THE MEDIATING ROLE OF PSYCHOLOGICAL RESILIENCE BETWEEN DOMESTIC PSYCHOLOGICAL ABUSE AGAINST WORKING WOMEN AND BURNOUT
Psychological abuse as a form of domestic violence against working women is prevalent but underreported almost all over the world. The present study was conducted to examine the relationship between domestic psychological abuse and burnout, and how psychological resilience mediates between them. One thousand married teachers from private secondary schools were selected through purposive sampling. Study results demonstrated that there is a relationship between domestic psychological abuse against working women and burnout, and that psychological resilience has a positive supportive effect in overcoming depersonalization among these women, yet the study also showed a lack of significant intervention in the relationship between psychological abuse and depersonalization of working women. This study confirms the absence of total or partial mediation to address psychological abuse and depersonalization of working women.
INTRODUCTION
Psychological abuse (also known as emotional abuse) is the most prevalent form of domestic violence and the primary factor in physical abuse against women all over the world (World Health Organization, 2021). There is an increasing emphasis on understanding emotional abuse as a distinct concept in order to recognize it as a violation of basic human rights, as expressed in several declarations of women's rights (e.g., United Nations Assembly 1993 on the Elimination of Violence against Women; 1995 Beijing Declaration and Platform for Action). Yet, in spite of these declarations, women's rights are still neglected almost all over the world (Das et al., 2015).
December 2003 was the first time that the United Nations General Assembly mandated, in Resolution 58/185, an extensive study of all forms and indicators of violence against women (Garcia-Moreno et al., 2006). Although some countries have made significant contributions by developing legislation, weak political commitment in many countries continues to allow an environment that is not conducive for women to live as peacefully and with integrity as men (Garcia-Moreno et al., 2006). This situation is underreported in the developing countries, especially in Asia (Koirala & Chuemchit, 2020). Many countries (e.g., Bangladesh, India, Sri Lanka, Nepal, Thailand, Malaysia, Indonesia, the Philippines, Japan, and Pakistan) maintain religious values and cultural norms that are the main sources shaping the role of men as the dominant ruler, whereas women are not allowed to seek justice for their legal rights in the courts. “Globally, 31% of women aged 15–49 have been subjected to physical and/or sexual violence from a current or former husband or intimate partner, or sexual violence from a non-partner, or both in their lifetime (defined as since the age of 15). That is almost one in every three women, or up to 852 million women worldwide. Intimate partner violence accounts for the largest proportion of this violence: an estimated 27% of ever-married/partnered women aged 15–49 experience physical and/or sexual intimate partner violence in their lifetime” (World Health Organization 2021). Similarity a study conducted in Serbia reported that of 6500 examined, 307 (4.9%) had experienced physical and/or psychological violence in the past 12 months(Jovanović et al., 2020a).
Several international organizations e.g.; Amnesty International, Feminist Majority Foundation, Global Alliance Against Traffic in Women, WHO, CEDAW are working on safeguarding the rights of women and reporting factual data about the situation, identifying the factors causing domestic violence, and assisting in developing strategies and policies to improve the situation through law enforcement.
There is a common misunderstanding of the problem of violence against women that only household women are the victims of this injustice, while the actual situation is that working women are equal sufferers. For example, teaching is a profession where teachers carry multiple roles with many pressures (García-Carmona et al., 2019). Many studies are available on domestic violence in different professions, but there is a lack of research on the experience of teachers, especially with regard to psychological abuse, which leaves no scars or bruises but affects the psychological and intellectual well-being of the women and decreases their passion for work (Wang et al., 2017). Although a few studies are available on how psychological abuse affects the well-being of women, no research is available on the relationship between domestic psychological abuse and burnout and how women overcome their effects. In Pakistan, for example, economic pressure forces families to allow women to work, but they are still restricted and face many resulting stressors (Pervez, Zia-ur-Rehman, Javed, & Raza, 2015).
OBJECTIVES OF THE STUDY
The study had the following objectives:
-
To explore the relationship between domestic psychological abuse against working women and burnout
-
To analyze how psychological resilience mediates between domestic psychological abuse against working women and burnout
THEORETICAL BACKGROUND
The theoretical foundation of this study is postmodern feminism, which offers strenuous criticism of any kind of abuse or violence against women. Postmodern feminism has gained recognition as a result of extensive research conducted by Haraway (2020). The theory supports females who are undergoing domestic abuse and violence by introducing healing responses using a relational approach and integrating a family driven by racial and cultural identity (Laing, 2000). Postmodern feminism reaffirms the centrality of women in the family, which is considered to be a foundational institution of human society (Stacey, 1988). The theory is especially relevant for women victimized by psychological abuse (Murray, 2006). Postmodern feminism describes the important role of women in a society that was originally established on the basis of ideas rooted in the Enlightenment and governed by an essentialist approach that endorsed patriarchal meanings of autonomy, identity, and involvement in the good life (Ashcraft & Mumby, 2004). Gender associated with the distribution of power and authority as well as supporting the relational approach (Ely & Padavic, 2007) is the fundamental axis supporting relational and power-exercising processes in the light of postmodern feminism (Flax, 1990). As a result, postmodern feminist theory helps to address the repudiation of emotional abuse faced by women working as the equals of men, and it also combats the marginalization of women and the dominance of men that culminates in the fallout of stress, exhaustion, depersonalization, and detachment of women (Meyerson, 1998).
LITERATURE REVIEW
Psychological Abuse
In 2021 the World Health Organization (WHO) reported that all forms of domestic violence are major violations of women's human rights and cause serious physical and mental health issues. The WHO also emphasized that this issue is prevalent globally with varying degrees of intensity regionally and culturally. This issue needs to be addressed on urgent basis because it is hindering the attainment of the United Nations' Sustainable Development Goal 5 (SDG 5; Gender Equality and Women and Girls' Empowerment) by 2030 (Bayeh, 2016).
Domestic psychological abuse is described as verbal, nonphysical behavior, harassment, belittling, excessive possessiveness, separation of a partner, deprivation of physical and economic resources, and an attitude that seeks to dominate the other person through fear and humiliation (Porrúa-García et al., 2016). The abuser's tactics to psychologically abuse the partner include embarrassment, control, concealing of information, isolation of the victim, exercise of financial control, stalking, creation of work restrictions, demeaning the victim, undermining the victim's self-confidence and sense of worth, and even seeking to convince the victim that she is crazy (Gómez et al., 2018).
According to a report by Taylor et al. (2013), many regions of the world (e.g., Eastern Mediterranean Region, Central Sub-Saharan Africa, East Asia, Caribbean and Central Asia) have not collected population-based data. Therefore, the authors recommended that countries in these areas acquire the necessary data for appropriate policy and decision making. Seeing this gap in research, we have designed this study to contribute to the international research data so that hindering SDG-5 can be minimized and gender equity, equality, and empowerment can be improved.
Prevalence of Domestic Abuse
The highest prevalence of domestic abuse is reported in Africa, the Eastern Mediterranean, and Southeast and Eastern Asia, where approximately 37% of women experience partner violence for a variety of reasons (e.g., young age, length of marriage, higher education, husbands' low education, working spouses, military occupation, fewer children, multiple spouses, smoking spouses, aggressive spouses, chronic illness in women or their spouse, and inadequate family income; Barnawi, 2017). Figure 1 shows the critical situation of domestic violence in different countries, highlighting the significance of the issue.
Citation: Performance Improvement Quarterly 36, 2; 10.56811/PIQ-21-0002
Psychological Abuse and Working Women
Emotionally abused working women express feelings of depression, fear, shame, loss of self-esteem, and anxiety (Karakurt & Silver, 2013), and all these factors impede their progress by hindering their potential (Aslan et al., 2019). A qualitative study conducted by Moe and Bell (2004) in the United States reported that working women manage the effects of abuse better because of their greater financial independence, and workplace support for domestically abused women can be a constructive contribution (see also Showalter, 2016).
Psychological abuse is least reported because of social barriers, criticism, privacy, fear, lack of confidentiality, and mistrust in the system (Vranda et al., 2018), especially in the education sector. The term domestic psychological abuse is described as a state of exploitation by an intimate partner. In Pakistan, an Islamic state, an intimate partner is described as a husband because culturally, religiously, and socially there is no concept of an intimate relationship for a woman with a man other than her husband.
Psychological Abuse and Burnout
Burnout is described as a psychological inability and social malfunctioning of employees because of stressful workplace experiences (Maslach & Leiter, 2016). It is significantly associated with a decline in quality of work, absenteeism, and an intention to leave the job. Burnout not only causes deterioration in the physical health of the employee but can also transmit a psychologically negative influence to coworkers, which becomes costly for the organization in terms of its effectiveness and growth (Chirico, 2016).
Maslach and Jackson (1981) described burnout as a prolonged psychological process with three dimensions: emotional exhaustion, feelings of cynicism, and reduced personal accomplishment. The first dimension concerns extreme feelings of emotional lethargy. The second dimension refers to an employee's isolated behavior and feelings of depersonalization leading to detachment from other people, including coworkers, clients, customers, subordinates, and supervisors. The third dimension is dominated by the employee's negative self-evaluation of their capabilities and feelings of ineffectiveness regarding expected and actual job performance.
Women who live in an abusive relationship have been observed to be depressed (Trevillion et al., 2012), to have low self-esteem, and to experience significant fatigue (Khalid, 2015). When psychologically abused working women experience burnout, it multiplies their feelings of depression, stress, and anxiety. Few researchers have found that this state negatively affects the work engagement and performance of these women (Wathen et al., 2015).
Psychological Resilience Among Working Women
Psychological resilience is characterized by the ability to successfully respond to extreme stress or trauma or adverse experience (Ran et al., 2020). Resilience is professional workers' ability to turn stressful events into opportunities for personal growth and to develop themselves as energetic, task-focused workers. In the past few decades, increased uncertainty, instability, and numerous routine life stress have heightened the significance of research on resilience (MacKinnon & Derickson, 2013), especially on psychological resilience in the workplace. Few studies are available on cultural and environmental factors contributing to teachers' resilience (Castro et al., 2010), and no study is available on how personal factors such as self-efficacy and self-confidence contribute to teachers' resilience. Teachers are reported to face a high level of adversity in professional settings that results in reduced psychological functioning (Hargreaves & Fink, 2006). Psychological resilience among teachers is described as the psychological stamina that helps them overcome stress and proactively develop ways to enhance their well-being (Gibbs & Miller, 2014). Resilience not only increases creativity in teachers but also supports their ability to manage excessive workloads, address workplace conflicts, and overcome burnout (Rajoo, 2020). Seeing the strengths of resilience (Yonezawa et al., 2011) suggests that instead of technical workshops, teachers should be trained on how to be resilient. Psychological resilience is a positive emotional trait of professionals, but unfortunately managers often do not know how to utilize this psychological capital to reduce employee burnout, which can damage the performance of the employees.
Some researchers have validated the relationship between resilience and burnout, but limited data are available on psychologically abused working women (e.g., teachers) and how burnout affects their psychological resilience. In addition, few studies are available on how resilience can be a mediating variable that contributes to a decrease in burnout (Zou et al., 2016).
SIGNIFICANCE OF THE STUDY
The percentage of women in the workforce is increasing every year in most of the world (Stockemer & Byrne, 2012). Therefore, this study is significant in identifying the causes of domestic psychological abuse and in knowing how teachers can overcome burnout utilizing their psychological resilience. Moreover, findings of this study can provide insight to educational administrators about how to plan the training of their teacher to overcome the effects of domestic abuse and how a supportive workplace environment can help in decreasing the effects of domestic psychological abuse in the workplace. Moreover, it is reported that lack of regular data collection hinders the planning of effective interventions (Mijatović-Jovanović et al., 2020).
Therefore, the study will help in designing appropriate strategies for mediation. In Pakistan, where only 21% of women are employed outside the home—which is very low compared to neighboring countries (Sarwar & Abbasi, 2013) such studies can give insight to policymakers about how to improve the situation and take some practical measures to safeguard the rights of women so they can contribute to the country's economic prosperity. The women in Pakistan are only 21% of the total population are employed (Sarwar & Abbasi, 2013), indicating the significance for providing an environment where they can come out to contribute in the country's economy.
RESEARCH HYPOTHESES
On the basis of the gap in the above-mentioned literature, the following hypotheses were developed (Figure 2):
Citation: Performance Improvement Quarterly 36, 2; 10.56811/PIQ-21-0002
Ho1: There is a significant relationship between domestic psychological abuse against working women, burnout, and psychological resilience.
Ho2: Psychological resilience mediates between domestic psychological abuse against working women and emotional exhaustion.
Ho3: Psychological resilience mediates between domestic psychological abuse against working women and personal accomplishment.
Ho4: Psychological resilience mediates between domestic psychological abuse against working women and depersonalization.
METHODOLOGY
A quantitative descriptive cross-sectional study was designed to explore the hypotheses by applying the survey data collection technique. The study uses sequential mediation analysis to examine the domino effects of emotional exhaustion, personal accomplishment, and depersonalization, with psychological abuse considered as a prime factor triggering a chain of reaction that affects the psychological resilience of working women (Baron & Kenny, 1986).
With regard to this research study, Model 1 of the Process Macro has been executed to validate the tentative hypotheses using sequential mediation in a single model to obtain the estimates for complete total effect, direct effects, and indirect effects of the constructs on the theoretical model. A total of 5,000 bootstrap samples were run on the model with a 95% confidence level. We applied Baron and Kenny's (1986) approach to mediation analysis, which involves four steps. First, the correlations between the independent variable and each of the dependent variables were validated in order to establish an effect that had to be mediated. Second, the correlation between the independent variable and the mediating variable was confirmed in order to establish the mediating variable as an outcome variable. In the third step, it was determined whether the mediating variable was significantly correlated with each of the dependent variables, because both mediating and dependent variables are outcomes of the independent variable. During this step, the independent variable was controlled while the correlations between the mediating and dependent variables were proven. The fourth step involved the authentication of the complete mediation paths across all the variables.
Before steering sequential mediation analysis to test the proposed hypotheses, customary linear regression model assumptions were tested comprising a linearity test, a normality test, a test of the homogeneity of variance of error, and a test of error independence. All assumptions were confirmed, suggesting the proof of the individuality of the mediational model. The analysis section comprised the frequency statistics depicting the demographic landscape of the sample respondents. The study employed descriptive analysis highlighting the mean and standard deviation values of the constructs, followed by determining the Kaiser-Meyer-Olkin (KMO) value for validating the sampling satisfactoriness of the collected data by administering exploratory factor analysis (EFA). Reliability analyses of the current data were also conducted according to Cronbach's alpha coefficient. Finally, correlations among the variables were determined.
Population and Sample of Study
The population of the study was comprised of the married teachers from the private secondary schools of Lahore, Pakistan. The secondary schools were selected through simple random sampling from the list of private schools provided on the website of the Board of Intermediate and Secondary Education. Ten married teachers were purposely selected from each school. A total of 1,000 questionnaires were administered, and 504 were returned (essentially a 50% response rate).
Instruments for Data Collection
Data were collected using three adapted tools self-administer by the teachers. Domestic psychological abuse was measured by the widely used Index of Psychological Abuse (IPA; Katz, Arias, & Beach, 2000). The IPA comprises 33 brief statements developed by Sullivan, Parisian, and Davidson (1991) (Cronbach's alpha = .83) and uses a 4-point Likert scale: 1 (never), 2 (rarely), 3 (sometimes), 4 (often), and 8 (not applicable). Respondents are instructed to indicate the extent to which they agree or disagree with each statement. The Pearson product–moment correlation method was used to confirm the validity of the survey items (Wijaya & Kloping, 2021).
Burnout was measured by the widely used Maslach Burnout Inventory (MBI) developed by Maslach, Jackson, and Leiter (1996) comprising a 7-point Likert scale with 22 items. The inventory consists of three dimensions: emotional exhaustion, depersonalization, and personal accomplishment. The subscales of the inventory are well established by several studies conducted in different cultural contexts. Cronbach's alpha of the scale is .85.
The Resilience Evaluation Scale (RES) developed by Van der Meer et al. (2018) was applied to measure psychological resilience. The RES was administered because of its underlying constructs: self-confidence and self-efficacy. Moreover, this 9-item 5-point Likert scale was simple in language and easy to understand by the participants. Cronbach's alpha of the scale is .87.
Statistical Analyses
The study used sequential mediation analysis to examine the domino effects (Slåtten & Lien, 2021) in emotional exhaustion, personal accomplishment, and depersonalization, with psychological abuse considered as a prime factor triggering a chain reaction involving the psychological resilience of working women (Baron & Kenny, 1986). The analysis section comprises initially the frequency statistics depicting the demographic landscape of the sample respondents. Furthermore, the study employed descriptive analysis highlighting the mean and standard deviation values of the constructs, followed by determining the KMO value for validating the sampling satisfactoriness of the collected data. The study also included reliability analysis of the current data using the Cronbach's alpha coefficient and, lastly, correlations among the variables.
Ethical Considerations
The selected sample was assured of the confidentiality of the data.
Demographics
Data were collected through a questionnaire-based survey administered on a randomly chosen sample of 525 female participants, including married female teachers from 100 randomly selected secondary schools of Lahore. The analysis of age demographics showed that approximately 28% of the teachers were 31 to 35 years of age, the largest group of participants. About 23% of the teachers were 26 to 30 years of age, the second-highest percentage among all participants. Moreover, 22%, 18%, and 9% of the respondents were in age brackets 36 to 40 years, 41 to 45 years, and 21 to 25 years, respectively. The summary of the demographic characteristics of the study respondents are shown in Table 1.
Means and Standard Deviations
The results of the descriptive statistical analysis, including mean and standard deviation values of all the theoretical variables considered in this research study, as determined by the responses and opinions shared by married working females, can be seen in Table 2. The only predictor variable (i.e., Psychological Abuse) shows a mean of 2.16, with a standard deviation equal to 0.46. Psychological Resilience taken as the intervening variable records a mean of 1.51 and a standard deviation of 1.28. Among the three criterion variables, the mean and standard deviation of Emotional Exhaustion are 2.14 and 0.91, respectively. For Personal Accomplishment, the mean value is 2.2207 and the standard deviation is 1.01. Lastly, in reference to the mean and standard deviation values of Depersonalization, they have been reported as 2.04 and 0.86, respectively.
Sample's Data Sufficiency Using KMO Indexes
To assess the data adequacy of the selected sample, KMO indexes were examined by performing EFA. Observing the pattern matrix of Psychological Abuse constructed from the results, we observed that six items were not involved from the originally adapted measuring instruments, because their item loadings showed a value less than 0.5. Thus, the KMO measure for evaluating the sampling adequacy of Psychological Abuse was improved to 0.77 (p = .000). The verified index of KMO analysis for Psychological Resilience was 0.89 (p = .000). Likewise for Emotional Exhaustion, the KMO measure approached 0.67 (p = 0.05). The KMO index of Personal Accomplishment was reported as 0.76 (p = .000), and for Depersonalization this measure of KMO was 0.667 (p = .000). Summary results illustrating sampling sufficiency measures using KMO are shown in Table 2.
Construct Reliability Coefficients
Interitem reliability for each of the theoretical variables was measured and analyzed based on their coefficients of Cronbach's alpha. The results are shown in Table 2: alpha coefficient of Psychological Abuse is 0.83; Psychological Resilience is 0.91; Emotional Exhaustion is 0.64; Personal Accomplishment is 0.76, and Depersonalization is 0.63. The reliability coefficients of each of the measurement scales divulge to the interpretation of the extent of each scale to be the consistent and reliable measure of the concepts. It can be inferred that all the scales are dependable measures of the variables.
Correlation Analysis
Correlation analysis has been used to assess the significance and extent to which the constructs and measurement variables bear a relationship with each other in the current research study. On the basis of the results of correlation analysis, we recognized that psychological abuse (Psy-Abuse) does not have any significant relationship with psychological resilience (Psy-Res, −7.3%) and Depersonalization (−1.9%) based on a significance value greater than .05 (p = .096 and p = .661, respectively). Moreover, Psy-Abuse has a significantly negative relationship (p < .05) with emotional exhaustion (E-E) and psychological accomplishment (P-A, with correlation values of −9.4% (p = .032) and −12.6% (p = .004), respectively. Furthermore, Psy-Res has significantly positive relationships with E-E (i.e., 39.7%, p = .000), P-A (i.e., 55.6%, p = .000), and Depersonalization (i.e., 42.3%, p = .000). The results of correlation analysis are shown in Table 2.
MULTIPLE REGRESSION ANALYSIS AND MEDIATION ANALYSIS
Path 1: Intervening Role of Psychological Resilience on Relationship Between Psychological Abuse and Emotional Exhaustion
Multiple regression analysis was conducted to investigate the impact of Psy-Abuse faced by female teachers on their E-E, P-A, and Depersonalization. In addition, based on the main objective of the study, the mediating role of their Psy-Res was also taken into consideration for assessing the effect of Psy-Abuse on the E-E, P-A, and Depersonalization expressed by these victimized females using Baron and Kenny's (1986) mediation analysis. As described above, this analysis involves four steps. First, the correlations between the independent variable (i.e., Psy-Abuse) and each of the dependent variables (i.e., E-E [r = −9.4%], P-A [r = −12.6%] and Depersonalization [r = −1.9%]) was validated in order to establish an effect that had to be mediated. Second, the correlation between Psy-Abuse and the mediating variable (i.e., Psy-Res [r = −7.3%]) was confirmed in order to establish this mediating variable as an outcome variable. In the third step, it was determined that Psy-Res (i.e., the mediating variable) was significantly correlated with each of the dependent variables (i.e., E-E [r = 39.7%], P-A [r = 55.6%], and Depersonalization [r = 42.3%]), because both mediator (psychological resilience) and dependent variables (E-E, P-A, and Depersonalization) are outcomes of the independent variable (psychosocial abuse). During this step, the psychological abuse variable was controlled while the correlations between psychological resilience and emotional exhaustion, personal accomplishment, and depersonalization were proven. The fourth step involved the authentication of the complete mediation paths across all the variables Results are shown in Table 3. Because the results satisfied the first three steps of Baron and Kenny mediation, and because the direct effects of the predictor variable (i.e., Psy-Abuse) on the criterion variables (i.e., E-E, P-A, and Depersonalization) were not equal to zero, henceforth, partial mediation was observed based on the data collection. This also satisfies the consistency of mediational hypotheses (Baron & Kenny, 1986; Shabbir & Gardezi, 2020).
Accordingly, the first step, linear regression, was performed on the impact of Psy-Abuse on E-E (β = −0.185, p = .032), and results are shown in Table 3 and Figure 3. As a result, in the second step, regression analysis was performed on the impact of Psy-Abuse on Psy-Res (β = −0.201, p = .096) and then in the third step, the direct impact of psychological resilience on E-E was investigated (β = 0.280, p = .000). Lastly, the total effect of Psy-Abuse on E-E under the intervening role of Psy-Abuse showed a nonsignificant result (β = −0.129, p = .105). The results shown in Table 3 and Figure 3 reveal that Psy-Abuse has a direct negative effect on the E-E of married working females. In addition, psychological resilience, although independently, has a positive supportive effect on E-E, yet has no intervening role in the relationship between Psy-Abuse and the E-E level of the females. This confirms the absence of total or partial mediation as Psy-Abuse emerges as a direct predictor of E-E: {\rm{Amount\ of\ mediation}} = {\rm{indirect\ effect}}{\rm{Total\ effect}} = {\rm{direct\ effect}} + {\rm{indirect\ effect}}c = c^{\prime} + ab{\rm{Without\ mediation{\hbox :}\ Total\ effect}} = - 0.185{\rm{With\ mediation{\hbox :}\ Total\ effect}} = - 0.129 + \left( { - 0.201} \right)\left( {0.280} \right){\rm{Total\ effect}} = - 0.186The above expression shows that the total effect of Psy-Abuse with the mediating role of Psy-Res on E-E is almost the same as it would be without the intervention of Psy-Res.
Citation: Performance Improvement Quarterly 36, 2; 10.56811/PIQ-21-0002
Path 2: Intervening Role of Psychological Resilience on Relationship Between Psychological Abuse and Personal Accomplishment
Based on the second main objective of the study, the mediating role of Psy-Res was also taken into consideration for assessing the effect of Psy-Abuse on the P-A by these victimized women using Baron and Kenny Mediation. Accordingly, linear regression was performed on the direct impact of Psy-Abuse on P-A (β = −0.274, p = .004), and results can be seen in Table 3 and Figure 4 showing a significant negative impact of Psy-Abuse on the P-A of the females. As a result, in the second step, regression analysis was performed on the impact of Psy-Abuse on Psy-Res (β = −0.201, p = .096) and then in the third step, the direct impact of Psy-Abuse on P-A was investigated (β = 0.432, p = .000). Lastly, the total effect of psychological abuse on personal accomplishment under the intervening role of Psy-Abuse showed a highly significant result (β = −0.187, p = .018). The results shown in Table 3 and Figure 4 reveal that Psy-Abuse has a direct positive effect on the P-A of married working women. In addition, Psy-Res also has a positive supportive effect on P-A, and it has an even stronger positive intervening role in the relationship between Psy-Abuse and the P-A level of the women. This confirms the presence of a partial mediation as Psy-Abuse emerges as a strong indirect predictor of P-A. {\rm{Amount\ of\ mediation}} = {\rm{indirect\ effect}}{\rm{Total\ effect}} = {\rm{direct\ effect}} + {\rm{indirect\ effect}}c = c^{\prime} + ab{\rm{Without\ mediation{\hbox :}\ Total\ effect}} = - 0.274{\rm{With\ mediation{\hbox :}\ Total\ effect}} = - 0.187 + \left( { - 0.201} \right)\left( {0.432} \right){\rm{Total\ effect}} = - 0.274The above expression shows that the total effect of Psy-Abuse with the mediating role of Psy-Res on P-A is the same as that of the total effect without the mediation of Psy-Res.
Citation: Performance Improvement Quarterly 36, 2; 10.56811/PIQ-21-0002
Path 3: Intervening Role of Psychological Resilience on Relationship Between Psychological Abuse and Depersonalization
Based on the second main objective of the study, the mediating role of Psy-Res was also taken into consideration for assessing the effect of Psy-Abuse on the depersonalization experienced by these married women using Baron and Kenny Mediation. Accordingly, linear regression was performed on the direct impact of Psy-Abuse on depersonalization (β = −0.036, p = .661), and results can be seen in Table 3 and Figure 4 showing an insignificant impact of Psy-Abuse on the depersonalization experienced by the women. As a result, in the second step, regression analysis was performed on the impact of Psy-Abuse on Psy-Res (β = −0.201, p = .096) and then in the third step, the direct impact of Psy-Res on depersonalization was investigated (β = 0.286, p = .000). Lastly, the total effect of Psy-Abuse on Depersonalization under the intervening role of Psy-Abuse showed a nonsignificant result (β = 0.022, p = .769). The results shown in Table 3 and Figure 5 reveal that Psy-Abuse has a direct but insignificant effect on the Depersonalization of married working women. In addition, Psy-Res, although it independently has a positive supportive effect on Depersonalization, yet it displays an insignificant intervening role in the relationship between Psy-Abuse and the Depersonalization level of the women. This confirms the absence of a total or partial mediation as Psy-Abuse emerges as an insignificant predictor of Depersonalization. {\rm{Amount\ of\ mediation}} = {\rm{indirect\ effect}}{\rm{Total\ effect}} = {\rm{direct\ effect}} + {\rm{indirect\ effect}}c = c^{\prime} + ab{\rm{Without\ mediation{\hbox :}\ Total\ effect}} = - 0.274{\rm{With\ mediation{\hbox :}\ Total\ effect}} = 0.022 + \left( { - 0.201} \right)\left( {0.286} \right){\rm{Total\ effect}} = - 0.0355The above expression shows that the total effect of Psy-Abuse with the mediating role of Psy-Res on Depersonalization is slightly negative as compared to without intervention of Psy-Res, which results in a negative yet greater influence on P-A.
Citation: Performance Improvement Quarterly 36, 2; 10.56811/PIQ-21-0002
Zero-Order, Partial, and Part Correlations
Zero-order correlation is the correlation between two (i.e., independent and dependent) variables without regulating for the impact created by any other variable(s), hence is similar to Pearson correlation. It helps identify bivariable correlations between variables while controlling other variables. The results in Table 3 show that Psy-Abuse has a negative unsubstantial relationship (r = −9%) with E-E. Psy-Abuse has a negative zero-order correlation (r = −13%) with P-A. Likewise, Psy-Abuse also has a small negative relationship with depersonalization (r = −2%). Psy-Abuse also has a negative zero-order correlation with Psy-Res (r = −7%). Psy-Res, on the other hand, has a substantial positive relationship with E-E (r = 39.7%). Psy-Res has a considerable and highly positive zero-order correlation with P-A (r = 55.6%). Lastly, it can be seen that Psy-Res has a negative minimal zero-order correlation with Depersonalization (r = −3%).
On the other hand, partial correlation is a correlation between the independent and the dependent variables after regulating for influence of any other variable on both the independent and the dependent variable. The results in Table 3 show that Psy-Abuse has a negative, unsubstantial relationship (r = −9%) with E-E. Psy-Abuse has a negative partial correlation (r = −13%) with P-A. Likewise, Psy-Abuse also has a small negative relationship with Depersonalization (r = −2%). Psy-Abuse also has a negative partial correlation with Psy-Res (r = −7%). Psy-Res, on the other hand, has substantial positive relationship with E-E (r = 39.3%). Psy-Res has a considerable and highly positive partial correlation with P-A (r = 55.3%). Lastly, it can be seen that Psy-Res has a negative minimal partial correlation with Depersonalization (r = −3%).
This leads to identification of part correlation, also referred to as semipartial correlation. Like partial correlation, part correlation is correlation between the two variables (i.e., independent and the dependent) after controlling and regulating for one or more than one variable, only taking into account the influence of control variable(s) on the independent variable. In other words, part correlation does not have control and regulation for influence of any confounding variable(s) on the dependent variable. The results in Table 3 show that psychological abuse has a negative, unsubstantial relationship (r = −9%) with E-E. Psy-Abuse has a negative part correlation (r = −13%) with P-A. Likewise, Psy-Abuse also has a small negative relationship with Depersonalization (r = −2%). Psy-Abuse also has a negative part correlation with psychological resilience (r = −7%). Psy-Res, on the other hand, has a substantial positive relationship with E-E (r = 39.1%). Psy-Res has a considerable and highly positive part correlation with P-A (r = 55.8%). Lastly, it can be seen that Psy-Res has a negative minimal part correlation with Depersonalization (r = −3%).
Multicollinearity Diagnostics
Multicollinearity diagnostics involves identification of the presence of a considerable correlation between two or more independent or predictor variables. This can be detected using a statistical metric called the Variance Inflation Factor (VIF). The VIF measures correlation and the strength of this correlation between two or more independent or predictor variables in the regression model. Tolerance, one of the two multicollinearity diagnostics' tools, is also the measure of collinearity reported as 1-R2. A minor tolerance value designates that the variable being considered is nearly the perfect linear blend of independent variables present in the regression equation, such that it is not to be added in the regression equation. Table 3 shows that all the variables in the linear regression relationship exhibit a small value of tolerance. For the VIF, the variables in the linear equation indicate a value of 1, showing that there is no correlation between the predictor variables.
RESULTS
The following results of the current study indicated that domestic psychological violence had an impact on working women in regard to E-E, P-A, and Depersonalization. However, the total effect of Psy-Abuse on E-E under the intervening role of Psy-Abuse did not have a significant relationship. On the other hand, Psy-Res had a positive impact on E-E, yet had no intervening role in the relationship between Psy-Abuse and the E-E level of the women. Furthermore, Psy-Res had a significant negative impact through Psy-Abuse on P-A, while P-A had a highly significant relationship with the intervening role of Psy-Abuse. In addition, Psy-Res was also positively supported and had a strong intervening role in the relationship between Psy-Abuse and P-A. As a result, expressions of the study participants showed that domestic Psy-Abuse along with the affects of Psy-Res on Depersonalization was slightly negative compared to lack of mediation effects on P-A.
DISCUSSION
Adopting the feminist postmodern perspective, we investigated the relationship between psychological abuse (Psy-Abuse) and emotional exhaustion (E-E), personal accomplishment (P-A), and depersonalization (DP) while noting the mediating effects of psychological resilience (Psy-Res) among women in the workplace. In this study, mean scores were Psy-Abuse 2.16 (0.46), Psy-Res 1.51 (1.28), E-E 2.14, P-A 2.22 (1.01), and DP 2.04 (0.86). Thus, bullying is connected with psychological distress that causes job strain and reduces social support, while burnout is interconnected with workplace bullying and psychological distress (Bernotaite & Malinauskiene, 2017). It is observed that six items from the original measuring instrument with items loading were less than 0.5, while KMO for Psy-Res was reported as 0.674 (p = .000). On the other hand, KMO of E-E was observed (p = .05) as were P-A (p = .000) and Depersonalization (p = .000) as shown in Table 2. The growing literature affirms that a stressful work environment is often associated with workplace bullying because of worsened interpersonal relationships caused by stressful working conditions (Takaki et al., 2010). In another study, Järvelä & MacNeil (2021) showed that social support moderates between a stressful work environment and work-related well-being in teachers. Interitem reliability for each variable was analyzed with coefficients of Cronbach's alpha. Table 2 shows the alpha Coefficients of Psy-Abuse as 0.832, Psy-Res as 0.916, E-E as 0.647, P-Accomplishment as 0.762, and Depersonalization as 0.631. A similar study conducted by Bernotaite and Malinauskiene (2017) in Kaunas, Lithuania, observed significant high psychological distress among Kaunas teachers but Psy-Abuse was not significantly associated with Psy-Res (−7.3%) and Depersonalization (−1.9%) based on significance values greater than .05 (p = .096 and p = .661, respectively). Similarly, E-E and Depersonalization have reduced professional efficiencies (Martínez et al., 2020) as health-related problems in working life and create a delayed response to chronic emotional and interpersonal stressors that happen more than once in the workplace (Hillert et al., 2013). One study found that teachers have the highest burnout level as compared to other professions (Mojsa-Kaja et al., 2015). On the contrary, the study found a significantly negative relationship (p < .05) with E-E and P-A, with correlations values of −9.4% (p = .032) and −12.6% (p = .004), respectively. Furthermore, Psy-Res has significantly positive relationships with E-E (i.e., 39.7%, p = .000), P-A (i.e., 55.6%, p = .000), and Depersonalization (i.e., 42.3%, p = .000). Psy-Res is negatively associated with burnout. Another study found that a high level of Psy-Res leads to less burnout in the workplace, a negative association was observed between Psy-Res and Depersonalization, P-A, and E-E (Akgemci et al., 2013).
Initially, we used linear regression to investigate the relationship between Psy-Abuse and E-E (β = −0.185, p = .032) faced by female teachers. As a result, in the second step, regression analysis was performed to study the impact of Psy-Abuse on Psy-Res (β = −0.201, p = .096) and then in the third step, the direct impact of Psy-Res on E-E was investigated (β = 0.280, p = .000). Lastly, the total effect of Psy-Abuse on E-E with the intervening role of Psy-Abuse did not show a significant result (β = −0.129, p = .105). Afterward, the significant negative impact of Psy-Abuse on P-A was lessened by using the Baron and Kenny Mediation approach. Accordingly, Psy-Res, although it independently provided a positive supportive effect on P-A, it has an even more strong positive intervening role in the relationship between Psy-Abuse and P-A.
This confirms the presence of a partial mediation as Psy-Abuse emerges as a strong indirect predictor of P-A. This situation becomes serious when women are living in an abusive relationship at home. They are often observed with posttraumatic stress disorder, anxiety, and depressive disorders (Trevillion et al., 2012). They generally express themselves with low self-esteem and fatigue (Dichter et al., 2011). One study found that when domestic Psy-Abuse occurs among working women experiencing burnout, it multiplies their feelings of depression, stress, and anxiety (Saberi et al., 2019), and they exhibit low levels of self-efficacy and burn out more easily. In addition, the mediation effect of Psy-Res on Depersonalization is slightly negative as compared to the absence of the intervention of Psy-Res, which showed a negative yet higher influence on P-A. A negative linear association between Psy-Res and burnout has been observed in professional academics (Akgemci et al., 2013). Conversely, dispositional mindfulness shows moderate effects on the relationship between psychological risk factors and mental health (Daubenmier et al., 2014). Furthermore, a study conducted by Anasori et al. (2020) demonstrated that workplace bullying significantly predicted E-E, and resilience and psychological distress partially mediated this relationship. Undoubtedly, Psy-Abuse has a negative relationship (−9%) with E-E. Similarly, Psy-Abuse also has negative associations with Depersonalization (r = −2%), but Psy-Res has a substantial positive association with psychological resilience. According to Trevillion et al. (2012), the majority of women who live in an abusive relationship are depressed and stressed. Similar experiences have been observed by Ran et al. (2020). Psy-Res is described as the capability of responding to extreme stress or unfavorable experience and subsequently as the ability to bounce back from damaging emotional experiences (Tugade et al., 2004). Moreover, Psy-Res has a negative relationship with the Depersonalization of working women (Yonezawa et al., 2011).
LIMITATIONS OF THE STUDY
The study is limited to married teachers working in the education sector. Future researchers should conduct a comparative study between working and nonworking married women. Moreover, other work environments should also be considered for study, including banks, the health sector, higher education, and any corporate sector.
CONCLUSION
Extreme psychological bullying is considered to be a strong predictor of psychological distress, but burnout did not affect these associations (Berotaite & Malinaukiene, 2017). The findings from this study offer insight into the scope and significance of psychological abuse and its impact on burnout in female married employees, and teachers in particular. These findings provide public and private sector policymakers suggestions for remedial efforts to overcome burnout that affects womens' work performance, as more than half of the population of developing countries such as Pakistan are women. This study seeks to improve women's contributions in the workplace and in their overall efforts to support economic uplift to these nations.
Data Source: WHO (2005) and European Union Agency for Fundamental Rights (2014)
Theoretical Framework
Barron and Kenny (1986) Mediation Analysis Considering the Intervening Role of Psy-Res on the Relationship between Psy-Abuse and E-E
Barron and Kenny (1986) Mediation Analysis Considering the Intervening Role of Psy-Res on the Relationship between Psy-Abuse and P-A
Barron and Kenny (1986) Mediation Analysis Considering the Intervening Role of Psy-Res on the Relationship between Psy-Abuse and Depersonalization
Contributor Notes
FARHAT MUNIR is an assistant professor in the School of Social Science & Humanities, Department of Education, University of Management and Technology, Lahore, Pakistan. She is a passionate teacher and a researcher who is always looking for new areas of research and bringing reliable knowledge to contribute to transformation in the education sector by making comparative analysis of the findings. She is specialized in Educational Leadership and Management and engaged in multiple innovative research projects. She is keen in utilizing the research findings not only in her teaching but also in extending research and training activities. https://orcid.org/0000-0002-7998-3374. Email: zershawaiz@yahoo.com
ALEENA SHUJA is a lecturer based in the Hasan Murad School of Management at the University of Management and Technology, Lahore, Pakistan. She completed her master's of science degree in Management Sciences (MS) from COMSATS University, Islamabad, Lahore Campus, in July 2015. Her role revolves around teaching, supervising master's and MS projects and theses, batch advising, and research publication.
IQRA SAEED is a lecturer in the School of Media and Communication Studies, University of Management and Technology, Lahore, Pakistan. She completed her master's of science degree in Communication Studies with a research specialization from the University of Punjab, and she also PhD at the Lahore College for Women University. Her role revolves around teaching, supervising master's and MS theses, batch advising, and research publication.
