Growing up in Scotland: change in early childhood and the impact of significant events

CHAPTER 3: PARENTAL SEPARATION

In this chapter we investigate what happens to families when parents separate in the first five years of their children's lives. Undoubtedly there is a complex interaction between parental separation and other factors that can help increase or decrease the risk of poor child outcomes. Such factors include exposure to parental conflict, the timing of the separation, the quality of the parent-child relationship, as well as related issues such as strained finances and maternal mental health (Mooney et al., 2009; Coleman and Glenn, 2009).

3.1 Key findings

• The incidence of parental separation is highest in the first couple of years after the birth.
• Over one in ten children (11%) experienced parental separation in the first five years of their lives. In two-thirds of these families the mother did not re-partner during this time period.
• Families most likely to experience parental separation are those where the parents were unmarried, the family was living on low income or where the pregnancy had been unplanned.
• Families experiencing separation were more likely than stable couple families to experience subsequent income poverty and poor maternal mental health.

3.2 How many families experience parental separation?

The majority of GUS children were born into couple families where the parents ⁶ remained together throughout the five-year period. However, 11% of children experienced parental separation at some point during the first five years of their lives. Most children continued to live with their mother, and in two-thirds of these families the mother remained a lone parent throughout the period after separating (Table 3.1).

Table 3.1 Family status stability during first five years

Family status stability	%	Unweighted frequency
Stable couple family throughout	72	2862
Parents separated after birth - mother did not re-partner	7	235
Parents separated after birth - mother re-partnered	4	114
Originally absent parent moved in with child	4	102
Originally lone parent re-partnered	3	68
Stable lone parent throughout	9	230
Bases
Weighted	3609	3611
Unweighted	3611

Base: All families taking part in all five years.

Overall, the incidence of parental separation declines over the first five years after birth. However, this pattern is not apparent when looking at only those families where the mother did not re-partner (Table 3.2).

Table 3.2 Timing of parents' separation

Age of GUS child at time of separation	Mother did not re-partner %	Mother re-partnered %	Total %
Age 0-1	20	38	26
Age 1-2	22	30	25
Age 2-3	22	21	22
Age 3-4	15	8	13
Age 4-5	22	3	15
Bases
Weighted	265	134	399
Unweighted	235	114	349

Base: All separated families taking part in all five years.

Note: Column per cent.

Table 3.2 includes families where the parents separated between the birth and the first interview. Although the first interview was carried out when the GUS child was 10 months old, mothers who at that point reported that they were not living with the child's father were asked to describe their relationship to him at the time of the birth. However, for the remainder of this chapter, we focus only on families who experienced separation after the first interview (thus excluding the 3% of all families where parental separation occurred between the birth and the first interview). The reasons for limiting our analysis to families where the separation occurred after the first interview are analytical as it ensures the event occurs after 'baseline' information has been collected.

3.3 Which families are most likely to experience parental separation?

As shown in Table 3.3 and Table 3.4 below, the likelihood of parental separation varies significantly by a number of the parental and household background characteristics listed in section 2.1.3 above and measured in 2005/06 (when the child was aged 0-1).

The likelihood of separation is higher among families with younger mothers, mothers with no qualifications, mothers with poor mental health, cohabiting parents, and when the birth had not been planned. The likelihood of separation is also higher among families living in rented accommodation, workless families, those living on a low income or in most deprived areas as well as among large families.

However, many of these variables are likely to be associated with each other. For example, large families are generally more likely to live on relative low income, as are workless families. Next, we therefore turn to multivariate analysis to identify those factors which are independently related to separation when other, potentially confounding, factors are taken into account. The associations which were identified in the multivariate analysis as remaining significant when controlling for other factors are highlighted in Table 3.3 and Table 3.4 and discussed further below.

Table 3.3 Separation by background characteristics of parents

Base: All originally couple families (at sweep 1) taking part in all five years.

Note: Row per cent.

Note: Shaded rows show characteristics with statistically significant associations with relationship history, after controlling for other factors in multivariate regression analysis.

Table 3.4 Separation by background characteristics of household

Base: All originally couple families (at sweep 1) taking part in all five years.

Note: Row per cent.

Note: Shaded rows show characteristics with statistically significant associations with relationship history, after controlling for other factors in multivariate regression analysis.

For the multivariate analysis, and for the remainder of this chapter, we focus only on families who experienced separation and did not re-partner (seven per cent of all originally couple families). That is, the parents separated after the first interview and the mother did not re-partner during the period studied. These families are compared with stable couple families. The reason for limiting our analysis to families where the mother did not re-partner is that re-partnering (the entry of a step-parent into the household) constitutes another possibly significant event in the child's life, which can affect family life in ways that are likely to differ from the original separation event.

The multivariate analysis shows that after controlling for other factors, the background characteristics that remained significantly associated with a higher likelihood of parental separation were the parents' relationship status at the birth of the GUS child, whether the birth was planned and income at the time of the first interview (see Table C.1 in the technical appendix for full results).

• Cohabiting parents were more likely than married parents to separate ⁷ .
• Families where the birth of the GUS child had not been planned were more likely to experience separation compared with families where the birth had been planned.
• Parental separation was more likely among low income families.

3.4 What happens to children who experience parental separation?

Families that experienced parental separation were more likely than stable couple families to experience drivers of child outcomes (measured in 2009/10); notably:

• household chaos;
• income poverty;
• poor maternal mental health; and
• lower parent-child warmth.

There does not seem to be a substantial difference in terms of the Pianta conflict dimension (see Table 3.5).

Table 3.5 Drivers of child outcomes by relationship stability

		%
Home chaos (% high level chaos)	Stable couple family	32
	Parents separated	40
	All	32
Income poverty (% poor)	Stable couple family	19
	Parents separated	55
	All	22
Maternal mental health (% poor mental health)	Stable couple family	12
	Parents separated	28
	All	13
Pianta warmth (% lower or least warmth)	Stable couple family	21
	Parents separated	29
	All	21
Pianta conflict (% higher or most conflict)	Stable couple family	15
	Parents separated	17
	All	15
Bases
Weighted		2644
Unweighted		2880

Base: All originally couple families taking part in all five years.

Table 3.5 does not show the level of the driver measures prior to the separation event taking place, and so we can not judge whether there has been a change in the driver following the event. For this we turn to multivariate regression analysis which allows us to look at the relationship between the event and the drivers of child outcomes while taking other potentially confounding factors into account, and importantly, controlling for the level of the driver measure at year 1 (note that a measure of home chaos is not available in year 1).

Table 3.6 summarises the results from the regression models. The dependent variable for each model is the relevant driver of child outcomes named in the column headings, the separation event and the sweep 1 measure of the driver are listed down the left and the arrows indicate the direction of any significant association. All of the measures of family characteristics listed in Section 2.1.3 were initially entered into the models as control variables (see Tables D.1 to D.5 in the technical appendix for full results).

Parental separation is significantly associated with a higher likelihood of income poverty and poor maternal mental health, but not significantly associated with either home chaos, or mother-child relationship on either the warmth or conflict dimensions.

• Controlling for income poverty in year 1, and other factors, families that had experienced separation were much more likely to be income poor in year 5.
• Mothers who experienced separation were more likely to experience poor mental health in year 5 (controlling for maternal mental health in year 1, and other factors). Unsurprisingly, mothers with poor mental health in year 1 were more likely to have poor mental health in year 5. However, there was also an interaction between the separation event and the mother's mental health in year 1 which means that the negative effect of separation on mental health is lessened for mothers who had poor mental health prior to the separation. This might suggest some support for other research reviewed by Coleman and Glenn (2010) which indicated that smaller increases in depression have been noted following the separation from high-conflict relationships, or where a partner has been depressed during the relationship.

Table 3.6 Relationship between parental separation and drivers of child outcomes controlled for other factors

	Drivers of child outcomes
	High level home chaos	Income poverty	Poor maternal mental health	Parent-child - Low warmth	Parent-child - High conflict
Parents separated		?	?
Driver present at year 1	n/a	?	?	?	?
Interaction: Driver present (year 1) and parents separated	n/a		?

Note: All factors other than the Event are measured at the Sweep 1 interview (2005/06).

Note: Arrows indicate whether an event or year 1 driver category is associated with significantly higher (?) odds of the driver of negative child outcomes occurring, compared with the reference category.

Note: All factors with arrows (?) are significant at 5% level, unless otherwise indicated. Blank cells indicate no significant relationship.

The timing of parental separation was included in separate models (not shown). The timing of parental separation was significantly associated with income poverty, with those separating between the third and fourth interviews having the highest likelihood of poverty in year 5. The timing of parental separation was also associated with poor mental health, with more recent separation associated with higher likelihood of poor mental health. However, it should be noted that relatively few families separated in any single year so these findings should be interpreted with caution.