1. Background
From the perspective of sociology, any attempt to explain the reasons for divorce and marital incompatibility requires consideration of the nature of marriage as an organization within the sociocultural context (1). Couples have different needs, desires, and expectations from marriage depending on their own or spouse’s characteristics and the society in which they live. According to these expectations and attitudes, the decision to continue married life is made (2). The factors influencing the intention to file a divorce petition among couples include frequent conflicts, betrayal, weak love, and poor commitment to marital life (3). Based on research findings, the level of families’ interference in couples’ life, the difference in the status of couples, and the positive perception of the consequences of divorce directly affect the rate of divorce, among which the level of families’ inappropriate interference and the degree of difference in couples’ status have the greatest impact on their intention to file a divorce petition (4, 5).
The causes of increased intention to divorce also include distorted perception and irrational thinking of couples, that is, what makes marital relationships more disturbed is not the stimuli and external events, but the couples’ own way of thinking and irrational beliefs about events. Irrational beliefs refer to aims and intentions that turn into obliging priorities and then the definite goals, hence if remain unsatisfied, they lead to anxiety (6). To investigate effective factors on marriage failure and divorce, various educational and counseling models and theories are suggested (7, 8).
The theory of planned behavior (TPB) can be useful to analyze beliefs, values, and attitudes underlying the divorce intention behavior due to including the constructs that address important dimensions of predicting behavior. The TPB, proposed by Ajzen, is widely used in various studies, and meta-analyses show that the constructs of the TPB can predict behavior and post-behavior intentions (9, 10). The most important determinant of behavior is intention. Due to the roles of the constructs of this model in explaining certain social problems such as drug addiction, internet addiction, and condom use (11-14), it also seems useful to analyze effective factors on the intention to file a divorce petition.
2. Objectives
In the current study setting that was a counseling center in Yazd, Iran, to which the divorce applicants are referred by the Yazd Justice Department for divorce proceedings, a questionnaire was completed by the applicants to measure the effect of the three constructs of the TPB, including attitude, behavioral control, and subjective norms, on the applicants’ intention to file divorce petitions.
3. Methods
3.1. Research Design and Participants
The current cross sectional study was conducted on all divorce applicants referred to the counseling centers by the Judicature in order to attend counseling courses and certain training programs and then obtain the authorization to legally finalize their divorce. Inclusion criterion was the interest of the subjects to enter the study. The exclusion criteria were lack of providing consent to participate in the study, severe addiction with several unsuccessful quitting attempts or any conditions diagnosed by the expert, with lack of ability to achieve compatibility or continue marital life. Sample size was determined 140, with 95% confidence interval (CI) and 80% test power.
The participants were randomly selected using random number table.
3.2. Data Collection Instruments
The instruments used included a questionnaire developed by directed content analysis on the transcripts of interviews with 27 divorce applicants referred to the counseling center in Yazd from 22 May 2017 to 23 August 2017 (not included in the present study).
According to the constructs of the TPB, the questionnaire was codified in two sections; the first section included items on demographic characteristics and the second one consisted of 48 items on the constructs of the TPB, 14 of which addressed attitudes toward divorce petition, 21 were related to perceived behavioral control, seven were about subjective norms, and six investigated behavior intention to file a divorce petition.
The validity of the questionnaire was investigated by both quantitative and qualitative content validation measures. For qualitative validation, the questionnaire was evaluated by 10 experts with adequate expertise and experience with the subject of the study. To quantitatively estimate content validity, content validity ratio (CVR) and content validity index (CVI) were used with the help of an expert panel. To estimate the reliability of the questionnaire, Cronbach’s alpha coefficient was used. Alpha coefficients of the attitude, subjective norms, perceived behavioral control, and behavioral intention constructs were 0.82, 0.73, 0.84, and 0.93, respectively. The items of all four constructs are rated based on a five-point Likert scale; hence, the score on each item ranged 1 (absolutely disagree), 2 (disagree), 3 (no idea), 4 (agree), to 5 (absolutely agree). Certain items are scored inversely. The lower the total score, the less the tendency to divorce (the stages of that qualitative study and the different stages of the questionnaire development and validation are already published) (15, 16).
Since the education levels of the respondents were different and in order to ensure that all of them would interpret the items the same way, the questionnaire was filled out by help of the researcher in a quiet place in the counseling center within 30 minutes and no one else was present.
3.3. Statistical Analyses
The mean and percentage of the scores on the constructs of the TPB were analyzed using parametric tests including t test, ANOVA, and Chi-square with SPSS version 24. Then structural equation modeling (SEM), covariance based with AMOS version 24, P ≤ 0.05, and goodness-of-fit indices were used. P value < 0.05 was considered as the level of significance.
3.4. Ethical Considerations:
The protocol of the study was approved by the Research Ethics Committee of Shahid Sadoughi University of Medical Sciences, Yazd (ethical code: IR.SSU.SPH.REC.1395.43).
4. Results
4.1. Characteristics of Participants
The majority (n = 89; 64%) of the participants were female. The mean age of the participants was 29.97 ± 7.48 years.
The model’s overall goodness-of-fit was good (Figure 1).
Figure 1.
Path analysis of intention to file a divorce petition
In Figure 1, the structural equation model is illustrated. There are three observed variables (perceived behavioral control, attitude, and subjective norm) in this model, shown in rectangles, and also one latent variable. In addition, the structural equation model includes one measurement model, that is, the intention measured by five items and its factor loading are all in the optimal range (greater than 0.5), with a structural model showing the effect of exogenous variables on intention. Among the studied variables, attitude had the highest effect on intention (0.69) and subjective norms had the lowest effect (0.13). The effects of these variables were positive, that is, with increasing them, the tendency to divorce increased. Overall, independent variables could account for 58% of the variance in intention.
Table 1 provides a brief summary of the results.
Table 1.Results of Path Analysis
| Index | Hypothesis Model Values | Standard and Determination Result |
|---|---|---|
| Goodness-of-fit index | 0.959 | > 0.90, acceptable |
| Comparative-fit-index | 0.995 | > 0.95, acceptable |
| Normed-fit-index | 0.958 | > 0.90, acceptable |
| Root mean-square error of approximation | 0.031 | < 0.08, acceptable |
| Normed chi-square | 1.131 | < 3 |
The study also examined the correlation between gender and marriage length with the scores on the TPB constructs by independent t test and that of other demographic characteristics with the constructs by one-way ANOVA. Results of Tukey’s post-hoc tests revealed that the mean score of the subjective norms construct in participants under 25 years was higher than those of in the age groups 25 - 30 and above 30 years (P < 0.05). For the divorce intention construct, the mean score of participants under 25 years were higher than those of the other two age groups.
The mean score of the perceived behavioral control was higher in the subjects that had no child than in the ones with children (P < 0.05).
For the subjective norms of divorce, the participants with lower education levels attained higher scores compared to the ones with higher education levels (P < 0.05).
Table 2 presents a summary of the demographic results regarding attitude, intention, subjective norms and perceived behavioral control.
Table 2.Distribution of TPB Construct Scores by Demographic Characteristics of Participants
| Item/Categories | Attitude to Divorce | Intention to Divorce | Subjective, Norm to Divorce | Perceived Behavioral, Control to Divorce | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean ± SD | t | P Valuea | Mean | t | P Valuea | Mean ± SD | t | P Valuea | Mean ± SD | t | P Valuea | |
| Gender | 3.742 | 0.000 | 4.509 | 0.000 | 1.672 | 0.097 | 0.284 | 0.777 | ||||
| Male | 40.53 ± 7.744 | 16.12 ± 4.713 | 21.82 ± 3.892 | 63.33 ± 9.028 | ||||||||
| Female | 45.48 ± 7.419 | 19.79 ± 4.488 | 22.92 ± 3.650 | 63.73 ± 7.273 | ||||||||
| Duration of marriage, y | 6.623 | 0.011 | 0.375 | 0.541 | 4.639 | 0.033 | 3.357 | 0.069 | ||||
| < 5 | 42.20 ± 5.33 | 18.77 ± 5.11 | 20.25 ± 2.48 | 60.42 ± 8.21 | ||||||||
| > 6 | 44.56 ± 5.52 | 18.27 4.57± | 19.30 ± 2.68 | 57.90 ± 8.08 | ||||||||
| Mean ± SD | F | P Valueb | Mean ± SD | F | P Valueb | Mean ± SD | F | P Valueb | Mean ± SD | F | P Valueb | |
| Age, y | 1.759 | 0.176 | 4.068 | 0.019 | 2.173 | 0.118 | 6.913 | 0.001 | ||||
| < 25 | 43.06 ± 5.35 | 20.20 ± 5.27 | 19.31 ± 2.51 | 60.36 ± 8.22 | ||||||||
| 26 - 30 | 42.18 ± 5.35 | 17.66 ± 4.61 | 20.54 ± 2.41 | 62.45 ± 8.31 | ||||||||
| > 31 | 43.42 ± 5.54 | 17.79 ± 4.37 | 19.66 ± 2.75 | 56.52 ± 7.40 | ||||||||
| Education level | 0.497 | 0.610 | 0.416 | 0.661 | 4.465 | 0.013 | 0.283 | 0.754 | ||||
| Below high school | 44.36 ± 8.5058.010 | 19 ± 5.314 | 23.93 ± 5.532 | 62.83 ± 8.275 | ||||||||
| High school | 42.85 ± | 18.07 ± 4.871 | 22.04 ± 2.172 | 64.04 ± 8.618 | ||||||||
| Higher education | 44.05 ± 7.169 | 18.39 ± 4.731 | 21.77 ± 2.827 | 63.75 ± 6.745 | ||||||||
| Number of children | 0.172 | 0.842 | 0.834 | 0.437 | 1.230 | 0.295 | 4.066 | 0.019 | ||||
| 0 | 43.22 ± 7.121 | 19.09 ± 5.124 | 22.02 ± 4.228 | 65.74 ± 8.188 | ||||||||
| 1 | 43.88 ± 8.050 | 18.10 ± 4.836 | 22.54 ± 2.420 | 61.46 ± | ||||||||
| > 2 | 44.12 ± 8.852 | 17.90 ± 4.810 | 23.22 ± 4.126 | 62.66 7.308± 7.565 | ||||||||
aIndependent samples t test.
bOne-way ANOVA.
5. Discussion
The present study was conducted to investigate the determinants of divorce intention using the TPB. The results of SEM revealed that the constructs of the TPB could explain 58% of the variance in intention as a variable. With regards to the effect of the size of the index, the R2 coefficient was large. The impact of attitude to divorce on intention to divorce is demonstrated. For example, some studies show that an individual’s initial attitude toward divorce affects the probability of divorce during his/her married life (17-19). To answer the question of why women who file divorce petitions more frequently than men in Iranian society have adopted a different attitude toward divorce in the recent years. Despite the fact that divorce is discouraged according to the religious teachings and moral principles of the society, it can be argued that the values related to the maintenance and stability of the family with respect to the fundamental functions of the family undergo a transformation under the influence of globalization and modernization, and eventually influence the change in attitudes toward divorce (20, 21).
As confirmed by the results of SEM, and measured by structural coefficient, the effects of the attitude factor was greater than the total effects of the perceived behavioral control and subjective norms. Considering the effects of attitude on satisfaction (8, 22), it seems logical that after a change in attitude toward the divorce and assessment of the advantages and disadvantages of doing this behavior, attitude has a great effect on the intention to divorce.
The current study results showed that perceived behavioral control and subjective norms were comparatively weaker determinants of behavior intention than the attitude to the divorce behavior. Consistent with other studies, the current study found that in some cases, the influence of others, especially parents, was effective on the decision of the couples about getting a divorce, and is considered as one of the main causes of divorce (23).
Regarding the perceived behavioral control construct, in the current study, no pronounced impact was observed for this variable. It can be argued that, since another study also show, the level of control of individuals on decision making can be influenced by various factors. The control of individuals on decision making can be influenced by various factors that in many cases are beyond the control of the individual such as social trauma and economic and personality problems of the spouse (24).
Consistent with other studies (7) on the intention to divorce, although women comprised the majority of divorce applicants, their mean score of intention to divorce was lower. It can be argued that while they file divorce petitions more frequently, they simultaneously seek to maintain their marriage. Consequently, it is usually women who are eventually forced to sue for the divorce, if they fail to tolerate the situation.
There was no significant difference in the mean scores of the behavioral control and subjective norms constructs between women and men. That is, as observed in some other studies, it is likely that both groups face certain conditions that are beyond their control or are exposed to the common pressures and beliefs of the society (subjective norms) about divorce (24).
As the current study results showed, more than half of the participants were under 30 years, that is, as observed in some other studies divorce applicants were young on average (18). Moreover, among the age groups, the mean scores of different constructs, except for perceived behavioral control, were not significantly different, since as people get older, both the likelihood of having a child increases and the likelihood of taking risk decreases. However, currently in the Western societies, some events are observed, namely gray divorce, indicating an increase in the age of the applicants for divorce (25).
Only 30% of divorce applicants had academic education. The effect of education level on decrease in filing divorce petition is also observed in previous studies. For example, a study observed that most of divorce applicants had low educational level (4, 26).
Significant differences were observed in the scores of subjective norms as the education level decreased, that is, the higher the educational level, the less the effects of others and social environment in increasing divorce petition filing (4). It seems reasonable that higher education levels increase thinking power, problem solving skills, and prudence, and therefore decrease the intention to divorce.
The behavioral control construct was not a strong predictor of divorce intention in the current study, however, the subjects that had no children exhibited higher levels of behavioral control to file divorce petition. As observed in some other studies (27), one of the important factors beyond the control of the individual and effective on the individual’s intention to change behavior and refuse to file a divorce petition is the possession of a child or children and concerns about their future may affect the decision making of the individual. It was also observed in the current study that the marriage length was effective on the intention to file divorce petitions. As the national official statistics show, divorce mostly occurs in the first five years of marriage (28).
5.1. Limitations
The present study had a cross sectional design. Therefore, the findings should be generalized with caution, or the replication of this study in a larger number of samples may yield more conclusive results. The study was conducted on people experiencing a very tense situation and, therefore, it was difficult for them to cooperate with the study, which could have influenced the results. The current study did not obtain a model, but the model was tested; however, surely using the model in other target population needs further testing of the model.
5.2. Conclusions
The most predictive value with respect to behavior intention to file divorce petition was obtained for attitude toward divorce. It is hoped that the counseling centers, by working with the clients on their attitudes, enable them to think more and contemplate when they are going to make an important decision. It is suggested that educational interventions in this field should be considered effective structures according to TPB.
