The patterns and practices of post-separation parenting are central to ensuring children's ongoing wellbeing (Amato, 2000; Smyth, 2004). Yet, there is very little existing Australian literature on post-separation parenting practices among Indigenous families. On parenting arrangements, the only directly relevant literature is a presentation by Qu and Weston (2012) that compared the pre- and post-separation parenting circumstances of Indigenous and non-Indigenous mothers and fathers. This study found that separated Indigenous parents were younger, poorer, had younger children at the time of separation and were far more likely to be in a cohabiting rather than a married relationship at the time of separation.

There is also a limited literature on the interaction of Indigenous families with the family law system. A 2004 article by Family Law Court Indigenous Family Consultant Steven Ralph, for example, details some unique aspects of Indigenous family law disputes. These include the extent of involvement of extended family in disputes, cultural issues around Indigenous affiliation and identity, and the lack of fit between norms of child access arrangements for many Indigenous families. More recently, the Family Law Council (2012) report, Improving the Family Law System for Aboriginal and Torres Strait Islander Clients, found that family law system services are underutilised by Indigenous families and there is evidence of significant unmet need in Indigenous communities. The very few Indigenous practitioners in the field and the low levels of Indigenous cultural competence among non-Indigenous practitioners have exacerbated this gap.

The literature on child support is even more scant, with the only identified literature being a 2010 presentation by Esler, Robertson, and Shipley (2010) on behalf of the Child Support Agency (CSA). Recognising a lack of knowledge of Indigenous families' interaction with the CSA, the authors suggested that the limited existing data indicated that Indigenous parents were more likely to be unemployed and paying or receiving low rates of child support than non-Indigenous CSA clients. The need for specific knowledge on how Indigenous families are negotiating the ongoing financial support of their children post-separation is of even greater import following the 2006 family law reforms. These reforms, among other objectives, included significant changes to how child support is calculated, paid and sought in order to encourage greater involvement of both parents in their children's lives following separation (Kaspiew et al., 2011).

The evidence from these sources and the (albeit relatively) larger Australian literature on post-separation parenting (see, for example, Kaspiew et al., 2011; Ministerial Taskforce on Child Support, 2005; Parkinson, 2007; Smyth, 2004) suggest the direct applicability of their findings to Indigenous families is not a reasonable practice. Along with the aspects identified by Ralph (2004) and Qu and Weston (2012) above, socio-demographic data also indicate that Indigenous families, separated and together, have unique dimensions. Indigenous families are, for example, far more likely to be socio-economically disadvantaged, and more likely to live in extended family households, be a sole-parent family and record higher rates of ex-nuptial births (Australian Institute of Health and Welfare [AIHW], 2011).

The main purpose of this paper is to use Footprints in Time: The Longitudinal Study of Indigenous Children (LSIC) to provide baseline descriptive information on the post-separation arrangements of Indigenous households, including:

• the number, proportion and characteristics of children who have a parent living elsewhere;
• the level and pattern of contact between children and parents living elsewhere; and
• the patterns of payment and receipt of child support for children.

The study

LSIC is a national panel study that in Wave 1 (2008) surveyed the families of 1,670 Indigenous children from 11 sites across Australia. These locations ranged from very remote communities to major capital cities and include: the Northern Territory "Top End"; south-east Queensland; the south coast of New South Wales; Mt Isa, including Mornington, Doomadgee and Normanton; Western Sydney; Dubbo; Greater Shepparton; the Torres Strait and Northern Peninsula area; the Kimberley region; Adelaide, including Port Augusta; and Alice Springs, including Hermannsburg. The Level of Relative Isolation (LORI) classification was used to allocate LSIC households into one of five categories of isolation, ranging from 1 = "none" (e.g., the Brisbane metropolitan area); 2 = "low" (e.g., Shepparton); 3 = "moderate" (e.g., Derby), 4 = "high" (e.g., Doomadgee); and 5 = "extreme" (e.g., Moa Island) (Department of Families, Housing, Community Services and Indigenous Affairs [FaHCSIA], 2009). Data for all LSIC waves are collected via face-to-face interviews between the study child's primary parent or main caregiver and locally employed Indigenous Research Administration Officers. As can be seen in Table 1, the spread of study children households is similar in geographic distribution to the total Indigenous population (Australian Bureau of Statistics [ABS], 2010). Around about 24% of LSIC households are in remote and very remote locations, but the majority (76%) are located in areas classified as having none or low levels of remoteness.

Source: LSIC Wave 1

The LSIC design divided the sample of study children into two cohorts: babies (B) and kids (K). At Wave 1, the majority of the B cohort were aged between 6 and 18 months and the K cohort between 3.5 and 4.5 years. Overwhelmingly, the main caregiver of the LSIC study children was their mother. As shown in Table 2, for 95% of the study children their nominated main caregiver was a birth parent, and in 98% of these cases that parent was their mother (n = 1,597). Around 5% of the study children were being primarily cared for by a person other than a birth parent, mostly a close female relative. Our analysis found that study children living in urban metro environments were just as likely to be in the care of another family member as those living in remote locations.

In this study we used data from Wave 2 (2009) of LSIC. We restricted analysis to Wave 2 because questions relating to parent-child contact and child support for study children who had a parent living elsewhere were not included in the Wave 1 survey. In Wave 2 (2009), 86% of families from Wave 1 were retained. In addition, to maintain the viability of the sample in remote regions, 88 new entrant families were interviewed. This provided a total of 1,527 study children and their families participating in Wave 2 (FaHCSIA, 2010).

Note: Percentages do not total exactly 100.0% due to rounding.

Source: LSIC Wave 1

Results

Study children whose other birth parent lives elsewhere

As shown in Table 3, 39% of the LSIC study children had a parent living elsewhere. In comparison, research using data from the Households, Income and Labor Dynamics in Australia (HILDA) survey, which uses a nationally representative population sample, suggests that around 25% of children have a birth parent living in another household (Walter, Hewitt, Natalier, Wulff, & Reynolds, 2010). Research using data from Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC) indicates that around 15% of children in its B and K cohorts did not live with both biological parents (Australian Institute of Family Studies [AIFS], 2011). Four hundred and twenty eight primary parents of 597 study children with a birth parent living elsewhere agreed to answer questions about the other parent. In 86% of these LSIC families, the parent living elsewhere is the study child's father, which is similar to that of the broader population, where 88% of children not living with both parents primarily reside with their mother (ABS, 2008). In the remaining 14% of households, the parent living elsewhere was the study child's mother or neither parent resided in the study child's household.

Note: ^a Note that 166 primary parents declined to provide information on the parent living elsewhere.

Source: LSIC Wave 2

The comparatively high proportion of study children with a birth parent living elsewhere is made more significant by the fact that LSIC study children are still quite young. In line with general trends on relationship separation, where parents are less likely to separate when they have young children, rates of having a birth parent living elsewhere could be expected to rise as study children move through their childhood years. In Table 4, we show the proportion of study children with a parent living elsewhere by cohort. We find that the main caregivers of the older K cohort reported higher rates of having a parent living elsewhere than the main caregivers of the B cohort. This difference (38% cf. 41%), however, is only three percentage points. This result suggests that, for the B cohort at least, parental separation happened very early in the study child's life. Or, alternatively, the study children's parents may have never lived together in the same household. While Qu and Weston's (2012) results support the finding of the younger age of the child at parental separation compared to non-Indigenous separated families, we were unable to ascertain the pattern of these variables, as LSIC Wave 1 or 2 did not collect data on the time since the study children's parental separation.

Note: Percentages may not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Socio-economic and demographic characteristics

While most LSIC family households were disadvantaged, we found a differentiation in the level of socio-economic position depending on whether the study child lived with both birth parents or not. As shown in Table 5, overall, those study children who had a parent living elsewhere tended to live in environments that were more disadvantaged than study children whose parents lived together. This was consistent across a wide range of socio-economic variables. For example, when the public and private renter figures are amalgamated over 80% of all LSIC households were renting. The proportion of renters was 82% in households where the study child's birth parents resided together compared to 88% for those with a birth parent living elsewhere. Moreover, the households where a birth parent lives elsewhere were statistically significantly more likely to be living in public rental housing than households where the parents reside together. Additionally, while fewer than half all LSIC households gave salary and wages as the major source of household income, the figures vary dramatically by parental household type. Fifty-six per cent of households with two birth parents reported salary or wages as the main source of household income compared to 20% of households where a birth parent of the study child lived elsewhere. Primary parents in households where the study child's other birth parent lived elsewhere were also less educated, less likely to be employed and more likely to live in public housing.

Notes: * Chi-square test indicates that this patterning is statistically significant at p < .05. ^a We excluded the 17 respondents with missing data on parent living elsewhere. ^b Note that numbers may not add up because respondents were allowed to nominate more than one source of main household income. ^c This is housing tenure at Wave 1. Percentages may not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Study child households where a birth parent was living elsewhere, therefore, were experiencing higher rates of socio-economic disadvantage than those where both birth parents resided together. It is critical, however, to reiterate that nearly all the LSIC families were socio-economically disadvantaged. The higher levels of socio-economic disadvantage for families where the study child's birth parent lives elsewhere, therefore, come off a very low starting point. These results indicate that Indigenous families where the child's other parent lives elsewhere are occupying the extreme end of socio-economic disadvantage.

We also found demographic differences between the two parental types of households. As shown in Table 6, for study children with a parent living elsewhere, a higher proportion of their primary parents were Indigenous compared to those living in households where both parents lived together. There are differences also in the age profile of primary parents. The ages of the primary parents of the study children who had a parent living elsewhere are concentrated in the younger and older age groups. This pattern can be attributed for the younger primary parents to the generally younger age profile of Indigenous sole parents (ABS, 2011), and for the older group to grandparents being the primary parent in the majority of the study children households where both parents lived elsewhere. The self-reported health of the primary parent in both types of parental households was similar.

Notes: * Chi-square test indicates that this patterning is statistically significant at p < .05. Percentages may not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Post-separation parenting arrangements

In LSIC Wave 2, 72% (n = 428) of the primary parents of the study children who had another parent living elsewhere agreed to answer questions about the contact between that child and their other parent. The results we report in this section are from the data supplied by these parents. In additional analysis comparing respondents who provided information and those who did not, we found that parents who did not agree to answer questions about the parent living elsewhere tended to have lower incomes than those who did agree.¹

As shown in Table 7, 56% of LSIC children with a birth parent living elsewhere see that parent at least once a week (41%) or at least once a month (15%). Around a quarter, however, never saw their other parent. In comparison, around half of all Australian children who have a birth parent living elsewhere see that parent at least once a fortnight and just over a quarter (28%) either never see that parent or see them less than once per year (ABS, 2008). While Qu and Weston (2012) found Indigenous separated mothers are more likely to report that the father spends no time with the child, from these data, the pattern of the frequency of contact between the LSIC study children and their other parent is not dissimilar to that found in total population figures.

Note: Includes responses only from primary parents of study children who had another parent living elsewhere who agreed to answer questions about the other parent (n = 428). Study children who had both parents living elsewhere (n = 33) were also excluded from this table. Percentages do not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

The type of contact, not just its frequency, is an important aspect in how post-separation parenting is practised. Patterns of overnight contact between study children and the parent living elsewhere varied by the age of the study child. As shown in Table 8, 42% of the B cohort stayed overnight with the other parent once a month or more. This figure increased to 49% in the K cohort, with one-quarter of primary parents reporting that the study child stayed overnight with the other parent at least once per week.

Note: Includes responses only from primary parents of study children who had another parent living elsewhere who agreed to answer questions about the other parent (n = 428). LSIC Wave 2 data on this variable categorised missing and refused responses into a single category and so they are excluded from this table. Study children who had both parents living elsewhere (n = 33) were also excluded from this table. Percentages do not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Payment and receipt of child support

In this section we investigate the receipt of child support by the primary parent in relation to LSIC study children. Within the broader Australian population, nearly 95% of separated parents with children register with the Child Support Agency. Among this group, roughly half pay or receive child support monies via the CSA, and the other half select to make or receive payments directly between the parents (Fehlberg & Behrens, 2008). Rates of assessed child support to be paid or received depend on a number of factors, including the income of the parent living elsewhere. Around 40% of Australian payers pay only the minimum child support liability of $312 per annum ($6 per week). Though not able to be fully quantified, the research also suggests that a substantial minority do not always pay their liabilities in full and/or regularly (Ludwig, 2008; Ministerial Taskforce on Child Support, 2005; Parkinson, 2007; Walter, 2002). In 2008, 58,000 CSA parents identified as Indigenous (Esler et al., 2010). Of the 597 LSIC study children in Wave 2 who had a parent living elsewhere only around half (281) had primary parents who agreed to answer questions about child support arrangements. The results we report in this section are from the data supplied by these parents. Our analysis of respondents who provided information and those who refused permission, suggest that those who refused permission were significantly more likely to report lower incomes.

As shown in Table 9, the majority of LSIC primary parent respondents whose study child had a birth parent living elsewhere received child support (58%). Of those who received child support the average amount of child support received was $59 per week. While not directly comparable due to data collection timing variation, this amount is lower than the median amounts of child support reported in other Australian literature (see Ministerial Taskforce on Child Support, 2005; Natalier et al. 2008). Such lower amounts are likely explained by the socio-economic disadvantage in Australian Indigenous families relative to the general population. A further one-quarter of parents did not know how much they received or stated that the amount varied. These data suggest that these parents were receiving irregular and/or part payments. Of the 114 primary parents who reported that they did not receive any child support for the study child, 24% indicated that they were not meant to receive it. The main reasons given for not receiving payments were that the other parent could not afford to pay (10%), child support couldn't find them (11%), they helped out (presumably in other ways) (10%), they didn't want to pay (14%) other reasons (23%) or the primary parent did not know (9%). In only one case was the response of "exempt due to cultural reasons" selected by a primary parent.

Note: ^a Includes LSIC study children in Wave 2 with a parent living elsewhere whose primary parent agreed to answer questions about child support arrangements, excludes 2 cases who refused to answer questions about child support and 1 parent who pays child support (n = 281). Percentages do not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Table 10 details how the LSIC parents received their child support payments and where they went for help and advice in relation to child support. As shown, the most common method of transfer was through the Child Support Agency (49%), followed by private arrangements (38%) and Centrelink (98%). Around 65% of LSIC primary parents had sought advice or help in relation to getting child support. The majority of these parents had gone directly to the Child Support Agency for that help or advice.

Note: ^a Includes LSIC study children in Wave 2 with a parent living elsewhere whose primary parent agreed to answer questions about child support arrangements and who reported receiving child support payments. Excludes 2 cases who refused to answer questions about child support and 1 parent who pays child support (n = 281). ^b Multiple responses were allowed. ^c This sums to 170, not 167 due to multiple responses by 3 parents, who answered that they received child support payments through the Child Support Agency and by private arrangement. ^d LSIC Wave 2 data on this variable categorised missing and refused responses into the single category so they have not been included in this table.

Source: LSIC Wave 2

Whether or not the primary parent receives child support has implications for broader family and child outcomes. As shown in Table 11, in relation to household income, those who did not receive child support were more likely to have a lower household income than those who did receive child support (the chi-square test indicates that this difference is marginally significant). It should be noted though that the receipt of child support did not make as much difference for the higher income households. This suggests the payment of child support may provide a strong financial buffer for households who receive it, particularly households with low to medium level incomes.

Notes: * Chi-square test indicates that this differs at p < .1. Percentages may not total exactly 100.0% due to rounding.

Source: LSIC Wave 2

Discussion

The answers to this study's first question on the proportion and characteristics of Indigenous children who have another parent living elsewhere have significant implications. Nearly 40% of LSIC study children live in households apart from at least one of their biological parents; a rate much higher than total Australian population estimates of 25%. The relatively young ages of the study children indicates that this disparity will likely widen further over the course of their childhood years. These figures are particularly concerning, because, even though almost all LSIC study children were living in socio-economically disadvantaged households, this disadvantage was greater where study children had a parent living elsewhere. Thus, these Indigenous children are particularly vulnerable economically and socially. Significant and systematic disadvantages are already accruing in the young lives of these study children. Their families and communities will likely require substantial support systems and deployment of resources to aid them in their task of nurturing their children to strong, healthy, capable young adulthood.

The findings related to our second question - the level and shape of the interaction between these study children and the parent living elsewhere - also provide important new knowledge. Our results indicate similarities, but also differences, between the patterns of post-separation parenting of the LSIC study children and that of all Australian children living separately from one parent. First, consistent with total population findings (ABS, 2008), the primary parent of the vast majority of LSIC children with a parenting living elsewhere was the child's mother. Second, in line with mainstream data (ABS, 2008), around half of the study children saw the other parent regularly and around one-quarter rarely or never saw the other parent.

A subtle, but important difference, however, was found in the effect of the child's parent's previous relationship on post-separation parenting. The broader Australian research finds that ex-nuptial fathers are significantly less likely than their married counterparts to maintain ongoing contact with their child/ren (Walter, 2000). The LSIC data did not allow us to ascertain the status of the study children's parents' previous relationship, but we already know from other data that Indigenous parents are less likely to be in de jure marriage relationships than non-Indigenous parents (ABS, 2010). Additionally, in this study, the high proportion of the B cohort already living apart from their other parent suggests a substantial number never lived with the other parent in the same household. Despite these (imputed) high rates of ex-nuptial parenting, the level and patterns of ongoing contact between the study child and the parent living elsewhere were at similar, or perhaps even higher, levels than those found in total population figures. Our findings, therefore, indicate that not being previously married and/or never living in the same household as the child's other parent does not affect Indigenous post-separation parenting practices in the same negative ways as it does the post-separation parenting practices of non-Indigenous parents who live apart from their child.

Our third question sought data on child support. While the level of median child support payments was lower for LSIC study children households than in the total CSA payee population, similarities were observed in the patterns of receipt in the LSIC families. The proportion of Indigenous primary parents whose method of transfer of monies is via the Child Support Agency is roughly the same as that found among all CSA clients (Ministerial Task Force on Child Support, 2005). Our results on the proportion of primary parents receiving child support were lower, but also not outside the parameters among non-Indigenous primary parents. CSA figures indicate that around 40% of all payers pay only the minimum child support liability (House of Representatives Standing Committee on Family and Community Affairs, 2003), which is currently $312 per annum or $6.00 per week, per payer.

While 40% of the LSIC parents reported receiving no child support, as suggested in mainstream literature (see Walter et al., 2010), many of these parents may be in fact receiving the minimum amount through the Centrelink system. At $6.00 per week, this amount is likely to be either not noticed or perhaps not classified as child support due to its relative financial insignificance. The low level of knowledge of entitlements among those not receiving child support monies suggests more targeted and Indigenous-appropriate educative and engagement work by agencies such as the CSA may be beneficial. The fact that the CSA was most commonly identified by a substantial proportion of separated parents as their main source of information suggests that study children's parents are open to CSA interaction.

The critical finding in relation to child support was the association between its receipt and rates of household income. Despite the very high levels of disadvantage among the households where study children had a parent living elsewhere, those who did receive some child support monies tended to have higher incomes than those who reported receiving no child support. This suggests that the, albeit limited, financial buffer provided by child support monies may make a disproportionate difference to households battling poverty and social marginalisation. As such, child support may be even more critical for Indigenous households than non-Indigenous households for reducing their economic and associated social vulnerability. This finding is tentative, as households receiving child support may have been higher income households before separation. However, the growing proportion of study children with a parent living elsewhere and the deep disadvantage of those households revealed by this study indicate that the need to further explore the potential importance of child support in such households is compelling.

Conclusion

This study builds on the knowledge base of Indigenous post-separation parenting practices. These practices are shown to vary in subtle but important ways from those of non-Indigenous families. In turn, the results of this study indicate that targeted, Indigenous-informed and culturally shaped research such as that delivered by LSIC is of crucial importance if public policy and its outcomes are to meaningfully assist such families to raise strong, resilient children with positive life chances. The task of future research is to examine the implications of the layers of disadvantage for those study children who not only have a parent living elsewhere but have little contact or support from them.