Psychometric properties of the Slovak version of the professional quality of life scale : Preliminary results

The model of professional quality of life distinguishes two aspects of working in helping professions: positive (compassion satisfaction) and negative (compassion fatigue consisting of burnout and secondary traumatic stress), which represent possible effects of helping on the lives of the helpers. Students of social work in external study form who worked in the field of social care and people working in helping professions (counsellors and social workers) participated in the research and completed Professional quality of life scale, Life orientation test—revised, Emotional habitual subjective well-being scales, Rosenberg self-esteem scale, State-trait anxiety inventory and Beck depression inventory. It was hypothesised that higher level of compassion satisfaction would be linked with optimism, self-esteem, higher frequency of experiencing positive emotions (joy, happiness, pleasure and energy), lower frequency of experiencing negative emotions (anger, fear, sadness, shame, guilt and pain) and low levels of anxiety and depression.


Introduction
Working in helping professions might have both positive and negative impact on the helping person.International research on positive and negative effects of helping on the helper is represented by the concept of the professional quality of life consisting of compassion satisfaction and compassion fatigue (burnout and secondary traumatic stress) (Figley, 1995;2002;Stamm, 2010).
Compassion satisfaction represents positive effects of helping in terms of pleasure resulting from ability to help others through the work (Stamm, 2010).Helper can also feel positively about colleagues and workplace or about contributing to the greater good of society through helping (Stamm, 2010).Compassion satisfaction is associated with higher levels of mindfulness and emotional separation of the helper (Thomas & Otis, 2010).
Compassion fatigue, consisting of secondary traumatic stress and burnout, represents negative consequences of helping which are connected with helper's exposure to trauma of his clients or patients (Bride, Radey & Figley, 2007;Figley, 1995;2002;Stamm, 2010).Increase in compassion fatigue is associated with the lower level of emotional separation (Thomas & Otis, 2010).Helper is at risk of compassion fatigue when his emotional connection with people, he helps, is high, and therefore has difficulties with distinguishing between his own feelings and feelings of the person he helps.
Secondary traumatic stress as a part of compassion fatigue is work-related stress resulting from secondary exposure of the helper to traumatic experience through helping people who experienced extremely stressful events (Stamm, 2010).It is important that helper does not experience traumatic events primarily, on his own, but indirectly by helping those who suffered (Figley, 2002;Stamm, 2010).Vicarious traumatisation of the helper is a result of his empathetic engagement with traumatic material of his clients (Figley, 2002).Helpers are at risk of developing secondary traumatic stress disorder as a result of helping and of being emotionally affected by traumatic experience of patients or clients (Figley, 2002).Symptoms of secondary traumatic stress disorder are similar to those of posttraumatic stress disorder-sleep difficulties, intrusive images, avoidance of the events or things that reminds helper of traumatic experiences of those he helped (Stamm, 2010).Increase in post-traumatic stress symptoms is correlated with lower compassion satisfaction (Lauvrud, Nonstad & Palmstierna, 2009).
Burnout as a part of compassion fatigue is associated with negative symptoms in various domains: cognitive (apathy, rigidity, lower self-esteem and preoccupation with trauma); behavioural (impatience, sleep difficulties and withdrawal); emotional (anxiety, fear, guilt, depression and hopelessness); spiritual (loss of purpose, loss of faith and questioning meaning of life); relational (isolation and mistrust); somatic (breathing difficulties, sweating and pain) and work (lower motivation and conflicts with colleagues) (Figley, 2002).Helper is exhausted, frustrated, feels that his work makes no difference, his workload is high and work environment is non-supportive (Stamm, 2010).Burnout is associated with lower levels of mindfulness and emotional separation (Thomas & Otis, 2010).
Professional quality of life scale (Stamm, 2005) is the most common measure of compassion satisfaction and compassion fatigue and has been translated into several languages.Slovak version of the instrument is not available yet.Objectives of present research were, therefore, to present preliminary results of on-going research study aimed at examination of psychometric properties of the Slovak version of the Professional quality of life scale.Paper is focused on the analyses of internal consistency and constructs validity of the instrument.
Compassion satisfaction was expected to be significantly positively correlated with optimism, selfesteem, positive state of mind (frequency of experiencing positive emotions) and negatively with negative state of mind (frequency of experiencing negative emotions), anxiety and depression.
Compassion fatigue, burnout and secondary traumatic stress were expected to be significantly positively correlated with negative state of mind (frequency of experiencing negative emotions), anxiety and depression and negatively with optimism, self-esteem and positive state of mind (frequency of experiencing positive emotions).

Participants and procedure
The sample consisted of 81 participants.Fifty-three (65.4 %) were students of social work in external form of study at Pavol Jozef Safarik University in Kosice, who worked in social care as social workers and 28 (34.6%) were people working in helping professions as counsellors and social workers in orphanages and counselling.All participants were female, aged 20-57 (M = 31.56;SD = 9.05) with 1-32 years of experience in helping professions (M = 3.89; SD = 6.65).
Participants received the informed content.Their participation was voluntary and anonymous.They were asked to complete professional quality of life scale, Life orientation test-revised, Emotional habitual subjective well-being scales, Rosenberg self-esteem scale, State-trait anxiety inventory and Beck depression inventory.

Professional quality of life scale
Professional quality of life scale (Stamm, 2010) is a 30-item measure for assessing level of positive (compassion satisfaction) and negative (compassion fatigue-burnout and secondary traumatic stress) effects of working in helping professions on the lives of the helpers.Respondents are asked to indicate frequency of positive and negative experience in their helping profession using five-point scale (1 = never; 5 = always).Three overall scores referring to the level of compassion satisfaction, burnout and secondary traumatic stress are calculated (responses to five items are reverse coded).Higher scores indicate higher levels of positive and negative effects of helping.Internal consistency estimates of Compassion satisfaction, Burnout and Secondary traumatic stress scales were 0.87, 0.90 and 0.87, respectively (Stamm, 2005).
English version of the Professional quality of life scale (Stamm, 2009) was translated into Slovak language using forward-translation and back-translation.Forward translation was conducted by professional counsellor speaking both English and Slovak language.Slovak translation was given to bilingual expert for back-translation into English language.Forward-translation, back-translation and original English version of the instrument were compared.Bilingual expert gave some suggestions for changes of the items of the Slovak version of the instrument in order to achieve its better conceptual equivalence with the original version.

Life orientation test-revised
Slovak adaptation of the Life orientation test-revised (Koverova & Ferjencik, 2013) was used to assess dispositional optimism.Life orientation test-revised is a 10-item measure for assessing dispositional optimism, i.e., generalised expectancies of good versus bad outcomes (Scheier & Carver, 1987;1993).Items are answered using five-point scale (0 = strongly disagree; 4 = strongly agree).Three items are reverse coded, total score is obtained as a sum of scale responses to six items (four items are fillers).Higher scores indicate higher level of optimism as a one-dimensional construct (Koverova & Ferjencik, 2013).Internal consistency estimates of the Life orientation test-revised were acceptable: 0.73 (Koverova & Ferjencik, 2013); 0.78 (Bailey, Eng, Frisch & Snyder 2007;Scheier, Carver & Bridges, 1994).Cronbach alpha of Life orientation test-revised in our research was 0.707. of Psychology Research: New Trends and Issues.6(2), 086-093.89

Emotional habitual subjective well-being scales
Emotional component of habitual subjective well-being was measured by original Slovak adaptation of Emotional habitual subjective well-being scales (Dzuka & Dalbert, 2002).Two separate scales (scale of positive state of mind and scale of negative state of mind) measure frequency of experiencing four positive (joy, happiness, pleasure and energy) and six negative (anger, shame, sadness, fear, guilt and pain) emotions or physical states (Dzuka & Dalbert, 2002).Respondents are asked to indicate the frequency of experiencing 10 emotions or physical states using six-point scale (1 = almost never; 6 = almost always).Two separate scores indicate the frequency of experiencing positive and negative emotions.Higher scores indicate higher frequency of experiencing positive and negative emotions.Internal consistency (Cronbach alpha) was 0.74 and 0.77 for scale of negative state of mind and scale of positive state of mind, respectively (Dzuka & Dalbert, 2002).In our research, Cronbach alpha estimates of scale of positive state of mind and scale of negative state of mind were 0.819 and 0.763, respectively.

Rosenberg self-esteem scale
Self-esteem was assessed by Slovak adaptation of Rosenberg self-esteem scale (Halama & Biescad, 2006).Ten items are answered by four-point scale indicating the level of agreement with the statement (1 = strongly disagree, 4 = strongly agree).Scale responses to five statements are reverse coded before sum of the overall score.Higher scores refer to higher level of self-esteem.Rosenberg self-esteem scale has adequate internal consistency (α = 0.76; Halama & Biescad, 2006;α = 0.83;Hatcher & Hall, 2009).Cronbach alpha of Rosenberg self-esteem scale in our research was 0.819.

State-trait anxiety inventory
Slovak version of State-trait anxiety inventory, part X-2 (STAI X-2; Mullner, Ruisel & Farkas, 1980) was used to assess trait anxiety, i.e., tendency of a person to experience anxiety in situations perceived as threatening.Twenty items of STAI X-2 refer to usual feelings of the person and are answered using four-point scale ranging 1-4 (1 = almost never, 4 = almost always).Sum of the scale responses to all items refers to the level of trait anxiety (seven items are reverse coded).High scores indicate high level of trait anxiety.State trait anxiety inventory, trait version is a reliable instrument for assessing trait anxiety (α = 0.91; Bados, Gomez-Benito & Balaguer, 2010; α = 0.88; Heretik, Ritomsky, Novotny, Heretik & Pecenak, 2009).Internal consistency of STAI X-2 in our research was 0.879.

Beck depression inventory
Slovak translation of Beck depression inventory (Prasko, Bulikova & Sigmundova, 2009) was used to assess the severity of depression.It consist of 21 groups of statements, each consisting of four items rated 0-3 according to severity of depression symptoms.Sum of scale responses refer to severity of depression.Higher scores indicate increasing severity of depression.Beck depression inventory is a reliable measure of depression (α = 0.90; VanVoorhis & Blumentritt, 2007;α = 0.94;Joe, Woolley, Brown, Ghahramanlou-Holloway & Beck, 2008).Cronbach alpha of Beck depression inventory in our research was 0.817.

Statistical analyses
Data were analysed using SPSS Statistics 20 software.Reliability of Professional quality of life scale was tested in terms of internal consistency (Cronbach α).To test construct and discriminant validity of Professional quality of life scale, bivariate correlation analyses (Pearson correlations) were computed.

Results
Descriptive statistics (means, standard deviations) and internal consistency estimates (α) for all measures are presented in Table 1.Scores on Professional quality of life scale were moderate for 90 Compassion satisfaction scale (M = 36.62;SD = 4.99) and Compassion fatigue (M = 45.16;SD = 8.02), lower to moderate for Burnout scale (M = 24.08;SD = 4.08) and lower for Secondary traumatic stress scale (M = 21.07;SD = 5.00).
Reliability estimates (Cronbach α) of Compassion satisfaction scale and Secondary traumatic stress scale were adequate (0.806 and 0.754, respectively).Internal consistency of Burnout scale was lower (0.556).Internal consistency estimate of Compassion fatigue (burnout items together with secondary traumatic stress items) was 0.785.
To test construct validity of Professional quality of life scale, Pearson correlations between measures were computed.Results of the correlation analyses are presented in Table 2.
Results of correlation analyses also supported discriminant validity of the Slovak version of the Professional quality of life scale.Compassion satisfaction was negatively correlated with anxiety (r = −0.314;p < 0.01) and depression (r = −0.225;p < 0.05).Compassion fatigue was negatively associated with optimism (r = −0.324;p < 0.01) and self-esteem (r = −0.257;p < 0.05).Negative relationships were found between burnout and optimism (r = −0.347;p < 0.01), self-esteem (r = −0.310;p < 0.01) and positive state of mind, i.e., frequency of experiencing positive emotions (r = −0.271;p < 0.05).Secondary traumatic stress and optimism were negatively correlated (r = −0.236;p < 0.05).However, significant negative correlations between compassion satisfaction and negative state of mind did not emerge, neither did negative correlations between secondary traumatic stress, selfesteem and positive state of mind, nor negative associations between compassion fatigue and positive state of mind (see Table 2).

Discussion
Objective of the present research was to examine psychometric properties (internal consistency and construct validity) of the Slovak version of the Professional quality of life scale on the sample of people working in helping professions.Preliminary results of on-going research showed that the Slovak translation of the instrument was reliable (in terms of internal consistency).Internal consistency estimates of Compassion satisfaction and Compassion fatigue were acceptable.However, when reliability estimates of the Compassion fatigue subscales (Secondary traumatic stress scale and Burnout scale) were analysed separately, Cronbach alpha of Burnout scale decreased.These results suggest that the Compassion fatigue scale of the Slovak translation of the Professional quality of life scale is more reliable as a whole.
Results also supported both convergent and discriminant validity of the instrument.Based on Figley's (2002) categorisation of compassion fatigue symptoms, we focused primarily on emotional (anxiety, depression, positive and negative states of mind) and cognitive (optimism, self-esteem) correlates of professional quality of life in the analyses.
Higher levels of compassion satisfaction were associated with higher levels of optimism and selfesteem, higher frequency of experiencing positive emotions (joy, happiness, pleasure and energy) and lower levels of anxiety and depression.These findings correspond with Stamm's (2010) conceptualisation of compassion satisfaction that emphasises mainly its positive emotional consequences on the life of the helper.Our findings brought evidence that compassion satisfaction is associated not only with positive emotional, but also with positive cognitive outcomes (optimism and self-esteem).
Higher levels of compassion fatigue (burnout and secondary traumatic stress) were associated with lower levels of optimism and self-esteem, higher levels of depression and anxiety and more frequent negative state of mind (i.e., experiencing negative emotions-fear, anger, shame, sadness, guilt and pain).
Higher levels of burnout were related to higher levels of anxiety and depression, more frequent negative emotional state (feelings of anger, shame, sadness, fear, guilt and pain), less frequent positive emotional state (feelings of joy, happiness, pleasure and energy) and lower levels of optimism and selfesteem.
817 Note: PROQoL_CS = professional quality of life scale-compassion satisfaction; PROQoL_CF = professional quality of life scale-compassion fatigue; PROQoL_B = professional quality of life scale-burnout; PROQoL_STS = professional quality of life scale-secondary traumatic stress; LOT-R = life orientation testrevised; RSES = rosenberg self-esteem scale; SPSM = scale of positive state of mind; SNSM = scale of negative state of mind; STAI = state-trait anxiety inventory, trait version; BDI = beck depression inventory.

Table 2 . Pearson correlations among measures
PROQoL_CS = professional quality of life scale-compassion satisfaction ; PROQoL_CF = professional quality of life scale-compassion fatigue; PROQoL_B = professional quality of life scale-burnout; PROQoL_STS = professional quality of life scale-secondary traumatic stress; LOT-R = life orientation testrevised; RSES = Rosenberg self-esteem scale; SPSM = scale of positive state of mind; SNSM = scale of negative state of mind; STAI = state-trait anxiety inventory, trait version; BDI = beck depression inventory.*p < 0.05.**p < 0.01.