Document Type : Research/Original/Regular
Authors
1 Counselling, Faculty of education & psychology, Alzahra university, Tehran, Iran.
2 Professor,Psychology Dept.,The faculty of Educational Sciences, Alzahra University, Tehran, Iran
Abstract
The current study aimed to determine the psychometric properties of the attachment behaviors scale in Iranian married people. The research method was correlational, and the statistical population is all married people in Iran. According to the research problem, the researchers have determined their target population as married people among the university centers of Tehran, which among whom (from staff and students) 271 individuals (156 women and 115 men) were selected through convenience sampling from Imam Hussein University Center of Strategic Studies and Alzahra University dormitory. The research instruments were the Attachment Behaviors Scale by Sandberg, Novak, Davis, & Busby (2012) and the Four ENRICH Couple Scales by Fowers and Olson (1989). Data were analyzed using confirmatory factor analysis. Results showed that Attachment Behaviors-Self and Attachment Behaviors-Partner were two distinct factors. The confirmatory factor fitness indicators except for RMSEA of Self subscale were in the accepted range. Cronbach's alpha coefficients for Attachment Behaviors-Self and Attachment Behaviors-Partner were respectively.71 and .82. Considering the psychometric properties of the attachment behavior scale and its focus on objective behaviors in married relationships, using this scale in research work and planning more accurate and efficient interventions to strengthen secure attachment bonds in married people is recommended to family counselors and psychologists.
Highlights
This study aimed to investigate the psychometric properties of the Brief Accessibility, Responsiveness, and Engagement (BARE) scale among Iranian married people. The confirmatory factor analysis showed that the subscales of Self-Attachment Behaviors and Partner-Attachment Behaviors are two separate constructs consistent with Starnberg's research (Sandberg et al., 2012). Model fit indices showed that both subscales had a good fit. The results of correlation analysis showed that the BARE scale has good convergent and divergent validity. Examination of the internal consistency of the subscales using Cronbach's alpha and composite reliability showed that their reliability is appropriate.
In general, the present study results showed that the BARE scale has good validity and reliability in Iranian married people living in Tehran and can be used to examine Self and Partner Attachment Behaviors (availability, responsiveness, and emotional engagement behaviors). Since Attachment Behaviors showed a high correlation with marital satisfaction, the present scale can be used for future research purposes and planning more accurate therapeutic interventions in married people. Since this validation is a prelude to national validation, it is suggested that the random sampling method be used in different cities of Iran in future research.
Keywords
Main Subjects
Extended Abstract
Background and purpose
Numerous studies have shown that having an attachment figure in adulthood can increase security, calm, and positive emotions, leading to a better relationship between couples (Allen, 2018). When a couple is anxious and distressed, their spouse should be physically, emotionally, and psychologically available and responsive to their needs. These moments lead to forming new attachment bonds or strengthening the existing attachment bond, known as emotional engagement, and the third key element that can be measured in couples' relationships (Johnson, 2011).
Some researchers use attachment interviews to assess the concept of attachment, and others believe in using self-report tools such as questionnaires. These existing measures have focused more on identifying and distinguishing secure, anxious, and avoidant styles than measuring attachment behaviors (Sandberg et al., 2016). As a result, Sandberg et al. (2012) developed the Brief Accessibility, Responsiveness, and Engagement (BARE) scale to evaluate attachment behaviors. Attachment behaviors are associated with higher marital satisfaction and stronger relationship and attachment bonds. Attachment Behaviors also moderated the impact of destructive variables on relationships (Knapp et al., 2015). Sandberg et al. (2017) showed that attachment behaviors explain more variance of marital satisfaction than attachment styles.
Due to the significant relationship between marriage quality with physical and mental health, increasing marriage quality across the world, including Iran, has become a significant concern. Focusing on attachment behaviors can help therapists more precisely and accurately identify behaviors that lead to increased attachment security and develop more effective interventions to improve them. Since in the studies conducted, no suitable tool was found to measure attachment behaviors in Iran, this study aimed to investigate the psychometric properties of the Brief Accessibility, Responsiveness, and Engagement (BARE) scale among Iranian married people.
Method
The research method was correlational, and the statistical population was all married people in Iran. According to the research problems, the researchers determined their target population as married people from universities in Tehran. Among the target population, 282 individuals (161 women and 121 men) were selected through convenience sampling from Imam Hussein University Center of Strategic Studies and Alzahra University. To estimate the sample size, at least 10 participants per questionnaire item were considered (Boateng et al. 2018), and the following instruments were used to investigate the psychometric properties of the BARE scale.
The Brief Accessibility, Responsiveness, and Engagement (BARE): It is a self-report, short and systematic scale that measures attachment behaviors in couple's relationships on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). This scale was developed by Sandberg et al. (2012) and included 12 items. Six items measure Self-Attachment Behaviors, and the remaining six items evaluate the perceived Partner-Attachment Behaviors. In Sandberg et al. (2012, 2016), BARE showed appropriate reliability and validity in normal and clinical samples. In the present study, Cronbach's alpha coefficient for Self-Attachment Behaviors and Partner-Attachment Behaviors were .71 and .82, respectively.
ENRICH Couple Scales: Developed by Fowers & Olson (1989), this measure has 35 items that examine marital satisfaction. This instrument includes four subscales, Marital Satisfaction, Communication, Conflict Resolution, and Idealistic Distortion. In the present study, Cronbach's alpha coefficients were .85, .87, .72, and .85, respectively.
Procedure. With the permission of Dr. Sandberg, BARE was translated into Persian using the Brislin method. The questionnaire was distributed among 30 people who were not part of the original sample to preliminary evaluate the psychometric properties and assess its reliability through another sample. Preliminary results indicated that the reliability of the subscales and the factor loadings of the items were appropriate. After distributing the scale between the main research participants, the data were analyzed using SPSS and AMOS software.
Results
The mean age of the participants was 36.21 (±10.65), and their age range was from 19 to 66. The participants had different levels of education. Missing data was less than 1%, which was filled by the series mean imputation method. After checking the normality of the distribution, the confirmatory factor analysis was used to examine the fit indices.
Table 1.
Results of confirmatory factor analysis for measurement models of BARE scale
Subscale |
CR |
AVE |
c2/df |
GFI |
AGFA |
CFI |
RMSEA |
NFA |
SAB |
.78 |
.52 |
3.190 |
.982 |
.925 |
.960 |
.09 |
.945 |
PAB |
.89 |
.72 |
.529 |
.995 |
.987 |
1 |
.000 |
.990 |
Threshold |
> .7 |
> .5 |
< 5 |
> .9 |
> .9 |
> .9 |
0.03-0.08 |
> .9 |
Note. SAB = Self Attachment Behaviors; PAB = Partner-Attachment Behaviors
Based on the results presented in Table 1, both BARE subscales had acceptable internal consistency. Also, considering that the AVE's was higher than .50 and lower than CR's, BARE subscales had good convergent validity. Additionally, the correlation between BARE subscales and Enrich subscales was calculated. The results of the Pearson correlation test showed that all correlations are positive and significant and are in the range of .45 to .75. Therefore, the discriminant validity of BARE was confirmed.
Conclusion
This study aimed to investigate the psychometric properties of the Brief Accessibility, Responsiveness, and Engagement (BARE) scale among Iranian married people. The confirmatory factor analysis showed that the subscales of Self-Attachment Behaviors and Partner-Attachment Behaviors are two separate constructs consistent with Starnberg's research (Sandberg et al., 2012). Model fit indices showed that both subscales had a good fit. The results of correlation analysis showed that the BARE scale has good convergent and divergent validity. Examination of the internal consistency of the subscales using Cronbach's alpha and composite reliability showed that their reliability is appropriate.
In general, the present study results showed that the BARE scale has good validity and reliability in Iranian married people living in Tehran and can be used to examine Self and Partner Attachment Behaviors (availability, responsiveness, and emotional engagement behaviors). Since Attachment Behaviors showed a high correlation with marital satisfaction, the present scale can be used for future research purposes and planning more accurate therapeutic interventions in married people. Since this validation is a prelude to national validation, it is suggested that the random sampling method be used in different cities of Iran in future research.
Ethical Considerations
Compliance with ethical guidelines
Participants were not required to write their first and last names. They became aware of the objectives of the research and were free to leave the study.
Funding: The present study was conducted at the personal expense of the authors.
Authors' contribution: Seyedeh Monavar Yazdi and Fatemeh Hashemi Golpayegani have been responsible for designing, conceptualizing, assessing the content validity, data collection, data analysis, and final reviewing of the present article.
Conflict of interest: The authors declare that there is no conflict of interest.
Acknowledgments: We thank Dr. Sandberg for allowing us to validate the BARE scale and all the participants who cooperated sincerely with us.