personal therapy, taking time for interpersonal relationships, creating variety in the workday, participating in extracurricular activities, and engaging with professional organi- zations (Coster & Schwebel, 1997; Norcross, 2000; Norcross & Guy, 2007).

In the absence of a comprehensive measure of self-care, the limited research in this area has typically taken the approach of examining specific self-care behaviors or activities. Employing different terms, including well-functioning strategies (Coster & Schwebel, 1997), career-sustaining behaviors (Kramen-Kahn & Hansen, 1998; Stevanovic & Rupert, 2004), and self-care strate- gies (Goncher, Sherman, Barnett, & Haskins, 2013; Mahoney, 1997), studies have presented a list of strategies or behaviors and asked psychologists or other mental health professionals to rate the importance or frequency of participation in each. Given the emo- tional demands of psychological work, the majority of these be- haviors have tended to focus on emotional or psychological well-

being and include behaviors such as maintaining work–life balance (e.g., engage in hobbies, spend time with friends), building pro- fessional competence and resources (e.g., seek case consultation, remain active in professional development), or employing cognitive strategies for keeping perspective and coping with work demands (e.g., maintain a sense of humor, maintain self-awareness). Results of these studies have found that psychologists view many well- functioning strategies or career sustaining behaviors as important for their professional well-functioning, particularly maintaining a work life balance and employing cognitive strategies (Rupert & Kent, 2007; Rupert, Miller, Hartman, & Bryant, 2012).

Although the self-care literature is somewhat fragmented, sev- eral themes do emerge. First, self-care is multidimensional and multifaceted because it involves many areas and dimensions of personal and professional life (Godfrey et al., 2011). Furthermore, self-care is purposeful in that it contains an intentionality compo- nent, a planful decision to engage in specific activities or behaviors (Godfrey et al., 2011; Lee & Miller, 2013; Wise, Hersh, & Gibson, 2011). Self-care also is a process involving self-reflection, aware- ness, and adaptation to one’s changing needs, experiences, and values (Coster & Schwebel, 1997; Norcross, 2000; Skovholt et al., 2001). Finally, the goal of self-care is promotion of healthy func- tioning and enhancement of well-being. In the context of helping professionals, the goal of self-care is not only to maintain resil- ience in the face of stress, but also to flourish in personal and professional life (Wise et al., 2011). Integrating these themes, self-care can be defined as a multidimensional, multifaceted pro- cess of purposeful engagement in strategies that promote healthy functioning and enhance well-being.

Item Generation and Expert Feedback

Using this general definition as a guide, we sought to further define the strategies or behaviors that fall within the realm of self-care. Although several self-care frameworks have been pre- sented in the literature, we adapted the self-care conceptual frame- work offered by Lee and Miller (2013) for the purposes of cap- turing the breadth of the construct and generating a comprehensive item pool. Originally proposed for social workers, this framework emphasizes two distinct but related self-care domains: personal self-care and professional self-care. We then turned to the wider self-care literature to describe unique dimensions of personal and professional self-care and to ensure that we identified all facets of self-care for the item generation phase. The dimensions within personal self-care were identified based on the general categories of self-care behaviors discussed in the conceptual literature as important for holistic health and well-being. The five critical dimensions that comprised our personal self-care domain were physical, psychological, spiritual, social, and recreational (e.g., Baker, 2003; O’Halloran & Linton, 2000; Richards, Campenni, & Muse-Burke, 2010). In regard to professional self-care, the dimen- sions were conceptualized based on the demands inherent in the field and strategies more directly related to work that can be employed to appropriately and effectively aid in fostering well- being at the workplace. The four dimensions of our professional self-care domain were psychological, social, work–life balance, and developmental (Coster & Schwebel, 1997; Elman, Illfelder- Kaye, & Robiner, 2005; Lee & Miller, 2013; Norcross & Guy, 2007). This broad framework of personal and professional do-

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

326 DOROCIAK, RUPERT, BRYANT, AND ZAHNISER

 

 

mains, each with specific dimensions, was adopted to capture all aspects of self-care and thus served principally as a template for item generation. Using this proposed definition and framework, items were then generated based on the empirical literature, rele- vant theories, consultation with experts and the target population, examination of related instruments, and rational deduction, as recommended by Holmbeck and Devine (2009).

Following item generation, the preliminary phase next consisted of content evaluation by experts and finalization of the item pool. The purpose of the expert evaluation was to examine face and content validity of the items, modify problematic wording, and reduce the number of items prior to administration to the initial sample. Three expert, doctoral-level psychologists who had pub- lished or presented in the area of self-care and professional well- being were provided with the self-care definition and framework to use in evaluating the items. A total of 80 items were generated for expert evaluation, with an equal number of items in the personal and professional self-care domains. Within the personal and pro- fessional domains, the number of items in each specific dimension varied depending on the complexity and breadth of the dimension. For each item, experts were asked to rate both the clarity and relevance on a 7-point scale, to indicate whether or not to include the item in the final scale, and to offer suggestions or wording changes to improve the item. Items with a mean clarity or rele- vance rating below 4, items marked “no” for inclusion by the majority of experts, and items with significant conceptual redun- dancy were modified or deleted. Based on the expert ratings and comments, the 80 items were reduced to 52 items (26 personal items and 26 professional items) for administration to the devel- opment sample in Study 1. For each item, participants were asked to rate the frequency with which they engage in the stated behavior on a scale from 1 (never) to 7 (almost always).

Study 1

Method

Procedures. Following approval from a university review board via exemption, potential scale items and validation measures were sent via U.S. mail to a sample of practicing psychologists in the state of Illinois. To secure this sample, names and addresses for all licensed psychologists were obtained from the Illinois Depart- ment of Financial and Professional Regulation. From this list of approximately 4,800 psychologists, two random samples of 1,500 participants, one for Study 1 and a second for Study 2, were drawn. The “Professional Self-Care and Well-Being Survey” (preceded by a prenotification postcard and followed by a reminder postcard) was sent in March 2015 to psychologists in Sample 1.

Participants. A total of 438 psychologists returned the sur- veys for a 29.5% response rate. Sixteen surveys were not included in the analyses because the participants were no longer engaged in clinical practice or left the survey unanswered. Of the 422 partic- ipants eligible for participation, 69.9% were women. The sample was primarily White (87.2%), with the remaining participants identifying as African American (4.7%), Latino (2.1%), Asian (2.1%), Multiracial (1.9%) and “Other” (0.7%). In regard to mar- ital status, 69.2% were married/partnered, 6.8% were in a com- mitted partnership, 12.3% were single, and 7.3% were divorced. In terms of work setting, 33.6% were in solo independent practice,

19.0% were in group independent practice, 13.7% were in a hospital setting, 1.9% were in a community center, 9.7% were in an outpatient clinic, and 21.8% marked other. The mean age was 50.48 years (SD � 14.50), mean years since licensure was 16.71 years (SD � 12.39 years), and mean number of hours worked per week was 44.13 hours (SD � 14.38 hr).

Measures. The survey included the 52 self-care items plus a series of demographic questions regarding age, gender, marital status, children, racial/ethnic background, primary specialty area, years since licensure, primary and secondary work settings, hours worked in these settings, and hours in different work activities. In addition, a number of measures were included to explore the validity of the self-care factors. Given the lack of well-validated measures of self-care, concurrent validity was examined by includ- ing measures of constructs that we expected to relate to self-care (i.e., perceived stress, life satisfaction, and burnout). As a result, the following validity measures were included in the survey.

The Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermel- stein, 1983) is a 10-item measure designed to assess perceived stress, conceptualized as the degree to which situations in an individual’s life are perceived as uncontrollable and overwhelm- ing. The PSS has significant correlations with life satisfaction, alcohol use, and other health-related outcomes (Cohen & William- son, 1987). Cronbach’s alpha has been reported as r � .91 (Cohen & Janicki-Deverts, 2012), with good internal consistency in the current study (� � .86).

The Satisfaction With Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985) is a five-item measure designed to assess the life satisfaction component of subjective well-being. Extensive evidence has found that an individual’s satisfaction with life score positively correlates with a range of life outcomes, including mental and physical health (Pavot & Diener, 2008). Cronbach’s alpha has ranged from r � .79 to .89 (Pavot & Diener, 2008). This scale yielded good internal consistency in the current study (� � .87).

The Maslach Burnout Inventory—Human Service Survey (MBI-HSS; Maslach & Jackson, 1996) is a 22-item measure, which assesses the three components of burnout: emotional ex- haustion (EE), depersonalization (DP), and a reduced sense of personal accomplishment (PA). The MBI-HSS is the most widely employed measure of burnout among human service professionals and has consistently reported sound psychometric properties (Maslach & Jackson, 1996). In the current study, the internal consistency for the emotional exhaustion subscale was good (� � .89), for the depersonalization subscale was adequate (� � .71), and for the personal accomplishment subscale was adequate (� � .73).

Results

In line with best practices (Worthington & Whittaker, 2006), the planned data analyses included the following: evaluation of items and missing data, assessment of the factorability of the data, selection of appropriate extraction and rotation methods, identifi- cation of the factor structure, retention and deletion of items, and optimization of scale length.

Item analyses and missing data. Item analyses examined each item’s mean, standard deviation (SD), skewness, and kurtosis values. Absolute skewness values above 3 or kurtosis values above

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

327PROFESSIONAL SELF-CARE SCALE

 

 

8 were considered “extreme” (Kline, 2011). Based on these crite- ria, two items were omitted for a total of 50 remaining items. No participants were included if they were missing more than 20% of the self-care items (Parent, 2013; Peng, Harwell, Liou, & Ehman, 2006). On the self-care scale, 398 participants (94.3%) had no missing data, 18 (4.3%) had one missing item, 4 (0.9%) had two missing items, and 4 (0.9%) had three missing items. To establish whether there was a meaningful pattern to the small amount of missing data, the missing completely at random (MCAR) test was used (Little, 1988). Little’s MCAR test was nonsignificant, �2(821, 422) � 859.96, p � .168, indicating that the self-care data were missing completely at random. The expectation-maximization (EM) algorithm was then used to impute missing data. This procedure was chosen over multiple imputation because it is more efficient, always produces the same result, and is more definitive (Allison, 2012; Schafer & Graham, 2002). Similar item analyses and missing data procedures were followed for all measures included in the study.

Exploratory factor analysis (EFA). As described previously, a broad conceptual framework comprised of personal and profes- sional domains, each with several dimensions, was used to gener- ate an initial pool of items that would capture the full range of self-care activities. It is important to note that this framework was used only for the purpose of item generation—that is, in pursuit of comprehensiveness across a wide range of aspects of self-care— and did not represent a prediction regarding the structure of the resulting measure of self-care. Given the variety of approaches that researchers have used to conceptualize self-care, we had no a priori hypothesis about the specific factor structure that would emerge. In measure development, principal axis factoring (PAF) is the pre- ferred analytic method when the goal is to establish the underlying factor structure and was therefore selected as the appropriate EFA method (DeVellis, 2012). An oblique, promax rotation was per- formed to allow the components to intercorrelate and to allow items to load on multiple scales (Thompson, 2004). Bartlett’s test confirmed that EFA was appropriate for the sample, �2(1225) � 10,818.30, p � .001, and a Keiser-Meyer-Olkin (KMO) test indi- cated that the data were likely to yield reliable factors (KMO � .89).

Factor structure. Following assessment of the factorability of the data and selection of the appropriate extraction and rotation method, PAF was performed and the optimal factor structure identified. Factor retention was based on the scree plot, eigenvalue-greater-than-one rule (e.g., Netemeyer, Bearden, & Sharma, 2003; Nunnally & Bernstein, 1994), parallel analysis, and interpretability (e.g., Bryant & Yarnold, 1995; Tabachnick & Fidell, 2007). Factors were also required to have more than two items, because it has been argued that at least three items are needed to identify common variance (e.g., Comrey, 1988; Yong & Pearce, 2013).

The initial promax rotation yielded 12 factors with eigenvalues greater than one. The eigenvalues for these 12 factors were as follows: 12.92 (25.85% of variance accounted for), 3.94 (7.88%), 2.34 (4.68%), 2.05 (4.11%), 2.00 (4.00%), 1.64 (3.27%), 1.59 (3.17%), 1.37 (2.74%), 1.23 (2.47%), 1.14 (2.28%), 1.07 (2.14%), and 1.04 (2.08%). Parallel analysis was also conducted to establish how many factors should be extracted. Because employing PAF for parallel analysis may lead to the overextraction of factors (Steger, 2006), principal components analysis (PCA) was con- ducted. The PCA parallel analysis with 1,000 simulated random

data sets indicated that the first seven factors had eigenvalues that exceeded chance values (O’Connor, 2000). When examining the interpretability of the factors in light of the eigenvalues, PCA parallel analyses, and scree plot, both five and six factors were plausible. However, the five-factor solution met multiple addi- tional criteria such that each factor contained a minimum of three items, exhibited adequate internal consistency, and was interpre- table and consistent with our initial conceptualization of self-care (Tabachnick & Fidell, 2007). Therefore, we identified the five- factor solution to best represent the data.

Item retention and deletion. After the five-factor solution was identified, follow-up PAFs used the interpretable factors from the previous PAF to force the extraction of the appropriate number of factors and to finalize the items that loaded onto each factor. Items were selected based on the factor pattern matrix using the following criteria: (a) items had a minimum factor loading of .32, which signifies 10% overlapping variance of the item with the other items in the factor (Tabachnick & Fidell, 2007); (b) no items had cross-loadings such that the difference between the item’s highest and second-highest factor loadings was less than .15 (Wor- thington & Whittaker, 2006); and (c) no items had absolute load- ings higher than .32 on two or more factors (Tabachnick & Fidell, 2007; Worthington & Whittaker, 2006). In the subsequent PAFs, factors were still required to have at least three items. Factor reliabilities were checked after dropping items to ensure that the item elimination did not greatly affect the reliability of the factor.

Following the identification of the five-factor solution, the sec- ond PAF forced the remaining 34 items onto a five-factor struc- ture. Eight items were eliminated based on the previously noted item criteria (i.e., four items had cross-loadings within .15 of each other; four items had absolute loadings higher than .32 on two or more factors). A follow-up PAF forced the remaining 26 items onto five factors, eliminating three additional items (i.e., two items had cross-loadings within .15 of each other; one item had a loading less than .32). When imposing a five-factor exploratory solution on the remaining 23 items, all items met criteria for acceptable fit.

Optimization of scale length. Following the refinement of items, optimization of scale length was conducted to assess the trade-off between length and reliability (Worthington & Whittaker, 2006). The items on a given subscale were compared to identify those that contributed the least to internal consistency, had the lowest factor loadings, had the highest cross-loadings, and had low conceptual consistency with other items on the factor. Two items were subsequently dropped, because their elimination resulted in an increase in the internal consistency of their respective subscales.

A final PAF forced the remaining 21 items to load onto five factors. The resulting solution accounted for 61.5% of the variance in the self-care items and consisted of the following factors: a five-item Professional Support Subscale (� � .83), a five-item Professional Development Subscale (� � .80), a four-item Life Balance Subscale (� � .81), a four-item Cognitive Awareness Subscale (� � .72), and a three-item Daily Balance Subscale (� � .70). The pattern matrix of the final five factors, items, and loadings is presented in Table 1. Factor subscales scores were created based on totaling the items comprising each factor. The correlations among items in the factor as well as the correlations between each item and the factor total were calculated as a check on item performance. Correlations were also run among all factors

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

328 DOROCIAK, RUPERT, BRYANT, AND ZAHNISER

 

 

to assess their interrelatedness. These correlation results are pre- sented in Table 2.

Validity. The initial validity of the PSCS Subscale scores was assessed by examining the significance of the correlation coeffi- cients relating the factor subscale scores to the validity measures: PSS, SWLS, and the MBI-HSS (EE, DP, and PA Subscales). Both the theoretical and empirical literature emphasize that professional and personal domains interact and spill over (e.g., Duncan & Goddard, 1993; Grzywacz & Marks, 2000; Stevanovic & Rupert, 2004). Thus, while some self-care factors focus more on work- place strategies (i.e., Professional Support and Professional Devel- opment) and other factors represent behaviors that occur mainly in the personal life domain (i.e., Life Balance), given the research on the interaction between professional and personal life, it was

hypothesized that all five of the self-care factors would relate to well-being outcomes across both personal and professional life domains. Specifically, higher scores on each of the self-care fac- tors were expected to relate to less emotional exhaustion, less depersonalization of clients, greater sense of personal accomplish- ment, lower perceived stress, and greater satisfaction with life.

Consistent with expectations, all five self-care factor scores had significant correlations with each validity measure in the expected direction (Table 3). Of particular note were the relatively stronger correlations of the Life Balance, Cognitive Awareness, and Daily Balance factors, with the well-being outcomes of emotional ex- haustion and perceived stress. In contrast, although statistically significant, the correlations of Professional Support with the well- being outcomes were relatively weaker (ranging from .11 to .26 in

Table 1 Final Item Factor Loadings

Item

Sample 1 EFA Sample 2

CFA1 2 3 4 5

Professional Support 48. I cultivate professional relationships with my colleagues. .80 .09 �.08 .12 �.03 .75 32. I avoid workplace isolation. .77 �.02 �.03 �.18 .13 .63 30. I share work-related stressors with trusted colleagues. .63 �.10 .12 .20 �.18 .72 26. I share positive work experiences with colleagues. .63 .01 �.03 �.03 .07 .76 14. I maintain a professional support system. .56 .10 .08 .12 �.07 .81

Professional Development 6. I participate in activities that promote my professional development. �.06 .77 �.07 .12 .08 .71

12. I connect with organizations in my professional community that are important to me. .05 .76 .08 �.14 �.10 .68 8. I take part in work-related social and community events. .15 .62 .09 �.24 .04 .59

42. I find ways to stay current in professional knowledge. �.09 .58 �.05 .31 �.04 .66 44. I maximize time in professional activities I enjoy. .002 .43 .10 .26 .01 .69

Life Balance 11. I spend time with people whose company I enjoy. �.05 .10 .81 �.06 �.07 .62 17. I spend time with family or friends. �.01 �.02 .73 �.12 .10 .64 43. I seek out activities or people that are comforting to me. .18 �.03 .66 �.11 .06 .78

1. I find ways to foster a sense of social connection and belonging in my life. �.07 .20 .66 .20 �.01 .77 Cognitive Strategies

45. I try to be aware of my feelings and needs. �.03 �.13 .12 .77 .04 .69 2. I monitor my feelings and reactions to clients. �.02 .07 �.20 .62 �.03 .48

34. I am mindful of triggers that increase professional stress. .12 �.06 �.003 .54 .08 .60 40. I make a proactive effort to manage the challenges of my professional work. .03 .10 .14 .37 .14 .67

Daily Balance 50. I take breaks throughout the workday. .10 .12 �.11 �.02 .71 .45 49. I take some time for relaxation each day. �.003 �.08 .04 .11 .69 .80

4. I avoid overcommitment to work responsibilities. �.08 �.07 .18 .01 .50 .54 Percentage of variance 32.6 9.9 7.0 6.1 5.9 Eigenvalue 6.84 2.08 1.47 1.29 1.23 Cronbach’s alpha .83 .80 .81 .72 .70

Note. EFA � exploratory factor analysis; CFA � confirmatory factor analysis.

Table 2 Correlations Among Professional Self-Care Factors

Professional self-care factors

Professional Support

Professional Development

Life Balance

Cognitive Awareness

Daily Balance

Professional Support — Professional Development .55�� — Life Balance .50�� .44�� — Cognitive Awareness .42�� .40�� .48�� — Daily Balance .22�� .24�� .36�� .41�� —

� p � .05. �� p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

329PROFESSIONAL SELF-CARE SCALE

 

 

magnitude). Overall, the correlation analyses provided evidence that the first five factors consistently related to well-functioning outcomes in expected directions. The result was a five-factor, 21-item Professional Self-Care Scale (PSCS), which was then tested in Study 2.

Study 2

Method

Procedure and measures. Following approval from a univer- sity review board via exemption, the “Professional Well-Being Survey,” including the newly constructed PSCS, was sent via U.S. mail to the second sample of 1,500 Illinois licensed psychologists in July 2015. As previously described, this scale consisted of 21 items, which reflected five domains of self-care: Professional Support, Professional Development, Life Balance, Cognitive Awareness, and Daily Balance.

Participants. A total of 403 psychologists returned the sur- veys for a 27.5% response rate. Twenty-nine participants were excluded because they left the surveys blank or were no longer engaged in clinical practice. Of the final sample of 374 partici- pants, 70.9% were women. The sample was primarily White (86.9%), with the remaining participants identifying as Asian (8.7%), African American (2.1%), Latino (1.9%), and “Other” (3.5%). In regard to marital status, 76.7% were married/partnered, 5.1% were in a committed partnership, 9.4% were single, and 7.2% were divorced. In terms of work setting, 36.6% were in solo independent practice, 26.7% were in group independent practice, 11.0% were in a hospital setting, 1.9% were in a community center, 6.1% were in an outpatient clinic, and 17.6% marked other. The mean age was 51.55 years (SD � 13.21), mean years since licensure was 18.13 years (SD � 12.04 years), and the mean number of hours worked per week was 36.20 hours (SD � 12.74 hr).

Results

The five-factor, 21-item structure identified in Study 1 was evaluated in this second sample and compared with three alterna- tive measurement models: a unidimensional model, a higher-order factor model with one second-order factor and five first-order factors, and a bifactor model. First, a unidimensional or single- factor model examined whether all self-care strategies loaded onto

a general self-care factor. Additionally, as the first eigenvalue in the Study 1 EFA was three times that of the second eigenvalue, it was possible that the item pool was dominated by a strong general factor (Cho et al., 2015). To examine whether the structure in- cluded a general self-care factor comprised of several highly related dimensions, both a higher-order model and a bifactor model were tested. A higher-order model with one second-order factor and five first-order factors was included to determine if a single higher-order factor of self-care would explain the intercor- relations between factors in the five-factor model. A bifactor model was also examined as a potential alternative representation of the data. Rather than suggesting that the effects of a general factor are mediated through a specific factor, as is the case for the higher-order model, a bifactor model separates the reliable vari- ance in each item-level response into an overarching latent con- struct (self-care) and into one of several specific content domains (e.g., one of the five self-care factors; e.g., Reise, Morizot, & Hays, 2007).

Confirmatory factor analysis (CFA). Confirmatory factor analysis was conducted using LISREL 8.8 (Jöreskog & Sörbom, 1993) and robust maximum-likelihood (RML) estimation to cor- rect for distortion in fit indices and standard errors because of multivariate nonnormality. To examine the relative fits of the four models to the data, the Satorra-Bentler scaled maximum-likelihood chi-square (SB-ML �2; Bryant & Satorra, 2012; Satorra & Bentler, 1994) was calculated, as well as seven indices of model fit. The statistical significance of the model’s overall chi-square value was not employed as the primary index of model fit because this statistical test of perfect fit is “too strong to be realistic” (Hu & Bentler, 1998, p. 425) and is not typically used to assess model fit in applied research (Brown, 2006). As recommended by Hu and Bentler (1998), model fit was assessed using two indices of abso- lute fit (root mean square error of approximation [RMSEA] and standardized root-mean-square residual [SRMR]) and two indi- ces of relative fit (comparative fit index [CFI] and non-normed fit index [NNFI]). In assessing goodness-of-fit, RMSEA � .08 (Browne & Cudeck, 1993), SRMR � .08 (Hu & Bentler, 1998), CFI � .90 and NNFI � .90 (Marsh, Hau, & Wen, 2004) were considered acceptable. In addition, the Akaike Information Cri- terion (AIC), AIC corrected for sample size (CAIC), and the Bayesian Information Criterion (BIC; a parsimony-adjusted predictive index of fit) were used to compare model fits, with

Table 3 Correlations of Self-Care Factors With Validity Measures

Professional self-care factors

EE DP PA PSS SWLS

� � .89 � � .71 � � .73 � � .86 � � .87 N � 420 N � 420 N � 417 N � 419 N � 421

Factor 1 (Professional Support) �.12� �.11� .22�� �.17�� .26��

Factor 2 (Professional Development) �.22�� �.25�� .27�� �.24�� .29��

Factor 3 (Life Balance) �.34�� �.24�� .35�� �.42�� .48��

Factor 4 (Cognitive Awareness) �.34�� �.34�� .44�� �.36�� .32��

Factor 5 (Daily Balance) �.45�� �.21�� .25�� �.42�� .31��

Note. EE � emotional exhaustion; DP � depersonalization; PA � personal accomplishment; PSS � Perceived Stress Scale; SWLS � Satisfaction With Life Scale. � p � .05. �� p � .01.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

330 DOROCIAK, RUPERT, BRYANT, AND ZAHNISER

 

 

smaller values of AIC, CAIC, and BIC denoting better model fit (Burnham & Anderson, 2004).

Table 4 reports the fit of the four competing models. The five-factor model met all criteria for acceptable model fit, SB-ML �2; (374) � 416.31; RMSEA � .061; SRMR � .058; CFI � .97; NNFI � .96 (see Table 1 for factor loadings). In contrast, the three other measurement models did not meet criteria for acceptable fit Table 4). Scaled difference chi-square testing (Bryant & Satorra, 2012) indicated that although the bifactor model fit the data of Sample 2 better than the second-order model, �SB-ML �2 � 30.87, �df � 15, p � .001, the five-factor model fit the data significantly better than did the bifactor model, �SB-ML �2 � 54.39, �df � 10, p � .001. Furthermore, the AIC, CAIC, and BIC estimates supported a preference for the oblique five-factor model. Factor reliabilities for all five subscales were acceptable (�s � .85, .79, .80, .71 and .69 for Professional Support, Professional Devel- opment, Life Balance, Cognitive Awareness, and Daily Balance, respectively). In summary, based on model fit indices, the data suggest that the PSCS best conforms to an oblique five-factor structure.

General Discussion

The current study reports the development of a 21-item self-care scale for professional psychologists, the PSCS. An initial list of 52 self-care items was generated and refined via PAF, ultimately yielding five factors on which a total of 21 items met all item-level criteria. The final five factors were identified as Professional Support (five items), Professional Development (five items), Life Balance (four items), Cognitive Awareness (four items), and Daily Balance (three items). Cross-sectional analyses from Study 1 also provided preliminary evidence regarding the validity of the PSCS factor subscale scores. A second study employed CFA with a second sample of clinical psychologists and found that the five- factor solution provided the best fit to the data as compared with three alternative measurement models.

The factors that emerged are consistent with many of the themes emphasized in the professional self-care literature and involve behaviors or strategies that represent key aspects of personal and professional life. The Professional Support factor emphasizes the importance of supportive colleagues and includes strategies such as avoiding isolation, cultivating relationships with colleagues, and sharing both rewarding and stressful work experiences. Profes- sional Development demonstrates the importance of engaging in

work activities that are enjoyable, participating in professional organizations and events, and staying current in professional knowledge. The Life Balance factor highlights the importance of having more than a professional identity but also a personal identity. This factor underscores the importance of social support outside of the work place and emphasizes strategies that serve to build a balance between work and nonwork life. Cognitive Aware- ness emphasizes the importance of psychological self-care and involves monitoring workplace stress and emotions, having a proactive approach to managing challenges, and maintaining awareness of feelings and needs. The fifth factor, Daily Balance, in contrast to the Life Balance factor, encompasses smaller-scale, microfocused strategies that can be incorporated throughout the workday to manage demands while maintaining awareness and replenishing resources.

Of note, the final PSCS does not include a physical self-care factor. Although our framework for item generation included a physical self-care dimension and the initial item pool included items focused on diet, exercise, sleep and physical health, these items did not group together to form a meaningful physical self- care factor and were thus eliminated. However, this should not be interpreted to mean that physical self-care is not important. In fact, the importance of taking care of oneself physically—eating well, exercising, getting sufficient sleep—is consistently emphasized in the self-care literature. Physical self-care behaviors, however, clearly lie in the personal life domain and may vary markedly across individuals based on factors such as health status, age, and personal preference. Thus, the complexity of physical self-care may not be adequately assessed through a small number of items and may be more appropriately assessed through more extensive, focused measures of health and lifestyles behaviors.

The five factors of the PSCS and, most importantly, the specific items on each factor, are consistent with the conceptualization of self-care as a proactive, ongoing process. In fact, items represent- ing strategies that are more likely to occur as a reactive response to stressors or challenges (e.g., seeking consultation or supervision when professionally challenged, seeking guidance or counseling when necessary, reducing workload in the face of professional stress) did not load significantly on any of the five factors and were thus not included in the final scale. Although these types of strategies, which are often discussed in the professional self-care literature (e.g., Norcross & Guy, 2007), may be important at times of stress, they are more reactive in nature and thus less likely to be

Table 4 Confirmatory Factor Analysis Assessment of Model Fit

Model SB-ML �2 df RMSEA� SRMR� CFI�� NNFI�� AIC CAIC BIC

One factor, unidimensional model 1010.49 189 .137 .106 .82 .80 1585.66 1597.13 1515.74 Five correlated factors 416.31 179 .061 .058 .97 .96 531.80 549.69 799.53 One second-order factor with five first-order factors 515.12 184 .073 .084 .95 .94 643.16 657.64 886.45 Orthogonal bifactor model with one general factor and five

domain-specific factors 485.72 169 .073 .087 .95 .94 625.55 651.56 934.57

Note. SB-ML �2 � Satorra-Bentler scaled maximum likelihood chi-square value; RMSEA � root mean square error of approximation; SRMR � standardized root mean residual; CFI � comparative fit index; NNFI � nonnormed fit index; AIC � Akaike Information Criterion (smaller values reflect better fit); CAIC � AIC corrected for sample size (smaller values reflect better fit); BIC � Bayesian Information Criterion (smaller values reflect better fit). � RMSEA and SRMR value of �.08 is acceptable. �� CFI and NNFI value of �.90 is acceptable.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

331PROFESSIONAL SELF-CARE SCALE

 

 

employed on a routine, ongoing basis. The PSCS adopts a preven- tative perspective on self-care, with items reflecting strategies or behaviors that may be integrated into one’s professional and per- sonal life on a more ongoing and proactive basis to promote well-functioning.

Based on the conceptual literature, our measure development process began with a general definition that viewed self-care as a multidimensional, multifaceted process. Consistent with this, the present research found that that the self-care items clustered into meaningful factors representing different facets of life and types of behaviors. Although these factors related significantly to each other, the higher-order and bifactor analyses determined that an overarching factor did not emerge based on the combination of the five self-care factors. Thus, the findings suggest that self-care is best conceptualized and assessed as a multidimensional rather than a unitary construct. In employing the scale, it is not appropriate to compute a total self-care score, as individuals may vary in terms of their needs, preferences, and engagement in the various domains of self-care. Rather, individual subscale (factor) scores are more meaningful in describing and understanding self-care.

Self-Care and Its Relationship to Personal and Professional Well-Being Outcomes

Study 1 established that the five self-care factors were related to a variety of personal and professional well-being outcomes. These findings suggest that the scale assesses behaviors reflective of meaningful dimensions that have the potential to predict important life outcomes. The initial correlational analyses conducted in Study 1 indicated that the first five factors were related as expected to the well-being outcomes. Although the varying strengths of the correlational coefficients provide rough evidence of the differen- tial value of specific self-care factors, future research is needed that proposes specific differential hypotheses in order to examine the unique properties of each factor. At present, examination of the correlation coefficients between the self-care subscales and the outcomes suggests that Life Balance, Cognitive Awareness, and Daily Balance may be especially important in lowering stress and burnout. It is interesting that although the importance of support is consistently emphasized in the self-care literature (e.g., Norcross & Guy, 2007), Professional Support was not as strongly related to well-being outcomes. This is, however, consistent with the burnout literature, which has yielded inconsistent findings regarding the relationship of workplace support and indices of burnout (e.g., Rupert, Miller, & Dorociak, 2015). It may be that the importance of professional support, as well as other self-care strategies, varies depending on characteristics of the individual and work environ- ment. Thus, research is needed to investigate the relationship of self-care subscales to well-being outcomes across different mental health professionals and work environments.

Limitations

Before concluding, three limitations must be considered. First, our assessment of the validity of the PSCS subscale scores is limited. Because validity of an instrument’s score is often assessed by examining the relationship between the new measure and another measure of the same underlying construct, the lack of other empirically based measures of professional self-care posed signif-

icant challenges. Preliminary evidence for the validity of the scale was thus established through examination of the correlation coef- ficients between the PSCS and other theoretically related con- structs. Because self-care is promoted as a means of reducing stress and enhancing well-being, we expected that self-care would correlate with positive well-being outcomes, which was the case. However, a more comprehensive assessment of validity is needed that proposes specific differential hypotheses and more compre- hensively explores the validity of the subscale scores. Additional research is needed to further validate the PSCS factor subscale scores, understand the factors’ differential effects in relationship to important personal and professional outcomes, and establish the unique, distinctive properties of the individual components of self-care.

Second, our validity data are limited by the reliance on cross- sectional self-report data. Correlations between self-care factors and outcomes may have been inflated because of the variance shared by the common self-report measurement method. Although the variability in strength of correlations indicates that shared or common method variance most likely did not completely account for significant relationships, future research may benefit from a multimethod approach to the assessment of self-care and related constructs. In addition, examining how the scales relate to out- comes over time or how they relate to other important non-self- report variables would increase confidence in the results.

Finally, we should note that both samples used in this measure development project were comprised of licensed psychologists. Furthermore, both samples were primarily White, female, and highly educated (all participants had a doctoral level degree). Thus, caution should be taken in generalizing the findings and using this measure with more demographically diverse groups as well as with other groups of mental health professionals. Although the PSCS factors reflect many key themes in the self-care litera- tures for mental health professionals in general, the strategies or behaviors described in specific items may vary in importance and meaning for individuals from different cultural or ethnic back- grounds and perhaps from different educational or professional backgrounds. It is also possible that our items may not have captured or adequately represented some strategies that are impor- tant for individuals from groups that are underrepresented within our samples. Further research is thus needed to evaluate items, replicate the factor structure, and explore the validity and gener- alizability of the PSCS factors with broader, more diverse partic- ipant pools and with other populations of mental health profes- sionals at different levels of training and experience. Additional work of this nature will help to demonstrate the utility of this measure, not just for licensed clinical psychologists, but for a broad range of mental health care providers. Given the growing body of literature concerning the need for self-care among mental health professionals of all levels, this is one area in which contin- ued study of the Professional Self-Care Scale can have its greatest impacts.

Conclusions

The study of professional self-care relies on an adequate assess- ment tool. The development and initial validation of the Profes- sional Self-Care Scale is an important first step in allowing pro- fessionals to assess self-care and to promote well-being in and

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

332 DOROCIAK, RUPERT, BRYANT, AND ZAHNISER

 

 

outside the work environment. The results of both Studies 1 and 2 provide strong support for the five factor structure of the PSCS and the internal consistency of these factors. In addition, the process of constructing items plus the evaluation by experts provides support for the face validity and content validity of the PSCS items, with Study 1 offering preliminary evidence that self-care is related to personal and professional well-being outcomes. Although further work is needed to establish the validity of the factor subscale scores and to replicate the factor structure of this scale with more diverse populations of professionals, these initial results are prom- ising. It is hoped that the development of this scale will foster research aimed at understanding critical components of self-care and examining predictors and outcomes of self-care. Ultimately, such knowledge will be helpful in offering specific self-care rec- ommendations and developing ways of promoting self-care for mental health professionals.

References

Allison, P. D. (2012, April). Handling missing data by maximum likeli- hood. Paper presented at the SAS Global Forum, Orlando, Florida. Retrieved from http://www.statisticalhorizons.com/wp-content/uploads/ MissingDataByML.pdf

American Counseling Association. (2014). ACA Code of Ethics. Alexan- dria, VA: Author.

American Psychological Association. (2002). Ethical principles of psy- chologists and code of conduct. American Psychologist, 57, 1060–1073. http://dx.doi.org/10.1037/0003-066X.57.12.1060

Baker, E. K. (2003). Caring for ourselves: The therapist’s guide to per- sonal and professional well-being. Washington, DC: American Psycho- logical Association. http://dx.doi.org/10.1037/10482-000

Barnett, J. E., Baker, E. K., Elman, N. S., & Schoener, G. R. (2007). In pursuit of wellness: The self-care imperative. Professional Psychology: Research and Practice, 38, 603–612. http://dx.doi.org/10.1037/0735- 7028.38.6.603

Barnett, J. E., Johnston, L. C., & Hillard, D. (2006). Psychotherapist wellness as an ethical imperative. In L. VandeCreek & J. B. Allen (Eds.), Innovations in clinical practice: Focus on health and wellness (pp. 257–271). Sarasota, FL: Professional Resources Press.

Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York, NY: Guilford Press.

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage.

Bryant, F. B., & Satorra, A. (2012). Principles and practice of scaled difference chi-square testing. Structural Equation Modeling, 19, 372– 398. http://dx.doi.org/10.1080/10705511.2012.687671

Bryant, F. B., & Yarnold, P. R. (1995). Principal-components analysis and exploratory and confirmatory factor analysis. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 99–136). Washington, DC: American Psychological Association.

Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference under- standing AIC and BIC in model selection. Sociological Methods & Research, 33, 261–304. http://dx.doi.org/10.1177/0049124104268644

Canada Psychological Association. (2000). Canadian code of ethics for psychologists. Ottawa, Canada: Author.

Carroll, L., Gilroy, P. J., & Murra, J. (1999). The moral imperative: Self-care for women psychotherapists. Women & Therapy, 22, 133–143. http://dx.doi.org/10.1300/J015v22n02_10

Cho, S. J., Wilmer, J., Herzmann, G., McGugin, R. W., Fiset, D., Van Gulick, A. E., . . . Gauthier, I. (2015). Item response theory analyses of the Cambridge Face Memory Test (CFMT). Psychological Assessment, 27, 552–566. http://dx.doi.org/10.1037/pas0000068

Cohen, S., & Janicki-Deverts, D. (2012). Who’s stressed? Distributions of psychological stress in the United States in probability samples from 1983, 2006, and 2009. Journal of Applied Social Psychology, 42, 1320– 1334. http://dx.doi.org/10.1111/j.1559-1816.2012.00900.x

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. http://dx.doi.org/10.2307/2136404

Cohen, S., & Williamson, G. M. (1987). Perceived stress in a probability sample of the United States. In S. Spacapan & S. Oskamp (Eds.), Psychology and health (pp. 31–67). Newbury Park, CA: Sage.

Comrey, A. L. (1988). Factor-analytic methods of scale development in personality and clinical psychology. Journal of Consulting and Clinical Psychology, 56, 754 –761. http://dx.doi.org/10.1037/0022-006X.56.5 .754

Coster, J. S., & Schwebel, M. (1997). Well-functioning in professional psychologists. Professional Psychology: Research and Practice, 28, 5–13. http://dx.doi.org/10.1037/0735-7028.28.1.5

DeVellis, R. F. (2012). Scale development: Theory and applications (3rd ed.). Thousand Oaks, CA: Sage.

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71– 75. http://dx.doi.org/10.1207/s15327752jpa4901_13

Duncan, S. F., & Goddard, W. H. (1993). Stressors and enhancers in the marital/family life of family professionals and their spouses. Family Relations: An Interdisciplinary Journal of Applied Family Studies, 42, 434–441. http://dx.doi.org/10.2307/585345

Elman, N. S., Illfelder-Kaye, J., & Robiner, W. N. (2005). Professional development: Training for professionalism as a foundation for compe- tent practice in psychology. Professional Psychology: Research and Practice, 36, 367–375. http://dx.doi.org/10.1037/0735-7028.36.4.367

Godfrey, C. M., Harrison, M. B., Lysaght, R., Lamb, M., Graham, I. D., & Oakley, P. (2011). Care of self-care by other-care of other: The meaning of self-care from research, practice, policy and industry perspectives. International Journal of Evidence-Based Healthcare, 9, 3–24. http://dx .doi.org/10.1111/j.1744-1609.2010.00196.x

Goncher, I. D., Sherman, M. F., Barnett, J. E., & Haskins, D. (2013). Programmatic perceptions of self-care emphasis and quality of life among graduate trainees in clinical psychology: The mediational role of self-care utilization. Training and Education in Professional Psychol- ogy, 7, 53–60. http://dx.doi.org/10.1037/a0031501

Grzywacz, J. G., & Marks, N. F. (2000). Reconceptualizing the work– family interface: An ecological perspective on the correlates of positive and negative spillover between work and family. Journal of Occupa- tional Health Psychology, 5, 111–126. http://dx.doi.org/10.1037/1076- 8998.5.1.111

Holmbeck, G. N., & Devine, K. A. (2009). Editorial: An author’s checklist for measure development and validation manuscripts. Journal of Pedi- atric Psychology, 34, 691–696. http://dx.doi.org/10.1093/jpepsy/jsp046

Hu, L. T., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3, 424–453. http://dx.doi.org/10.1037/1082- 989X.3.4.424

Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Hillsdale, NJ: Erlbaum.

Kline, R. B. (2011). Principles and practice of structural equation mod- eling (3rd ed.). New York, NY: Guilford Press.

Kramen-Kahn, B., & Hansen, N. D. (1998). Rafting the rapids: Occupa- tional hazards, rewards, and coping strategies of psychotherapists. Pro- fessional Psychology: Research and Practice, 29, 130–134. http://dx.doi .org/10.1037/0735-7028.29.2.130

Lee, J. J., & Miller, S. E. (2013). A self-care framework for social workers: Building a strong foundation for practice. Families in Society, 94, 96–103. http://dx.doi.org/10.1606/1044-3894.4289

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

333PROFESSIONAL SELF-CARE SCALE

 

 

Little, R. J. (1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Associa- tion, 83, 1198 –1202. http://dx.doi.org/10.1080/01621459.1988 .10478722

Mahoney, M. J. (1997). Psychotherapists’ personal problems and self-care patterns. Professional Psychology: Research and Practice, 28, 14–16. http://dx.doi.org/10.1037/0735-7028.28.1.14

Marsh, H. W., Hau, K. T., & Wen, Z. (2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler’s (1999) findings. Structural Equation Modeling, 11, 320–341. http://dx.doi.org/ 10.1207/s15328007sem1103_2

Maslach, C., & Jackson, S. E. (1996). Maslach Burnout Inventory–Human Services Survey (MBI-HSS). In C. Maslach, S. E. Jackson, & M. P. Leiter (Eds.), MBI manual (3rd ed.). Mountain View, CA: CCP, Inc.

Netemeyer, R., Bearden, W., & Sharma, S. (2003). Scaling procedures: Issues and applications. Thousand Oaks, CA: Sage.

Norcross, J. C. (2000). Psychotherapist self-care: Practitioner-tested, research-informed strategies. Professional Psychology: Research and Practice, 31, 710–713. http://dx.doi.org/10.1037/0735-7028.31.6.710

Norcross, J. C., & Guy, J. D. (2007). Leaving it at the office: A guide to psychotherapist self-care. New York, NY: Guilford Press.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill.

O’Connor, B. P. (2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instruments & Computers, 32, 396–402. http://dx.doi.org/10.3758/BF03200807

O’Halloran, T. M., & Linton, J. M. (2000). Stress on the job: Self-care resources for counselors. Journal of Mental Health Counseling, 22, 354–364.

Parent, M. C. (2013). Handling item-level missing data: Simpler is just as good. The Counseling Psychologist, 41, 568–600. http://dx.doi.org/10 .1177/0011000012445176

Pavot, W., & Diener, E. (2008). The satisfaction with life scale and the emerging construct of life satisfaction. The Journal of Positive Psychol- ogy, 3, 137–152. http://dx.doi.org/10.1080/17439760701756946

Peng, C.-Y. J., Harwell, M., Liou, S.-M., & Ehman, L. H. (2006). Ad- vances in missing data methods and implications for educational re- search. In S. Sawilowsky (Ed.), Real data analysis (pp. 31–78). Green- wich, CT: Information Age.

Reise, S. P., Morizot, J., & Hays, R. D. (2007). The role of the bifactor model in resolving dimensionality issues in health outcomes measures. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation, 16(Suppl. 1), 19–31. http://dx.doi.org/10.1007/s11136-007-9183-7

Richards, K. C., Campenni, C., & Muse-Burke, J. L. (2010). Self-care and well-being in mental health professionals: The mediating effects of self-awareness and mindfulness. Journal of Mental Health Counseling, 32, 247–264. http://dx.doi.org/10.17744/mehc.32.3.0n31v88304423806

Rupert, P. A., & Kent, J. S. (2007). Gender and work setting differences in career-sustaining behaviors and burnout among professional psycholo-

gists. Professional Psychology: Research and Practice, 38, 88–96.

"Is this question part of your assignment? We can help"

ORDER NOW