Couple relationship and transition to parenthood: Does workload at home matter?

Abstract

In this paper the association between the quality of couple relationships and household workload after the transition to parenthood is investigated. The sample consists of 251 men and women recruited through 21 child health‐care centres in South Bohuslän, Sweden. The results reveal a significant correlation between experiences of household workload and the quality of couple relationship for women when they have had both their first and their second child. The association clearly remains for first‐time mothers even when other characteristics, which may affect the couple's relationship, are included, such as stress, child's temperament, and social support. Stress also appears to have a significant connection with the satisfaction with the couple relationship, specifically for men. These results indicate that in a society with a high level of equality between men and women, household work and stress appear to be indicators of relationship happiness for persons who have recently become parents.

Introduction

The transition to parenthood is a psychologically very sensitive period in couples' lives. The relationship is exposed to a number of changes. As a result, the transition may affect the couple's relationship satisfaction, with a decline in marital satisfaction as a result. Belsky et al. (1985) found small yet reliable changes in a negative direction when couples became parents.

When evaluating the effects of decline it can, however, be difficult to distinguish normal decline, caused by time, from a decline due to the baby. During the first years of marriage, the relationship satisfaction is likely to decline to some degree (Kurdek 1993, McHale and Huston 1985), and it is also during these first years that most couples have their first children. Crohan (1996) studied newlywed couples over a period of three years after the marriage and included both those who became parents and those who remained childless. She found that marital satisfaction declined for parents and childless couples, but the decline was most pronounced for those couples that had become parents. Kurdek's (1993) study of marital relationships covered a period of four years and also included parents and childless couples. The results indicated that marital satisfaction declined in both groups, as did the expression of affection in general. One quality that changed more in the group of couples who had had a child related to the time spent in activities with each other. Couples who were parents had a steeper decline in terms of shared activities. This appears to be reasonable because child care of necessity takes time from the couple's shared activities.

There are also studies which report no decline in romantic love and lower separation rates in couples who have recently became parents (Belsky 1990, Lindahl et al. 1998). The probability of separation was lower, specifically during the first years, and may be explained by children acting as barriers to parental divorce (Belsky 1990). This suggests that the transition to parenthood may not be as distressing for marriages as has been previously suggested.

To summarize, apart from a general decline in satisfaction during the first years of marriage, and apart from the fact that a child may postpone a parental divorce, research shows that having a child has a negative effect on at least some aspects of couple relationships.

There are several different explanations for reduced satisfaction. In addition to more extensive social and economic demands, one cause of the development of problems is assumed to be that a child reduces the time the spouses can spend together. Another reason for dissatisfaction can be disagreements about child rearing (Belsky 1990). An additional source of conflict, specifically during the transition to parenthood, relates to the division of household labour (Belsky 1990, Kluwer and Heesink 1996, Voydanoff and Donnelly 1999). Considerable pressure is produced by disagreements between the spouses about the workload (Crohan 1996) and about how the household chores are supposed to be distributed between the spouses (Belsky et al. 1986).

Workload at home is, however, not the only thing to create stress when a couple has a child. Workplace demands and expectations can also be experienced as a burden (Yogev 1983, Haas et al. 2000). It has been maintained that studies of household work should not ignore the pressure of paid work (Kalleberg and Rosenfeld 1990). Compared with paid work, the division of household labour appears, however, to be a more dominant factor when it comes to marital discord (Kluwer and Heesink 1996). Women's participation in the labour market is not the only central factor when it comes to equality between the spouses. The husband's participation in housework and childcare is also a key issue (Menaghan and Parcel 1990).

Men and women experience the transition to parenthood differently, partly because they perform different roles as parents. Wives normally have a higher level of marital satisfaction than husbands before the birth of the baby and, after the birth, the wives' satisfaction declines more than husbands' satisfaction (Belsky et al. 1985). Wives' marital adjustment appears to be more sensitive to the effects of a child than the husbands' adjustment. This is thought to depend on the burdens imposed on mothers after the birth of the baby, as husbands appear not to change their contribution to the division of household labour (Belsky et al. 1983).

In addition, parents who have a second child can experience the workload at home and outside the home differently than first‐time parents. Marital quality declines when couples become parents, but this decline continues with additional children (Belsky et al. 1983). Bird (1997) suggested that increased economic burdens and problems with childcare are reasons for greater distress after the transition to parenthood. When more children arrive, economic burdens and childcare problems are probably emphasised to an even greater degree.

Children do not in and by themselves affect couple relationships negatively, but they do so as a result of increased social and economic burdens (Bird 1997), and by reducing the time the parents can spend with each other (Belsky 1990). Parents who experienced higher levels of psychosocial problems like economical difficulties, unemployment and insufficient social network also had higher stress levels (Östberg 1998). Sufficient social support from relatives, friends and others should mediate the impact children have on the quality of couple relationships. Social support generally has a favourable effect on marital satisfaction (Bird 1997, Bryant and Conger 1999), and it has been found to be particularly important for working mothers. Working women who reported lower levels of social support also reported a low level of marital adjustment (Graham et al. 2000).

Another important mediating factor for the influence of the child on the couple relationship is associated with how easy it is to take care of the child. Mothers of small children show clearly higher levels of parental stress when the child has behavioural problems (Östberg 1998). It can be assumed that the temperament of the child and that of the parents interact, but nevertheless, children with difficult temperaments impose real stress on marriages (Belsky 1990). Parents who have infants that are easy to handle have a more positive experience of their couple relationships (Howes and Markman 1989). When reactions to infant cries were investigated, mothers who responded more negatively to infant cries described their own infants as more difficult, and these mothers had a more negative perception of their marital quality (Pedersen et al. 1997). Fathers interacted more affectively and responsively with their children, when they perceived the child as less difficult (Volling and Belsky 1991) and it has been suggested that fathers' emotional investment in children is related to the quality of the relationship with the spouse (Bradley and Whiteside‐Mansell 1997).

The present study was performed in Sweden, where the economic support for parental leave and qualified daycare facilities is well‐established in society. The mother or the father can choose to stay at home with 80% of their salary for 390 days after the birth of the baby and an additional 90 days with a smaller allowance. Sixty days of the parental leave are designated exclusively for fathers. When the baby is one year old, municipal daycare of good quality is provided at a reasonable cost. These circumstances can have an important effect on the way husbands and wives experience the couple relationship, parenthood and the burdens related to it.

The aim of the present study was to investigate whether satisfaction with the couple relationship during the transition to parenthood is linked to household work, while taking account of stressful factors, the child's temperament and support from the environment. We compare first‐time parents with second‐time parents in order to see whether the possible association is stronger when the couple experiences the transition to parenthood for the first time.

Method

Participants

The sample comprised 258 Swedish speaking individuals, 128 mothers and 123 fathers, recruited through 21 child healthcare centres in South Bohuslän, Sweden. One mother and six fathers did not return the questionnaire or returned it unanswered, leaving 251 participants total. The median age of the mothers was 30.3 years while that of the fathers was 32.6 years. At the time of the study, the age of the babies ranged from 2 to 20 weeks, with a medium age of 9.1 weeks. Approximately half the participants were first‐time parents while 38% were second‐time parents; the remaining parents had more than two children. The couples had been married/cohabited for 6.4 years on average. Thirty‐seven percent of the men and 53% of the women had three years or more of university studies and 89% of the men and 84% of the women had had a paid job, full‐time or part‐time, before the birth of the child. Eleven percent of the men and 16% of the women were occupied by other types of issues like studies, unemployment, previous parental leave or long‐term sickness leave.

Measures

The quality of the couple relationship was assessed using the Dyadic Adjustment Scale (DAS; Spanier 1976), which is one of the most widely used measures of marital satisfaction. The scale is composed of 32 items, scored on a Likert‐type scale. The possible range of scores is 0 to 151, with a high rating representing positive adjustment. The scale can be divided into four sub‐scales: Dyadic Consensus, Dyadic Cohesion, Dyadic Satisfaction, and Affectional Expression. Spanier (1976) reported that the DAS had an alpha reliability of 0.94. The alpha reliability for the total scale in the present study was 0.92 (n = 250).

Household work

The following five questions were asked about the household work: ‘How challenging are your household tasks?’ (5 = slightly, 4 = little challenging, 3 = neither challenging nor not challenging, 2 = quite challenging, 1 = very challenging); ‘How enjoyable do you think it is to do your household tasks?’ (5 = very, 4 = quite, 3 = neither/nor, 2 = quite unenjoyable, 1 = not very enjoyable); ‘How strained do you get about doing your household tasks?’ (5 = slightly, 4 = not much, 3 = neither/ nor, 2 = quite, 1 = very); ‘On the whole, how satisfied are you with the division of household tasks between you and your partner?’ (5 = very, 4 = quite, 3 = neither/nor, 2 = quite dissatisfied, 1 = very dissatisfied); and ‘How satisfied do you think your partner is with how the household tasks are shared between you and him/her?’ (5 = very, 4 = quite, 3 = neither/nor, 2 = quite dissatisfied, 1 = very dissatisfied). The scores for these questions were added up to produce a total rating, with a low rating indicating distress about household work. The alpha reliability for the summed scores was 0.48 in total group (n = 249). The alphas varied between the sub‐groups and the item total correlations were low for the second question. The second question was, therefore, deleted, which produced alphas between 0.40 and 0.68 in the sub‐groups, the higher alphas for the females' and the lower for the males' responses.

Stress

When it came to stress, the participants were asked ‘Is there something within the following domains that you experience as particularly demanding: family relations (1 = no, 2 = yes), finance (1 = no, 2 = yes), work (1 = no, 2 = yes), health (1 = no, 2 = yes) or other areas?’ (1 = no, 2 = yes). The scores were then summed to build a total stress score. After the stress questions the participants were given an opportunity to comment specifically the stressful situation by an open‐ended request: ‘If you have answered yes on some of the questions above, please note what has been stressful’.

Difficult child

The Neonatal Perception Inventory (NPI) was developed by Broussard and Hartner (1970) to assess the mother's perception of her child as compared to the average child. Behavioural items such as crying, spitting, feeding, elimination, sleeping and predictability were included in the measure. The items were scored on a five‐point scale with the sum of the scores ranging from 6 to 30. When the score for the mother's own baby was subtracted from the score for an average baby the scores could range from −24 to 24. A low score indicated that the mother's own child was perceived as difficult. Alpha for the questions concerning the own baby was 0.75 and for the questions for an average baby 0.80 (n = 251).

Social support

In terms of social support the participants were asked: ‘How satisfied are you with the help or support you receive from: parents, relatives, friends, professionals or others?’ (5 = very satisfied, 4 = quite satisfied, 3 = neither satisfied nor dissatisfied, 2 = quite dissatisfied, 1 = very dissatisfied). The scores for the different sources of support were added up to produce a total evaluation. The alpha reliability for the summed scores was 0.62 (n = 251).

Transition to parenthood

To obtain an overall impression of how the parents experienced their transition to parenthood, we used the Transition to Parenthood questionnaire (TTP; Steffensmeier 1982). The scale measures changes in feelings of satisfaction and happiness in the couple's relationship after the baby's birth. It is composed of 25 items, scored on a Likert‐type scale. The summed scores range from 25 to 108. A low rating represents a positive attitude to change. The scale has three sub‐scales: Parental Responsibilities and Restrictions; Parental Gratification; and Marital Intimacy and Stability. Previous studies have reported alphas of between 0.75 and 0.82 for the sub‐scales (Zelkowitz and Milet 1997). In the current study, the alphas ranged from 0.65 to 0.82 for the sub‐scales and was 0.81 for the total scale (n = 245).

Procedure

The ethical approval of the study was applied and consented from the ethical committee at the medical faculty of Gothenburg University. The questionnaires were distributed over a period of seven weeks. The mothers were recruited by nurses when they visited the baby welfare clinic. In Sweden, practically every parent with a newborn baby visits a baby welfare clinic for regular check‐ups of the baby. Every mother who visited the particular nurse received an envelope containing an introductory letter, a questionnaire for the mother and the father with identical questions and addressed envelopes. Only one mother refused to accept the questionnaire. Both the mother and the father were asked to answer the questionnaire. Parents who had not answered were sent a reminder letter with questionnaires to each parent. Three child healthcare centres did not participate in the reminding procedure. Seventy percent of the couples returned the questionnaire. Those couples were rewarded with a book about infants' psychological development. When the returned questionnaires had been registered the names of the respondents were omitted.

Results

Means and standard deviations for the first‐ and the second‐time parents and for each gender are presented in Table 1. Scores on the Dyadic Adjustment Scale (DAS) and its four sub‐scales and the Transition to Parenthood Scale (TTP) and its three sub‐scales revealed slight variations between the groups, but t‐tests revealed no significant differences between the first‐ and second‐time parents or between the sexes. Nor did the perception of the infant differ between the sexes. Differences on household work, stress and support were measured with a non‐parametric test. The Mann–Whitney test showed that first‐time mothers were to some extent more content with household work than second‐time mothers (p = 0.046). Men in first‐time parent group were more stressed than women (p = 0.045). First‐time fathers reported greater social support than second‐time fathers (p = 0.001).

Table 1. Means and standard deviations for first‐ and second‐time parents.

	First child		Second child
	Women	Men	Women	Men
DAS
M	121.67	121.59	118.60	116.37
SD	12.95	14.97	13.87	14.77
(n)	(67)	(64)	(50)	(46)
Dyadic consensus
M	52.46	51.77	51.80	50.51
SD	5.61	7.93	5.75	6.46
(n)	(67)	(64)	(50)	(46)
Dyadic satisfaction
M	41.98	43.19	41.34	41.89
SD	5.26	4.32	5.206	5.61
(n)	(67)	(64)	(50)	(46)
Dyadic cohesion
M	17.99	17.69	16.94	16.54
SD	3.54	3.48	3.544	3.53
(n)	(67)	(64)	(50)	(46)
Affectional expression
M	9.12	8.95	8.52	8.50
SD	1.85	2.38	2.28	2.27
(n)	(67)	(64)	(50)	(46)
TTP
M	77.06	77.36	75.10	74.91
SD	8.60	7.66	8.338	7.92
(n)	(63)	(64)	(50)	(46)
Parental responsibilities and restrictions
M	22.29	23.11	22.74	22.74
SD	3.36	3.39	3.09	3.07
(n)	(65)	(64)	(50)	(46)
Parental gratifications
M	34.55	33.90	32.98	32.43
SD	3.94	3.61	3.460	3.83
(n)	(64)	(64)	(50)	(47)
Marital intimacy
M	20.31	20.34	19.38	19.64
SD	4.15	3.53	3.967	3.18
(n)	(64)	(64)	(50)	(47)
Household work
M	13.54	13.62	12.60	13.04
SD	2.68	2.52	2.35	2.07
(n)	(67)	(64)	(50)	(45)
Stress
M	5.63	5.89	5.70	5.83
SD	.85	1.25	.91	.97
(n)	(66)	(64)	(50)	(46)
Difficult child
M	5.63	5.28	6.04	5.70
SD	4.26	4.23	4.09	4.01
(n)	(67)	(64)	(50)	(47)
Support
M	19.52	19.16	18.58	17.28
SD	2.54	3.24	2.671	2.22
(N)	(67)	(64)	(50)	(47)

When the relationship satisfaction was compared between the couple the results showed a strong positive correlation (first‐time parents, r _s = 0.70, p<0.001, n = 59; second‐time parents, r _s = 0.78, p<0.001, n = 41). Table 2 presents Spearman rank correlations relating to the quality of couple relationships and work at home, stress, perception of the infant, and support from the environment. A strong link emerged between the quality of women's couple relationships and their experience of household work, specifically for first‐time mothers. A positive link between these variables was also significant for the first‐time fathers, but the correlation was weaker. The more satisfied the women were with their partners, the more content they were with their household chores. Another significant association related to stress was that first‐ and second‐time fathers and second‐time mothers were not happy with their couple relationship when they experienced their lives as stressful.

Table 2. Spearman rank correlations between couple relationship and workload at home, stress, experience of child and support for first‐ and second‐time parents.

	First child				Second child
	Women		Men		Women		Men
	r s	n	r s	n	r s	n	r s	n
Household work	0.47***	67	0.25*	64	0.27^a	50	0.10	44
Stress	−0.10	66	−0.28*	64	−0.28*	50	−0.49***	45
Difficult child	0.04	67	−0.05	64	0.05	50	0.05	46
Support	0.02	67	−0.19	64	0.05	50	0.18	46
^a p<0.10; * p<0.05; ** p<0.01; *** p<0.001.

When the independent variables were put in a logistic regression model (Table 3), the results show that household work had a strongly significant effect on first‐time mothers' relationship satisfaction. Stress was a significant predictor for relationship satisfaction for both mothers and fathers, but particularly for second‐time fathers. The model is significant in predicting satisfaction in couple relationship for all four groups.

Table 3. Logistic regression analyses to predict satisfaction in couple relationships from workload at home, stress, experience of child and support for first‐ and second‐time parents.

	First child								Second child
	Women(n = 66)				Men(n = 64)				Women(n = 49)				Men(n = 43)
	Odds ratio	Wald t‐ ratio	p	95% CI	Odds ratio	Wald t‐ ratio	p	95% CI	Odds ratio	Wald t‐ ratio	p	95% CI	Odds ratio	Wald t‐ ratio	p	95% CI
Household work	1.61	10.76	.001	1.21–2.13	1.18	1.74	.188	0.92–1.50	1.31	2.87	.090	0.96–1.78	1.10	0.23	.632	0.75–1.60
Stress	0.86	0.21	.650	0.44–1.66	0.48	5.05	.025	0.25–0.91	0.41	3.81	.050	0.17–1.00	0.31	7.21	.007	0.13–0.72
Difficult child	0.97	0.20	.657	0.83–1.12	0.98	0.73	.788	0.85–1.12	1.05	0.33	.566	0.90–1.22	1.04	0.15	.703	0.86–1.24
Support	0.97	0.05	.820	0.78–1.12	1.20	3.44	.064	0.99–1.45	1.02	0.03	.857	0.81–1.30	1.13	0.44	.507	0.72–1.64
Model: First child Women χ^2 17.43 (df = 4) p = 0.002; Men χ^2 14.12 (df = 4) p<0.007. Second child Women χ^2 9.86 (df = 4) p<0.043; Men χ^2 10.91 (df = 4) p<0.028.

Discussion

The aim of this study was to examine satisfaction with the couple relationship and the spouses' experience of the workload during the transition to parenthood. We wanted to extend the research literature by studying these questions in circumstances in Swedish society where economical support for parents through the parental leave system and the provision of professional daycare facilities are well‐established.

The main finding was an association between household work and the couple relationship for women. Why does household work have such a prominent association with women's satisfaction with their couple relationship? A newborn baby in the family is a full‐time project, and if we add household tasks to taking care of a baby it can become quite overwhelming. One representative example of this is a comment after stress questions made by one of the mothers: ‘I am at home, so I get to do too much of the household‐work and that results in quarrels’.

If household work creates problems when the mothers, and some of the fathers, are on parental leave, the problem may become even more crucial when both parents participate in paid work. ‘Is anyone doing the housework?’ (Bianchi et al. 2000) was the expressive title of a paper describing the trends in the division of household work. In some cases, husbands want their wives to play a traditional role, taking care of the household, even if the wife also works outside the home (Yogev and Brett 1985). Kluwer and Heesink (1996) emphasise the fact that conflicts about the division of household labour appears to be an important topic for future research because factors that influence relationship satisfaction influence the general well‐being of the entire family.

The association between household work and couple satisfaction in our study does not, however, appear to depend on how dissatisfied the individuals were per se with household tasks. There were no significant differences between the men's and the women's scores about the household work, but, when the values were correlated with relationship satisfaction, the link was significant for the women but not for the men. This indicates that even if men and women experience the household work in a similar way, the division of labour at home plays a stronger part in women's marital satisfaction than it does in men's. Previous research also indicates a similar tendency. Wives', but not husbands', dissatisfaction with the household workload was clearly associated with conflict between the spouses (Kluwer and Heesink 1996).

In the current study, stress was negatively associated with satisfaction in couple relationships, and the comments showed that it specifically concerned stress relating to work. It is often difficult for both women and men to combine work and family. In work environments where males are in the majority, there are fewer policies that facilitate the combination of employment and domestic issues. Men's colleagues and employers may also, even if only implicitly, discourage them from reducing their working hours (Yogev 1983, Kluwer and Heesink 1996). Even part‐time jobs are less common for men than for women (Van Dijk and Siegers 1996). In our study, the reasons the fathers gave for work stress clearly pointed to a similar tendency. Finance can be a stressful component in itself, but the fathers did not refer to finance as a primary source of distress with work. As a matter of fact, the majority of these fathers reported that the stress stemmed from the opposing demands they faced. There were fathers who were exposed to heavy demands at work, but they wanted to be at home with the child. There were fathers who regarded their work as stressful at the same time as they had demands at home and they felt they did not have the same time to spend on their work as before. Some fathers reported that they were just generally tired and exhausted.

Fathers' desire to be with the baby could be relieved if they took parental leave to a larger extent, but the most common choice in Sweden is for the mother to take most of the the leave. The fathers' share was 11.9% (SCB 1999) of the total number of leave days available. It has been suggested that the family finance is one of the main reasons for this unbalanced distribution of parental leave between mothers and fathers. This is associated with the upper income limit for benefits. If men have a higher income the family income is reduced to a greater degree if the father stays at home. This explanation is, however, not quite relevant. Fathers with higher education take parental leave to a greater degree in Sweden than fathers with lower education (Haas 1992). The main reason for the uneven distribution of leave between the parents is probably, as pointed out above, the fact that colleagues and employers are not used to men reducing their working hours and therefore dislike it.

Also women commented on their work‐related stress. How can women, who are on parental leave, experience workplace stress? A closer investigation of the reasons for stress revealed that when women described stress they expressed general worries about their future work or about their economic situation. They did not, however, mention the double burden as the reason for the work‐related stress, which men emphasised.

Other stress‐related comments were associated with family relations. Specifically second‐time fathers who were not happy with their partners made remarks related to relationships both with their in‐laws and their own family. In general, women appear to be more likely to experience trouble with the in‐laws than are men. One‐third of the women reported tension with parents‐in‐law (Marotz‐Baden and Cowan 1987) and the distress was specifically associated with wives' experience of a less satisfactory marital relationship (Bryant et al. 2001). In the present study, relationship dissatisfaction and comments about stress related to the partner may be quite obviously linked to each other. The link is more implicit when it comes to relationships with in‐laws. Further studies in this area would provide more specific information about relationships with in‐laws and the way these relationships are associated with each partner's experience of marital happiness.

Previous studies report an association between the child's temperament (Pedersen et al. 1997) and social support (Bird 1997, Graham et al. 2000) on the one hand and the couple relationship on the other. Our results only partly confirm these findings. First‐time fathers who reported a supportive environment experience their couple relationships somewhat more positively, but the child's temperament was not significantly linked with either men's or women's couple relationships. This matches with Belsky's (1990) suggestion that, when a couple receives good support from the environment a child with a more demanding temperament does not need to affect the couple relationship negatively. Another fact is that, if the couple relationship is experienced positively in the first place, a difficult child does not affect the parent's satisfaction with the relationship to any large degree. Cowan and Cowan (1995), for instance, maintain that it is unlikely that a child makes well‐functioning marriages bad. Apparently, a child's temperament and social support may be of importance but compared with household labour and difficult relatives, their significance is weaker.

Limitations of this study must be taken into consideration. Measures of household work as well as of stress are rather inadequate and may have influenced the results. The household work measure appears to be more reliable for females than for males. More sophisticated measures need to be utilized in future research.

When summing up our results we would like to refer to Stanley et al. (2000), who point out that even if interesting results are presented the findings should be interpreted with caution. After all, the results often provide only part of the information required to draw conclusions. Taking this into consideration, current research could have some practical implications. Social policies are already relatively well established in Sweden. Therefore, it would be possible to improve these policies if we could develop a more understanding attitude from both employers and work mates concerning parental leave. Another matter is the couple relationship in itself; it is always to advantage if the couple can communicate with each other, but specifically during the stressful period of having small children, conflict resolving is easier when the couple has good communication, characterised by reciprocity and sharing.