The Relationship of the Current and Desired Number of Children with the Quality of Marital Relationship and Fertility Motivations Among Employed Women and their Husbands

In recent years, there have been significant changes in fertility; accordingly, fertility has been on a declining trend worldwide (1). Based on the statistics, Iran has experienced the fastest decline in fertility rates among other countries between 1995 and 2015 (2). In this regard, the fertility rate in Iran has been less than most Asian, European, and Latin American countries over the past three decades (1). The population growth rate has declined to 1.24% in the 2016 census (3), and Iran was ranked 147^th among 228 countries regarding fertility rates (4). This rank is next to those of the UK, Iceland, and Uruguay and equal to those obtained for the countries in Europe with the lowest population growth rates (4).

Some of the consequences of reduced fertility include the lack or decline of economic growth and development, population aging, social disadvantages, socialization issues and social communication disorders in children of small families, psychological problems of future generations, cost of elder care, weakness of the country's defense forces, decrease of the country's young and genius individuals, and even the decline of the human race (2). In addition, some causes of lower fertility rates in the world are an increase in the cost of living, increased marriage age, economic difficulties, one-child policy, and a decline in the tendency for childbearing (1).

Fertility, as a social reality, can be examined at the macro and micro levels. The macro level includes the impact of environmental, social, political, and cultural factors on fertility. On the other hand, the micro level consists of the influence of such internal factors as motivations, desires, and personal attitudes towards fertility behavior (5). According to previous studies, the most important factors are the fertility motivations which determine the couples' fertility ideals, tendencies, and behaviors (6).

Psychologists and behavioral scientists have always emphasized the role of motivation in explaining human behavior (7). The voluntary nature of fertility highlights the major role of motivation in this domain (8). Motivation is defined as an intrinsic force that induces or stimulates a certain behavior in individuals (9). According to Miller (2016), people can have different levels of positive and negative motivations (10). Miller in another study (1995) on fertility motivations in the US found that positive fertility motivation had a significant relationship with a higher desire for childbearing, higher number of children, and shorter birth intervals. In the aforementioned study, negative fertility motivations had a statistically negative relationship with the desire for having children and the desired number of children (7). Each of these fertility motivations has a significant effect on one's fertility behavior (11).

Nowadays, the increase of some environmental and social behaviors, such as individualism and individual independence, couple conflict, and divorce, has reduced the childbearing potential in Iran (1). In addition, the elevation of some other social factors, such as marriage age, education level, economic expectations (1), age of first-time mothers, knowledge of contraceptives, and marital relationship quality, can affect the fertility motivation and behavior of individuals. According to Sandford (2006), marital relationship quality is a multidimensional concept (13).

Pato et al. consider marital relationship quality in four dimensions, namely self-care, care for the spouse, communication styles, and problem-solving plans (13). Regarding the link between the marital relationship quality and childbearing, Rijken in a qualitative study (2008) in the Netherlands and Eshaghi (2014) in Iran showed that couples with a good relationship are more likely to have children since they believe that their relationship will provide a favorable environment for the upbringing of children. On the other hand, couples with a poor relationship may assume children as a factor for strengthening the couple bond and saving the marriage (3, 12). There are contradictory results regarding this domain. For instance, Kariman (2014) introduced the concerns over increased marital instability as one of the causes of doubt in childbearing (14).

No research has been performed on the importance of marital relationship quality and its effect on childbearing in Iran. The childbearing decision is made collectively by the couples. However, there is dearth of knowledge and research regarding the role of men in fertility and quality of marital relationship. Accordingly, it is required to study the association of marital relationship quality, fertility motivations, and the associated effective factors. Moreover, regarding the policies announced by the supreme leader on the importance of fertility and childbearing, special attention must be paid to this issue. Therefore, the present study aimed to determine the relationship of the current and desired number of children with marital relationship quality and fertility motivation among employed women and their spouses in Mashhad, Iran, in 2017.

Materials and Methods

This correlational study was performed on 270 couples (i.e., a total of 540 subjects) in Mashhad in 2017-2018. The inclusion criteria consisted of being: 1) female, 2) employed, 3) married, 4) Iranian, 5) resident of Mashhad, 6) mother of ≥ 2 children, and 7) married to a literate person without any history of mental illnesses and addiction. On the other hand, the exclusion criteria were incomplete response to the questionnaire and withdrawal from the study.

The research settings were divided into three categories, namely healthcare (Omid and Ghaem hospitals and healthcare centers No. 5 and No. 2), cultural-educational (10 private and public schools and five universities), and administrative-financial (10 public and private central offices). For sampling purposes, a list was prepared from the women employed in the service-provider category, service section of hospitals, public and private offices, factories, schools, and universities using cluster sampling method to select the subjects. However, given the dispersion of the self-employed women throughout the city, the convenience sampling method was used for this group. The sampling was calculated based on the meta-analysis article of Kaylen J. Fellows et al. (2015) (20) using the following sample size formula:

r =-20.0

n = (1.0+96.84/w)²+3

w=1/2log 1+ γ /1- γ

n= (2.8÷ 0.21)²+ 3=180

The estimated sample size was tripled due to the sampling method. Therefore, the final sample size was 540 subjects (270 couples), and the number of samples in educational-cultural, healthcare, and administrative-financial categories was calculated. Subsequently, sampling was completed using a random clustered sampling method and a random number table. In the healthcare, educational-cultural, administrative-financial, service provider, and self-employed categories, 72, 85, 37, 22, and 54 participants were included in the study, respectively. The overall dropout rates of the employed women and their spouses in all groups were 35 and 40, respectively. Therefore, the sampling was repeated again since it was not possible to change the sample size.

Data collection tools included demographic questionnaire form (52 questions), the work-family conflict scale (WFCS; 18 items) (Carlson et al., 2000), marital relationship quality questionnaire (33 items), and Miller's fertility motivation questionnaire (49 items divided into two parts evaluating positive (28 items) and negative (21 items) fertility motivations, each of which was scored separately).

Validity of the WFCS questionnaire, marital relationship quality questionnaire, and Miller's fertility motivation questionnaire were confirmed in studies performed by Motesharrei in Ahwaz (2012), Khoshkam (2006), and Arghavani et al. (2012) (21, 13, 17). Furthermore, the internal consistency method (Cronbach's alpha coefficient) was used to confirm the reliability of all the questionnaires for all of the employed women and their spouses. Accordingly, the reliability of all the questionnaires were confirmed with an alpha coefficient of < 0.90. After obtaining permission and informing the subjects about the study objective, they were provided with the research unit selection form so that they could participate in the study if they met the inclusion criteria and obtained the consent of their spouses.

Afterward, the female participants were given the research unit selection form for their spouses. They were instructed to have their spouse fill out the forms under a calm condition. If they were eligible, the other questionnaires related to employed women and their spouses were given to them in separate envelopes to be delivered to their husbands. Therefore, all of the questionnaires were completed at home. The female participants were called 24 h later to be reminded to complete the questionnaires. Subsequently, the questionnaires were collected. After the questionnaires were returned by the couples, the researcher checked them to see if all of the items were answered.

The dropout rates in the healthcare, educational-cultural, administrative-financial, service provider, and other categories were 10 females vs. 11 males, 8 females vs. 8 males, 3 females vs. 5 males, 2 females vs. 1 males, and 12 females vs. 15 males, respectively. These participants were excluded from the study since they did not answer all of the questions. It was impossible to change the sample size; therefore, the sampling was repeated to reach the desired sample size. The subjects were required to respond to all of the questions completely. Finally, the collected data were analyzed in SPSS software (version 24).

The demographic data were presented as mean and standard deviation. Moreover, Kolmogorov-Smirnov test was used to evaluate the normal distribution of the data. The data were analyzed using the Pearson and Spearman correlation coefficient, independent t-test, Mann-Whitney U test, one-way analysis of variance (ANOVA), least significant difference (LSD), Kruskal-Wallis test, Chi-square test, and regression. In all the tests, a p-value less than 0.05 was considered to be statistically significant. Ethical considerations were respected in the current study. In this regard, all of the information was kept confidential, and written consent was obtained from all participants.

Results

A total of 540 subjects participated in this study, 270 of whom were female and the rest were male. The mean ages of the employed women and their husbands were 34.46±6.61 and 38.15±6.97 years, respectively. Moreover, the mean ages at the time of marriage were 22.61±4.46 and 26.69±4.79 in the women and their husbands, respectively. The mean marriage duration of the couples was estimated at 134.3±81.06 months. The mean gravidity was 1.61±0.92, with the mean number of children of 1.34±0.72. Moreover, the desired number of children of employed women was 2.79±1.32. The employed women and their spouses had the mean work experience of 9.08±7.51 and 13.57±7.90 years and worked 43.04±17.28 and 41.57±23.84 h a week, respectively. Only one male participant was unemployed. A total of 196 (72.6%) and 166 (61.5%) employed women and their spouses had tertiary education, respectively.

Spearman correlation coefficient showed that the current number of children had a statistically positive correlation with the age and work experience, and marriage duration of the couples. On the other hand, this variable was negatively correlated with the age at marriage in the couples. Moreover, the desired number of children had a statistically positive relationship only with the husband's age (r=0.11, P=0.03; Table 1).

There was a statistically negative relationship between the current number of children and marital relationship quality (r=-0.19, P=0.002). On the other hand, this variable was positively correlated with positive motivations in working women (r=0.16, P=0.008). In addition, the desired number of children had a statistically positive association with the marital relationship quality and positive motivations among the couples (Table 2).

Regarding the correlation of the above-mentioned factors with on another and with the current number of children, the relationship between these factors was investigated through linear regression analysis. Based on the results shown in Table 3, this analysis explains 42% of the variance in the current number of children. According to these results, four variables, namely marriage duration (P<0.001), marital relationship quality (P<0.001), positive fertility motivation (P<0.001), and working hours per week (P=0.017) among the employed women significantly affected the current number of children. Marriage duration and positive fertility motivation in working women also had a significantly positive effect on the current number of children.

 Moreover, the working hours per week and the marital relationship quality of employed women were found to negatively affect the current number of children.

Regarding the correlation of the studied variables with the desired number of children, the relationship between them was investigated through linear regression analysis. The model predicts 7% of the variance in the desired number of children. According to the results, the two variables of positive fertility motivation in employed women (P<0.001) and husbands’ working hours per week (P=0.01) could significantly predict the desired number of children. According to the results shown in Table 4, positive fertility motivation has a statistically positive effect on the desired number of children. In addition, this variable was negatively affected by working hours per week.

Quality of marital relationship from the women’s point of view was divided into three categories of weak, moderate, and good. Moreover, the current and desired number of children were compared in three groups. Mean numbers of current children were 1.40, 1.47, and

1.12 in the group of women with poor, moderate, and good marital relationships, respectively. The results of ANOVA test showed that the mean difference of the current number of children in the three above-mentioned groups (poor, moderate, and good) was significant (f=6.73, P=0.001).

The LSD follow-up test showed that the number of children in women with good relationships was significantly lower than that in women with poor and moderate marital relationships. The numbers of children in the women with a good marital relationship were 0.28 and 0.35 lower than those in the group with poor (P=0.006) and moderate (P=0.001) marital relationships, respectively. The ANOVA test showed that the mean numbers of children in the employed women belonging to the three groups of poor (2.1±69.01), moderate (3.1±04.36), and good (2.66±1.52) relationships were not significantly different (P=0.124).

Discussion

The present study aimed to investigate the relationship of the current and desired number of children with marital relationship quality and positive/negative fertility motivations in employed women and their spouses in Mashhad. The mean gravidity was estimated at 1.61±0.9 which is lower than the total fertility rate in Iran

(1.7) and Khorasan Razavi province (1.9) (15). Desired number of children in employed women was obtained as 2.79±1.32. According to a study conducted by Hosseini and Begi in Mahabad (2012), about 70% of the women desired two children (16). Based on the results, the desired number of children in the employed women living in Mashhad is about 3; however, currently, they have 1.5 children.

The number of current children showed a statistically positive correlation with the age and marriage duration of the couples. However, this variable had a statistically negative correlation with the age at marriage and work experience of the couples. The results of linear regression analysis showed that the current number of children was positively correlated with marriage duration (P<0.001) and positive fertility motivation in employed women (P<0.001). Moreover, this variable showed a statistically negative relationship with the total score of marital relationship quality (P<0.001) and working hours per week in employed women (P=0.017).

Based on the results, it can be concluded that the increase in working hours of employed women had a relationship with the decreased number of children. This could be due to the economic pressure on employed women in assisting in the fulfillment of family’s financial needs, thereby impeding childbirth. Moreover, it can be said that the age and marriage duration of the couples had a statistically positive relationship with the current number of children, which is consistent with the findings of a study conducted by Arghavani et al. (2013) in Mashhad (17).

In Iran, the establishment of a family is closely associated with childbearing, which is often one of the main motivations of marriage. In Iranian societies, couples are expected to have a child shortly after their marriage. According to the results of linear regression, the comparison of the number of children in groups of employed women with poor, moderate, and good marital relationships showed that the number of children was lower in the women with good marital relationships than in those with poor and moderate marital relationships. It is probably due to the fact that the couples with fewer children are more likely to spend time together and are more satisfied with their relationships. The direction of causality in this domain can be investigated through the implementation of longitudinal studies.

Desired number of children was positively correlated only with the age of the husband (P=0.03). This is inconsistent with the findings of obtained by Rijkin et al. (2009), which revealed that the age of the husband had no effect on future fertility in the Netherlands (12). As the results of regression analysis indicated, positive fertility motivation in employed women (P<0.001) and husband’s working hours per week (P=0.01) could significantly predict the desired number of children. Interpretation of these findings suggests that by increasing one standard deviation in the score of positive fertility motivation, the desired number of children will increase by 0.23 standard deviation. Moreover, if the standard deviation increases in the husband's working hours, the desired number of children will decrease by 0.16 standard deviation.

Marital relationship quality in the present study had no significant relationship with the desired number of children. Rashid (2006) found that women who had a romantic relationship with their husbands were more likely to have more children in order to increase the duration and quality of their relationship (18). However, the results of another study conducted by Claessens showed that unstable family conditions and poor relationships reduced fertility rates (1, 19). The relationship between reduced childbearing and increased spousal working hours in the current study may be due to men's longer and harder efforts to earn a living for the current number of children which makes them feel they cannot handle another child.

Results of a study conducted by Arghavani (2013) are consistent with those of the present study, regarding the significant relationship between positive fertility motivation and the increase of the desired number of children (17). Moreover, the findings of a study performed by Pezeshki et al. (2004) are also consistent with our results which could be due to the similarities in the social and cultural contexts of the two aforementioned studies and the present study. Furthermore, regarding the relationship between positive fertility motivations and the desired number of children, the findings of a study performed by Miller et al. are consistent with those of the present research (10). Spousal positive and negative motivations had no significant relationship with the desired number of children.

This study entails several research limitations, including individual differences, mental status, and accuracy of research units in answering questions, which could affect the responses of the research units. However, these issues could not be controlled by the researcher. Efforts were made to encourage the participants to fill out the questionnaires at a special time and place at home. However, the researcher was not informed of the conditions of the research units at the time of completing the questionnaire. The questionnaires were in form of self-report, and the researcher assumed that the participants' statements are accurate; nevertheless, there is no evidence regarding this issue. The results of the study are applicable only to the employed couples who possess the mentioned characteristics and cannot be generalized to housewives.

Conclusion

As the findings of the present study indicated, the current number of children had a statistically negative relationship with marital relationship quality. Moreover, this variable was positively associated with positive fertility motivation in the couples. Desired number of children showed a statistically positive relationship with marital relationship quality and positive motivations smong the couples. Accordingly, effort should be made to empower employed women to maintain and improve the quality of their marital relationships and reduce their working hours. Moreover, this group should be provided with workplace childcare facilities and the necessary support to increase their fertility motivation.

Results of the present study provide policymakers with useful information for revising and re-establishing the working hours of women. Moreover, they can empower working women with the aim of improving the quality of their marital relationships by stepping up to improve marital relationship quality and increase childbearing. This research could also serve as a basis for future studies on the barriers to childbearing and research on men as a group involved in fertility.

Acknowledgements

This article is derived from a thesis submitted by the second author in partial fulfillment of the requirements for the degree of Master of Science in Midwifery Counseling in 2017 (IR.MUMS.REC.1396.184). The authors would like to express their gratitude to the respected research assistant at Mashhad University of Medical Sciences who funded this study.

Conflicts of interest

Authors declared no conflicts of interest.