Differentiation of self (DoS) is a developmental construct the design of which relies on the Bowen’s family systems theory and included both intrapersonal and interpersonal dimensions. The DoS represents the capability of an individual to view him/herself apart from his/her family of origin (1, 2). Licht and Chabot (2006) described DoS as a compound-complex concept of unifying the development of individuation with the ability to distinguish between thoughts and emotions (3). Recent views and multiple research types have been trying to clarify the nature of DoS. Several types of study have been conducted on various samples regarding DoS as a developmental construct (4).

The Differentiation of Self-Inventory (DSI) questionnaire involves several dimensions or aspects of differentiation that focuses one's attention specifically on adults, their important relationships, and common relations with the family of origin. According to the evidence, it was shown that DSI scores indicate less emotional reactivity, break up, assimilation with others, as well as greater strength to obtain an “I-position”, and anticipate lower intensive anxiety, better psychological conformity, and greater marital satisfaction. In line with the Bowen’s theory, along with the potential company of the DSI for testing the Bowen’s theory, other results argued this theory as a clinical evaluation tool and an index of the psycho-therapeutic results (5).

The authors performed an online survey of university students. Accordingly, it was observed that the differentiation of self-medicated the effects of stress to comprehend stressful incidents, which is further contributed to the mental and emotional state of a person stress (6). Self-differentiation as an extra factor or circumstance had a significant effect on psychological health and occupational anxiety in a study conducted in Philippines that is also consistent with the findings of a study performed in the United States (7).

Junho and Sangmoo observed the mediating effect of DoS on the family of origin experiences and marital satisfaction. Empirical data have provided support for Bowen’s assertions (8). For example, the results of recent studies have shown significant statistical relationships between lower scores of DoS on the one hand and high levels of anxiety on the other hand (9-10). Differentiation is defined in a continuum in which one can feel strong in unfavorable conditions and maintain strong emotional ties (i.e., high differentiation). At the other end (i.e., low differentiation), the individual puts himself in situations that cause anxiety and emotional attachment to others in addition to overcoming or merging with others (11).

Multiple types of study were conducted in different fields of study with regard to DoS, such as the Bowen’s family systems model of generalized anxiety disorder and romantic relationship distress (12), relationship between self-differentiation and marital satisfaction in mutually compatible and incompatible spouses (13), effect of self-differentiation training on the level of marital satisfaction among married female students (4), relation between DoS and marital satisfaction, and correlation of self-differentiation and optimism with responsibility and marital satisfaction in nursing students (14).

Individuals with a high level of differentiation cope with life tensions better than others. In addition, they are more adaptable and flexible, act free from emotionality, and are less affected by it. A high level of self-differentiation is related to marital satisfaction, stable relationships with the opposite gender, and effective parent-child communication (15). According to the literature, the studies in the field of intervention have been rarely performed, which indicates the necessity of performing studies on the self and marital interventions. Therefore, the present study aimed to examine the effectiveness of Bowen's self-differentiation in self-differentiation and marital satisfaction.

Materials and Methods

This quasi-experimental was carried out using a pretest-posttest design. The study population (n=24) consisted of the women who referred to counseling centers in Tehran in 2018. The subjects were selected by the convenience sampling method and randomly assigned into two intervention (n=12) and control (n=12) groups. In this study, interviews were conducted with a number of volunteer married women who were willing to attend training sessions. Written informed consent was obtained from all the participants before enrollment, and confidentiality and anonymity were followed regarding the responses. Furthermore, the participants were informed of the purpose and design of the study, as well as the fact that the participation was voluntary. Moreover, they had a right to withdraw from the study at any time.

In this study, the counseling was based on Bowen’s self-differentiation performed in the form of group counseling in ten 90-120-minute sessions twice a week in the experimental group, which was considered an independent variable. After attending the training sessions, the questionnaires were answered, and about 1 week later, they were presented again. The results were analyzed following the above-mentioned process.

This study was conducted in accordance with the ethical standards of the Declaration of Helsinki (1975), revised in 2000. The protocol of this study was approved by the Ethics Committee of Shahid Beheshti University (Reference no.: REC.1397.168). Descriptive statistics and inferential statistics (i.e., multivariate analysis of covariance [MANCOVA]) were used in this study. Table 1 tabulates the context of the training sessions based on the Bowen's theory.

The ENRICH Marital Satisfaction Scalehas two types, including 115 and 125 items, which identify difficult and gratifying backgrounds and determine the need for couple counseling. According to Elson (16), the reliability of this questionnaire was obtained at 0.86 after 4 weeks. In addition, its validity was confirmed by reporting the Cronbach's alpha correlation coefficient from 0.41-0.60 (17). Soleimanian (18) also obtained a Cronbach's alpha

correlation coefficient of 0.93. Since the scale of 115 items made subjects tired a short version has been developed. Therefore, after calculating the correlation coefficient of each question, 47 questions were chosen with a high correlation.

The reliability of the above-mentioned questionnaire was obtained rendering a Cronbach's alpha coefficient of 0.95. In the present study, the 47-item questionnaire was used with the internal validity of 0.92. The answers to each option were considered "Totally agree," "Agree", "Neither agree nor disagree", "Disagree", and "Totally disagree." The contribution of each of the items from the right to the left is 1, 2, 3, 4, and 5. 

The DSI-Revised (19) is probably the most popularly used and well-legislated measurement tool for self-differentiation. This inventory enables respondents to report their relationship with other members of their family. Reliability of this instrument has been confirmed in different studies, including those conducted by Skowron and Friedlander (5), as well as Rafi (20), reporting the Cronbach’s alpha coefficients of 0.88 and 90%, respectively. In the present study, Cronbach's alpha coefficient was obtained at 0.80.

Each item is based on a 6-point Likert scale from "Very good for me" to "It is never true for me". The share of each question from the right to the left is 1, 2, 3, 4, 5, and 6. The maximum score in this questionnaire is 276. The participants responded to the items in each sub-section on a 6-point Likert-type scale, ranging from one (Not at all true for me) to six (Very true for me). The DSI full-rating score was calculated by converting crude scores on all items in Emotional Reactivity, Emotional Cut-off, and Fusion with Others sub-scales, as well as one item on the I-Position sub-scale and adding them together. Accordingly, the higher scores showed higher levels of differentiation (e.g., less emotional reactivity, less difficulty in maintaining I-positions, less emotional cut-off, and less contradiction) (5).

Results

Mean scores of the participants’ age were 30.58±3.61 and 29.92±3.18 years in the experimental and control groups, respectively. Furthermore, both groups were homogeneous regarding the educational level. In the intervention group, 33.3%, 41.7%, and 25% of the subjects were reported with high school education, bachelor’s degree, and master’s degree, respectively. However, 25%, 50%, and 25% of the subjects had high school education, bachelor’s degree, and master’s degree in the control group, respectively.

Table 1 tabulates the descriptive statistical indices of the intervention and control groups before and after the test (Table 1).

As it can be observed, the mean in the post-test stage was higher than that in the pre-test. Based on the presented results in the Table1, the implementation of self-differentiation education increased the marital satisfaction of married women.

Some of the assumptions were evaluated before performing these tests. Table 1 tabulates the results of analyzing the assumptions and MANCOVA of the test group (Table 2).

The assessment of data normality was performed by the Kolmogorov-Smirnov test, and the results are presented in Table 2. Based on the results, the significance level of each of the

study variables was higher than 0.05, and the data of both variables were normal (P<0.05) (Table3).

As it can be observed, the equality of variances is assumed for all the study variables (P<0.05) (Table 4).

The results of the Levene’s test performed on the study variables showed that the variances of the dependent variable’s error were equal in the two groups (P<0.05). The standard method for assessing the equality of covariance matrices is the Box’s M test in which P-value less than 0.05 is considered an index of heterogeneity or inequality index.

Considering the presented assumptions, it can be observed that the study data were capable of entering into MANOVA and examining the differences of the dependent variables between the two groups (Table 5).

The results of Table 5 revealed that after the elimination of the pre-test effect with MANOVA, there was a significant effect on self-differentiation. The findings of the present study showed that there was a significant difference between marital satisfaction and self-differentiation variables of married women who attended self-differentiation training sessions and those of married women in the control group (Wilk's lambda=0.31; P<0.05).

This covariate is linearly relevant to the dependent variables, and its inclusion into the analysis can increase sufficiently to find the diversity between the groups of an independent variable. A one-way MANCOVA is applied to distinguish whether there are any statistically considerable differences between the adapted means of three or more independent (unrelated) groups, which are controlled for a continuous covariate (Table 6).

The results of Table 6 show that there was a significant difference between marital satisfaction and self-differentiation variables in the intervention group.

As it can be observed in Table 5, the significance level obtained for self-differentiation and marital satisfaction was smaller than the significance level (0.025)

obtained from Benfroni correction (dividing the significant level of 0.05 into two dependent variables). Consequently, regarding the obtained meanings, it can be said that the scores of self-differentiation and marital satisfaction increased in the intervention group, compared to those in the control group.

Discussion

The present study was conducted to investigate the effect of Bowen’s self-differentiation training on the self-differentiation and marital satisfaction in Tehran. As stated in the results, the scores of women’s self-differentiation and marital satisfaction in the interventions, increased significantly compared to those of the pre-test. These findings are consistent with the results of studies conducted by several authors (21-25). The findings of a study conducted by Marcos (26) are inconsistent with the results of the aforementioned studies, and no relationship was observed between self-differentiation and marital satisfaction.

In addition, Yousefi et al. (2010) approved that the Bowen’s theory is appropriate to be practical in the Iranian community, and self-differentiation is in tune with mental health and marital quality (27). Zarei et al. (2011) showed that self-differentiation, confidence, and prudence have a considerable relationship with marital relationships (28). To explain this finding, according to Bowen, the concept of self-differentiation is useful for individuality and togetherness (2).

A more differentiated individual would be able to communicate with others while having a solid sense of self. This solid sense of self is characterized by individual identity, personal beliefs and opinions, as well as a strong conviction of what should be done or not done based on their own life experiences and reasoning (1). Differentiated individuals have a solid sense of self, react sedately in emotional situations, and make decisions based on their own beliefs and intellect (29). They are not overwhelmed by emotions and are able to take I-positions. This entails being able to own their thoughts and feelings and state what they believe without needing to conform to others or please them. Furthermore, those who are differentiated are not necessarily detached from emotions and are able to have relationships with appropriate boundaries (30).

On the other hand, the feelings of comfort and closeness in relationships can turn into anxiety and discomfort as the self of an individual is incorporated into the self of another individual (31) that results in hostility and possible rejection. This is what Bowen described as being undifferentiated. An undifferentiated person is considered someone who reacts emotionally and emotionally overwhelmed when triggered by external stimuli. They cut off interactions and intimacy when the relationship is too intense and are more likely to be faced with people with whom they have close relationships (32).

In this study, some limitations decreased interior and exterior credibility. Another limitation was the lack of men’s cooperation, which made this training less effective. Since the statistical population of this study were women referring to a counseling center in Tehran, the results could not be generalized with regard to individual, social, and intercultural differences. The final factor that limited the generalizability of the results was matching the age, gender, and educational level of the participants with the degree of self-differentiation. On the other hand, the strengths of this study were random sampling and homogenization of the groups to prevent intervening variables.

Based on the limitations observed in the present study, it is recommended to perform this study on men and compare their findings with those of the women. Since this study was carried out on one of the couples, it is suggested to conduct further studies on couples simultaneously. The present study was carried out on the subjects within the age range of 25-35 years, and it is recommended to perform the study again on different age groups. It is worthwhile to simultaneously compare this study with other methods and their findings.

Conclusion

The central concept of the Bowen’s theory is different in the family; this is what the DSI aims to measure. Differentiation in the family is created by knowing each other’s feeling that results in experiencing comfort and closeness. This theory is especially useful to improve an individual’s life through having an autonomous self and individuality, as well as the concept of connectedness and intimacy among couples (32).

Acknowledgements

This study was extracted from a master thesis. The authors would like to thank all the women who participated in this project. This study was conducted in agreement with the ethical standards of the Declaration of Helsinki (1975), revised in 2000. The protocol of this study was approved by the Ethics Committee of the University of Shahid Beheshti (Ref no.: REC.1397.168).

Conflicts of interest

      The authors declare no conflicts of interest.