Family-related life events

Insights from two Australian longitudinal studies
Research Report No. 22 – December 2012

3. Analyses of the Longitudinal Study of Australian Children

3.1 The Longitudinal Study of Australian Children

LSAC is a national study that, from 2004, has been following the development of around 10,000 children and their families. The study provides extensive information about two age cohorts of children—with each cohort comprising approximately 5,000 children. At the first survey wave in 2004, children in the younger cohort were aged 0–1 years (called the “B cohort”) and those in the elder cohort were 4–5 years old (called the “K cohort”). These two cohorts and their families are followed up every two years, with the most recent data available for analysis having been collected in 2010 (Wave 4). Information was collected from around 83–84% of the original sample in Wave 4, representing more than 8,000 families in total.

The majority of analyses presented in this report are based on Wave 4 data, although earlier waves of data are also used in parts. At Wave 4, the LSAC study children in each cohort were aged 6–7 years and 10–11 years respectively. As these children often have siblings, the families included in the study cover a wider range of ages.

Other information about LSAC, including additional details about methodology, sample sizes across the survey waves, retention rates for families, the number of respondents answering the set of questions on life events in each survey wave are shown in Appendix D.

As was the case for the results outlined for the HILDA survey data, the LSAC-related results derive from analyses of weighted data. The weights adjust the estimates produced from the sample, to take account of some biases in the sample that have been introduced through the initial sample selection and non-response.

In the study, many details are collected from “Parent 1”, who is the parent or guardian nominated as the one knowing the most about the LSAC study child. Information is also collected from and about “Parent 2”, if there is one in the household. Most often, Parent 2 is the primary carer’s partner, and therefore we refer to this person as partner. Generally, we refer to respondents as parents throughout the report.13

3.2 Questions relating to life events

In each survey wave of LSAC, the primary carer is asked whether or not, in the previous 12 months, they have experienced any one of a list of life events. Since Wave 2, respondents have been asked to indicate any of those that “happened to” either them or their partner during the previous 12 months.14 In Waves 2–4, it was therefore not possible to identify whether an event happened more directly to the respondent or their partner, although some events were obviously “family events”.

Appendix E ( Table E2) shows, for each cohort, the proportion of primary carers who indicated in Wave 1 that they had experienced each of the life events (taken separately) in the previous 12 months, and the proportions who indicated in Waves 2–4 that they or their partner had experienced these events in the previous 12 months.

The list of life events has been largely consistent across the waves of the study, although in Wave 4, some new items were included, as noted in the list below. The life events examined in the present study were:

  • health concerns or the death of someone close:
  • suffered a serious illness, injury or assault;
  • had a serious illness, injury or assault happen to a close relative;
  • had a parent, partner or child die;
  • had a close family friend or another relative die (e.g., aunt, cousin, grandparent);
  • family and household composition:
  • was pregnant or had a baby;
  • had someone new (other than a new baby or partner) move into the household (e.g., new step-child, foster child, other relative, friend or boarder);
  • started living with a new partner/spouse [Wave 4 only];
  • had a separation due to relationship or marital difficulties;
  • broke off a steady romantic relationship;
  • financial and social matters:
  • had a major financial crisis;
  • lost job, but not from choice (e.g., sacked, redundant, contract ended);
  • had something of value lost or stolen;
  • had someone in the household with an alcohol or drug problem;
  • residential matters:
  • moved house [Wave 4 only];
  • lived in a drought-affected area [Wave 4 only]; and
  • had home or local area affected by bushfire, flooding or severe storm [Wave 4 only].

Two of the experiences listed above may have commenced well before the 12-month interval investigated. These are: had someone in the household with an alcohol or drug problem; lived in a drought-affected area.

Respondents were also asked about a number of other life events that were considered to be not so directly relevant to this research project. A list of these other events is included in Appendix D.

3.3 Prevalence of life events

Table 16 shows the proportion of primary carers who reported in 2010 (Wave 4 data) that they or their partner had experienced each of the life events listed in the previous 12 months. Similar data for Waves 1, 2 and 3 are presented in Appendix E Table E2.

Table 16: Prevalence of parents’ life events by cohort, B and K cohorts LSAC, Wave 4
Parents’ life events in the previous 12 months B cohort (%) K cohort (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 11.3 12.7 12.0
Had a serious illness, injury or assault happen to a close relative 19.9 19.5 19.7
Had a parent, partner or child die 4.8 5.8 5.3
Had a close family friend or another relative die 26.0 24.7 25.4
Family and household composition
Was pregnant or had a child 10.4 4.6 7.5
Had someone new (other than a new baby or partner) move into the household 7.2 6.7 6.9
Started living with a new partner/spouse 1.9 1.9 1.9
Had a separation due to relationship or marital difficulties 5.9 5.0 5.5
Broke off a steady romantic relationship 4.3 3.6 3.9
Financial and social matters
Had a major financial crisis 10.8 12.5 11.6
Lost job, but not from choice 7.9 9.0 8.5
Had something of value lost or stolen 6.6 6.7 6.7
Had someone in the household with an alcohol or drug problem 3.0 3.7 3.4
Residential matters
Moved house 18.7 15.0 16.9
Lived in a drought-affected area 6.7 7.5 7.1
Had home or local area affected by bushfire, flooding or severe storm 5.9 7.1 6.5
Sample size ( N) 4,202 4,111 8,313

Across the sample, combining the B and K cohorts:

  • The death of a close family friend or another relative was the most commonly experienced event, with one-quarter of respondents reporting that this had occurred to them or their partner in the previous 12 months.
  • The next most commonly experienced event was that a close relative had become seriously ill or had been injured or assaulted (mentioned by 20% of respondents).
  • On a similar theme, 12% of respondents indicated that they or their partner had suffered a serious illness, injury or assault.
  • A slightly higher proportion of respondents (17%) reported a different sort of life event—that of moving house.
  • A major financial crisis was also reported by 12%. For some families this may have been related to the respondent or partner having lost his or her job—reported by 9% of respondents.
  • Relationship changes were less commonly experienced than those mentioned above: 6% referred to separation due to relationship or marital difficulties; 4% reported the experience of dissolution of romantic relationship; and 2% indicated that they had started living with a new partner or spouse.
  • The following diverse range of other events occurred to 5–8% of respondents: was pregnant or had a baby (8%); lived in a drought-affected area (7%); had their home or local area affected by bushfire, flooding or severe storm (7%); had someone other than new baby or partner move into the household (7%); had something of value lost or stolen (7%), and had a parent, partner or child die (5%).
  • The only other event examined in this analysis—having someone in the household with an alcohol or drug problem—was reported by 3%.

Source: LSAC Wave 4, B and K cohorts combined (2010)

3.4 Experiences of multiple life events

While some of the events outlined above would cause greater disruption than others, the overall level of disruption experienced would also be influenced by the experience of an accumulation of events. This section focuses on the accumulation of the different events listed in Table 16 that were experienced by respondents and/or their partners. Among respondents with partners, those reporting two events may have been referring to one event experienced by them and the other by the partner, or two events experienced by one member of this partnership. Several of the events listed would have occurred to both partners together (e.g., moving residence while in the relationship), and events occurring mainly to one partner would also have important repercussions for the other partner (e.g., loss of job) (see Table 17).

Table 17: Overall prevalence of multiple life events experienced in previous 12 months, B and K cohorts LSAC, Wave 4
Number of life events B cohort (%) K cohort (%) Total (%)
None 29.9 30.3 30.1
1 or more 70.1 69.7 69.9
1 29.0 30.2 29.6
2 19.7 19.6 19.7
3 11.1 10.9 11.0
4 5.5 5.0 5.3
5 or more 4.7 4.1 4.4
Totals 100.0 100.0 100.0
Sample size (N) 4,202 4,111 8,313

Source: LSAC Wave 4, B and K cohorts combined

Across the sample at Wave 4:

  • Seventy per cent of respondents indicated that at least one event had occurred to either them or their partner. In other words, 30% reported that they had not experienced any of the events listed.
  • Of the total sample, 30% reported one event, while approximately 20% indicated the experience of two events. That is, half the sample had experienced one or two of the events listed.
  • Some 21% of parents indicated that they or their partner had experienced three or more events (with four or more events occurring to 10%).

Of those who indicated that they (or partner, where applicable) experienced only one life event, the most commonly mentioned were:

  • having a close family friend or another relative die (representing 24% of all those who experienced only one event);
  • having a serious illness, injury or assault occur to a close relative (14%); and
  • moving house (13%).

Of those who indicated that they or their partner experienced two life events between them (if partnered), the most commonly mentioned combinations were:

  • having a serious illness, injury or assault occur to a close relative, together with having a close family friend or another relative die (this combination was reported by 11% who experienced two life events);
  • having a close family friend or another relative die, together with moving house (5%); and
  • having a serious illness, injury or assault happen to a close relative, together with suffering a serious illness, injury or assault (5%).

Of those who experienced three life events, the most commonly experienced combinations were:

  • suffering a serious illness, injury or assault, together with having a serious illness, injury or assault happen to a close relative, and having a close family friend or another relative die (with this combination of events being reported by 4% who experienced three life events);
  • having a serious illness, injury or assault happen to a close relative, together with having a close family friend or another relative die, and moving house (4%); and
  • having a parent, partner or child die, together with having a serious illness, injury or assault happen to a close relative, and having a close family friend or another relative die (3%).

Those experiencing four or more life events were too varied to allow a meaningful summary of the prevalence of different combinations. For example, the most commonly experienced combination of events was reported by only 12 of the 700 respondents who reported four events. (The events in that set were: having a serious illness, injury or assault happen to a close relative + having a close family friend or another relative die + having a major financial crisis + moving house.)

3.5 Socio-demographic circumstances associated with experience of life events

A particular focus of this report was to identify which parents were at greatest risk of experiencing particular life events. The following sections examine this, and this subsection provides a summary of the methods used for this purpose.

A set of characteristics was identified as being potentially important in explaining the likelihood of life events occurring. Broadly, the characteristics were: family form and composition, family background (parents’ country of birth and language), socio-economic circumstances, and residential location.

These measures are:

  • family form and composition: relationship status;
  • primary carer’s age;
  • age of youngest child in the family;
  • family background: parents’ country of birth and main language;
  • family socio economic circumstances:
  • household income;
  • main source of household income;
  • parental employment;
  • primary carer’s education;
  • housing tenure; and
  • residential location: remoteness of region.

The variables are described more fully in the presentation of results in the subsections that follow.

Information from Wave 3 of LSAC was used to identify these characteristics in respondents, in order to then relate them to reports of having experienced life events in the 12 months prior to Wave 4 of the study.

To ascertain which characteristics most strongly predicted experiencing each life event, multivariate analyses were used. These models allow us to explore to what extent each of the socio-demographic variables has a unique, independent association with the likelihood of having experienced that event. These analyses have been used to highlight the most significant factors, and we have then focused on those factors when describing the results. Information about how the multivariate results can be interpreted is presented in Box 1.

Box 1: Interpretation of multivariate results

For each life event, a variable was created that indicated whether a life event had been experienced. Each variable was coded as “1” for having experienced the event and “0” for not having that experience. For each of these variables, a logistic regression model was estimated. This form of model is appropriate given that the variables we are interested in are binary. The explanatory variables included in the models were the same for each life event, as described in the next sub-section.

Results of these analyses have been presented as odds ratios (Tables 18 to 21). The “odds” of experiencing a particular life event refers to the probability of experiencing it, expressed as a ratio of the probability of not experiencing it. That is, odds ratios represent estimates of how the “odds” vary for those with and without particular characteristics.

In these analyses, an odds ratio provides an indication of whether experiencing a particular life event is more likely (when the odds ratio is greater than 1) or less likely (when the odds ratio is less than 1) for those with a particular characteristic, compared to those in a comparison group (the reference category of the variable). When the odds ratio is equal (or close to) to 1, there is no (or little) difference between those with that characteristic and those in the reference group.

For example, for family type, the reference category is a married two-parent family. The odds ratios, then, compare the “odds” of each life event having occurred to those in either a cohabiting two-parent family or a single-parent family with the “odds” of these events having occurred to those in a married two-parent family. For the life event “was pregnant or had a baby” (Table 19), the odds ratio of 1.38 for cohabiting couples indicates that the odds of having a new child among cohabiting couples was 1.38 times that of married couples. The (non-significant) odds ratio of 1.07 for single parents indicates that there was not a significant difference between married couples and single parents in their odds of having a new child.

The stars in the table indicate the statistical significance of each odds ratio. If there are no stars on a figure, this indicates that, according to conventional levels of significance, this odds ratio does not differ significantly from 1. A greater number of stars indicate that we have greater confidence that this variable has a significant association with the prevalence of this life event. As noted in the example above, the odds ratio of 1.07 for single parents, compared to married couples, in the likelihood of having had a new baby was not statistically significant from 1; that is, the prevalence of this life event did not differ for these two groups.

The size of the odds ratio indicates how much the life event prevalence varies according to this characteristic. Thus, if the odds ratio is greater than one, the larger the number is, the greater is the difference in the prevalence of this life event between those with this characteristic and those in the reference group. If the odds ratio is less than one, the closer the number is to zero, the smaller is the relative likelihood of this life event having been experienced by those with this characteristic compared to those in the reference group. Put another way, the closer the number is to zero for a characteristic, the greater is the prevalence of this life event for those in the reference group for this characteristic.

Note that a limitation of this analysis is that these odds ratios only allow comparison back to the reference group in a strict sense, although the size and direction of the coefficients can be used as a guide to how the prevalence of life events compares across other groups. For example, looking at the likelihood of having a new baby for those in a cohabiting two-parent family or a single-parent family, the odds ratios (1.38 and 1.07 respectively) are based on comparisons for each group to the married couple families The relative size of these odds ratios suggests that cohabiting families have a greater likelihood of having a new baby than single parents. However, further statistical tests would be required to assert this with certainty.

We next provide an overview of the multivariate results. However, more detail about each of the variables and their associations with life events is presented in the subsections that follow, in which the associations between each variable and the likelihood of experiencing each life event are presented in figures and tables.

In addition to separately examining each life event, analysis of the number of life events experienced is included. However, as some life events may be viewed quite differently than others, three different counts of life events are used: (a) the overall total, including all possible life events; (b) the total excluding family and household changes that are not so inherently negative (having a new baby, a new partner or spouse, or another household member); and a further limited total that excludes the climate-related life events of living in a drought-affected area, or living in an area (and possibly home) that was affected by storms, floods or fire in the previous 12 months.

For multivariate analyses of the extent to which the number of life events experienced could be explained by the various socio-demographic factors examined, models were estimated with the same set of variables as used in the analyses of specific life events. For these models, ordinary least squares was used and each coefficient can be interpreted according to how much the number of life events differs (is higher or lower) for someone with a particular characteristic, relative to the reference group for that characteristic. This approach was also used in the analysis of HILDA data outlined in Section 2.

Overview of multivariate results

The multivariate results are presented in the following tables:

  • Table 18 includes life events related to health concerns or the death of someone close;
  • Table 19 includes life events related to family and household composition;
  • Table 20 includes life events related to financial and social matters;
  • Table 21 includes life events related to residential matters; and
  • Table 22 shows the results for the three counts of life events.

The present section provides a broad overview of the results, highlighting some of the key findings. This is followed by a more detailed description of specific results.

Consistent with findings concerning age-related experiences of respondents in the HILDA dataset, the prevalence of a range of life events varied according to the age of LSAC parents, with the younger parents being at greater risk than older parents of experiencing all the life events related to family and household composition (those in Table 19), and also being more likely than other parents to have moved house. In addition, younger LSAC parents were more likely than older parents to have had something of value lost or stolen. They were less likely, however, to have lived in an area affected by bushfire, flooding or severe storms. Younger parents were the least likely to have had a parent, partner or child die, but more likely than other parents to have experienced the death of a close family friend or other relative. Younger parents had also experienced a greater number of the life events examined than older parents.

The experience of several life events (taken separately) also varied with housing tenure, with parents living in rental accommodation being more likely than those living in a home that they owned or were buying to indicate that they or their partner had experienced a serious illness, injury or assault, the death of a close family friend or other relative, moving house, and life events related to family and household composition (those in Table 19) and to financial and social life events (those in Table 20). Given such consistency in trends, renters also experienced a greater number of life events than those who owned or were buying their home. Parents in “other” housing tenure arrangements also tended to have a higher number of life events when compared to those who own or are buying their home.

Relatively few of the characteristics we examined explained the likelihood of experiencing life events related to health concerns or the death of someone close (i.e., the four events listed in Table 18), perhaps because these life events are most likely predicted by more external factors; in particular, the characteristics of those other people.

Relationship status, on the other hand, was a key variable in explaining the life events related to family and household composition and also mattered to some of the life events related to financial and social matters. The nature of these trends are outlined in the next subsection. For other life events, a range of relationships emerged, which are described more fully in the subsections that follow.

Table 18: Multivariate analyses of socio-demographic characteristics and life events related to health concerns or the death of someone close, odds ratios
Characteristics as measured in previous wave Suffered a serious illness, injury or assault Had a serious illness, injury or assault happen to a close relative Had a parent, partner or child die Had a close family friend or another relative die
Family type (ref. = couple, married)
Couple, cohabiting 1.17 1.21 * 1.16 1.12
Single parent 0.94 0.87 0.65 0.89
Main source income (ref. = wages)
Government support 1.38 1.04 1.14 1.20
Other 1.08 1.03 1.13 0.90
Parental employment (ref. = full-time)
Jobless 1.09 1.05 1.45 0.83
Part-time only 1.26 * 1.12 1.25 0.90
Housing tenure (ref. = own/buy)
Renting 1.55 *** 1.12 1.19 1.16 *
Other 1.67 * 1.41 * 1.28 0.94
Parental income (ref. = middle)
Lowest quintile 0.87 1.15 0.88 1.15
Highest quintile 1.01 0.97 1.07 0.95
Education level of primary carer (ref. = highest)
Lowest 1.20 0.81 * 1.22 1.15
Middle 1.10 0.96 1.18 1.04
Age of primary carer (ref. = middle)
Younger 0.93 0.94 0.55 *** 1.31 ***
Older 1.10 1.06 1.31 * 1.02
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born 0.99 1.05 1.01 0.87 *
Main language other than English (Australia- or overseas-born) 0.80 0.87 0.88 0.89
Region (ref. = major cities)
Inner regional 1.22 * 1.09 1.02 0.97
Outer regional 1.12 1.04 0.99 1.11
Remote/very remote 0.84 0.92 0.89 0.84
Age of youngest child (years, continuous) 1.01 0.99 1.03 1.01
Constant 0.09 *** 0.23 *** 0.04 *** 0.29 ***
Sample size ( N) 8,022 8,022 8,022 8,022

Note: Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income information. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 19: Multivariate analyses of socio-demographic characteristics and life events related to family and household composition, odds ratios
Characteristics as measured in previous wave Was pregnant or had baby Had someone new move into the household Started living with a new partner/ spouse Had a separation due to rel./ marital difficulties Broke off a steady romantic relationship
Family type (ref. = couple, married)
Couple, cohabiting 1.38 * 1.09 2.27 * 2.29 *** 3.48 ***
Single parent 1.07 1.40 16.56 *** 1.24 6.66 ***
Main source income (ref. = wages)
Government support 1.63 * 2.07 *** 1.25 1.37 1.21
Other 1.12 1.29 0.94 0.63 1.20
Parental employment (ref. = full-time)
Jobless 1.00 0.72 0.58 1.13 0.69
Part-time only 0.91 0.87 0.59 0.95 1.04
Housing tenure (ref. = own/buy)
Renting 1.34 * 1.37 ** 1.20 1.39 * 1.52 **
Other 0.87 1.04 0.74 1.02 0.75
Parental income (ref. = middle)
Lowest quintile 0.94 0.64 ** 0.99 1.13 1.21
Highest quintile 1.14 1.03 1.37 1.07 0.87
Education level of primary carer (ref. = highest)
Lowest 0.89 1.54** 0.90 1.60* 1.77
Middle 1.03 1.36* 0.99 1.50* 1.69*
Age of primary carer (ref. = middle)
Younger 2.47 *** 1.69 *** 1.95 ** 1.50 ** 1.50 **
Older 0.36 *** 0.92 0.46 * 0.84 0.84
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born 0.89 1.43 ** 1.20 0.81 1.28
Main language other than English (Australia- or overseas-born) 0.91 1.40 * 0.96 0.71 0.71
Region (ref. = Major cities)
Inner regional 0.66 ** 0.94 1.21 1.68 *** 1.20
Outer regional 0.70 ** 0.95 1.27 0.91 0.88
Remote/very remote 0.90 1.85 ** 2.20 0.93 0.87
Age of youngest child (years, continuous) 0.86 *** 1.10 *** 1.16 ** 1.03 0.99
Constant 0.12 *** 0.03 *** 0.00 *** 0.02 *** 0.01 ***
Sample size ( N) 8,022 8,021 8,021 8,021 8,021

Note: Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income information. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 20: Multivariate analyses of socio-demographic characteristics and life events related to financial and social matters, odds ratios
Characteristics as measured in previous wave Had a major financial crisis Had something of value lost or stolen Had someone with an alcohol or drug problem Lost job, but not from choice
Family type (ref. = couple, married)
Couple, cohabiting 1.70 *** 1.34 2.48 *** 1.28
Single parent 1.82 *** 2.01 *** 1.17 0.97
Main source income (ref. = wages)
Government support 1.17 1.21 1.60 0.76
Other 1.53 *** 1.01 1.23 0.48 ***
Parental employment (ref. = full-time)
Jobless 0.72 0.97 1.27 0.92
Part-time only 0.89 1.03 1.05 1.00
Housing tenure (ref. = own/buy)
Renting 1.74 *** 1.44 ** 1.54 ** 1.44 ***
Other 1.08 0.69 1.54 0.84
Parental income (ref. = middle)
Lowest quintile 1.45 ** 1.09 1.46 0.91
Highest quintile 0.76 * 1.16 0.89 0.88
Education level of primary carer (ref. = highest)
Lowest 1.50 ** 0.72 1.45 1.16
Middle 1.47 *** 0.92 1.02 1.12
Age of primary carer (ref. = middle)
Younger 1.02 1.34 * 1.01 1.09
Older 0.95 1.16 0.91 0.95
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born 0.97 0.90 1.09 1.02
Main language other than English (Australia- or overseas-born) 0.96 0.92 0.95 1.11
Region (ref. = major cities)
Inner regional 1.01 0.78 1.22 0.95
Outer regional 0.88 0.83 0.97 0.78 *
Remote/very remote 0.70 1.15 0.54 0.83
Age of youngest child (years, continuous) 0.97 0.97 0.99 1.00
Constant 0.06 *** 0.06 *** 0.01 *** 0.07 ***
Sample size ( N) 8,021 8,021 8,021 8,021

Note: Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income information. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 21: Multivariate analyses of socio-demographic characteristics and life events related to residential matters, odds ratios
Characteristics as measured in previous wave Moved house Lived in a drought-affected area Had home or local area affected by bushfire, flooding or severe storm
Family type (ref. = couple, married)
Couple, cohabiting 1.08 0.76 1.04
Single parent 1.86 *** 1.10 1.18
Main source income (ref. = wages)
Government support 0.85 1.57 1.25
Other 1.01 1.95 *** 1.44 *
Parental employment (ref. = full-time)
Jobless 0.74 0.46 ** 0.83
Part-time only 1.05 0.67 * 1.10
Housing tenure (ref. = own/buy)
Renting 3.40 *** 0.94 0.97
Other 2.49 *** 1.86 ** 1.38
Parental income (ref. = middle)
Lowest quintile 0.72 ** 1.08 0.76
Highest quintile 1.42 *** 0.59 *** 1.04
Education level of primary carer (ref. = highest)
Lowest 1.18 0.70 * 1.02
Middle 1.03 0.83 0.91
Age of primary carer (ref. = middle)
Younger 1.35 *** 0.80 0.71 **
Older 0.76 ** 1.22 0.84
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born 0.96 0.75 * 0.91
Main language other than English (Australia- or overseas-born) 0.83 0.46 *** 0.44 ***
Region (ref. = major cities)
Inner regional 1.09 4.08 *** 1.41 **
Outer regional 1.07 8.00 *** 2.26 ***
Remote/very remote 1.11 10.41 *** 2.09 ***
Age of youngest child (years, continuous) 0.98 0.99 0.99
Constant 0.12 *** 0.04 *** 0.06 ***
Sample size ( N) 8,021 8,021 8,021

Note: Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income information. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 22: Multivariate analyses of socio-demographic characteristics and the number of life events, OLS
Characteristics as measured in previous wave Overall mean Excluding new baby, new partner, other new household member Also excluding drought, storms/floods
Family type (ref. = couple, married)
Couple, cohabiting 0.3 *** 0.3 *** 0.3 ***
Single parent 0.4 *** 0.2 *** 0.2 ***
Main source income (ref. = wages)
Government support 0.3 *** 0.2 * 0.1 *
Other 0.1 0.1 0.0
Parental employment (ref. = full-time)
Jobless –0.2 * –0.1 –0.1
Part-time only –0.0 0.0 0.0
Housing tenure (ref. = own/buy)
Renting 0.5 *** 0.4 *** 0.5 ***
Other 0.3 ** 0.3 *** 0.2 *
Parental income (ref. = middle)
Lowest quintile 0.0 0.0 0.1
Highest quintile 0.0 0.0 0.0
Education level of primary carer (ref. = highest)
Lowest 0.1 0.1 0.1 *
Middle 0.1 0.0 0.1
Age of primary carer (ref. = middle)
Younger 0.2 *** 0.1 * 0.1 ***
Older –0.0 –0.0 –0.0
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born –0.0 –0.1 –0.0
Main language other than English (Australia- or overseas-born) –0.2 *** –0.2 *** –0.1 **
Region (ref. = major cities)
Inner regional 0.1 *** 0.2 *** 0.1
Outer regional 0.2 *** 0.2 *** 0.0
Remote/very remote 0.2 * 0.1 –0.1
Age of youngest child (years, continuous) –0.0 –0.0 –0.0
Constant 1.1 *** 1.0 *** 0.8 ***
Sample size ( N) 8,021 8,021 8,021

Note: Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income information. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Prevalence of life events by characteristics: Detailed results

Family form and composition: Relationship status

To examine how life events vary for families of different forms, families were classified according to the parental relationship (of resident parents) as:

  • married couples;
  • cohabiting couples; or
  • single parents.

Couple (married or cohabiting) families include those in which there are two adults in a marriage or marriage-like relationship, even if one of these adults is not considered to be a parent to any resident children. In the case of single parents, we make no distinction according to the living arrangements of resident children, who may spend some of their time living in the home of a non-resident parent. Across the pooled data, at Wave 3 (since we use these characteristics to examine life events reported at Wave 4), 74% of parents were married, 12% were cohabiting and 15% were single parents.15 (See Appendix E).

Table 23 indicates that the prevalence of a number of life events varied with relationship status (the statistical significances are also shown). Table 24 presents the average number of life events experienced by parents in these three relationship status groups.

Table 23: Prevalence of parents’ life events by parental relationship status
Parents’ life events in previous 12 months Married (%) Cohabiting (%) Single parent (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 10.6 15.0 15.6 12.1 ***
Had a serious illness, injury or assault happen to a close relative 18.7 22.5 21.7 19.7 *
Had a parent, partner or child die 5.3 6.1 4.9 5.4
Had a close family friend or another relative die 24.2 28.8 25.1 25.4 *
Family and household composition
Was pregnant or had a child 6.1 12.7 8.9 7.5 ***
Had someone new (other than a new baby or partner) move into the household 5.5 6.7 13.0 7.0 ***
Started living with a new partner/spouse 0.6 1.6 8.3 2.0 ***
Had a separation due to relationship or marital difficulties 3.7 11.1 8.2 5.5 ***
Broke off a steady romantic relationship 1.2 6.2 13.5 3.9 ***
Financial and social matters
Had a major financial crisis 8.2 16.2 21.7 11.7 ***
Lost job, but not from choice 8.2 11.6 8.0 8.4 *
Had something of value lost or stolen 5.0 8.0 12.7 6.8 ***
Had someone in the household with an alcohol or drug problem 2.1 7.3 6.5 3.5 ***
Residential matters
Moved house 13.8 19.2 26.4 16.9 ***
Lived in a drought-affected area 7.3 6.6 6.1 7.1
Had home or local area affected by bushfire, flooding or severe storm 6.3 6.9 6.1 6.5
Sample size ( N) 6,244 847 975 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

In the multivariate analyses, the life events most strongly related to relationship status were those that represented changes in family composition or relationship, followed by the set of events listed under the heading “financial and social matters”. The key findings that emerged from the multivariate analyses are listed below.

Events pertaining to family and household composition issues varied significantly with relationship status in the following ways:

  • Cohabiting parents were the most likely to experience a pregnancy or have a child, followed by single parents and then married parents.
  • Single parents were the most likely to have somebody new, other than a new baby or partner, move into the household.
  • Single parents were also the most likely to commence living with a new partner or spouse.
  • A separation due to relationship or marital difficulties, or having a romantic relationship end were more likely to be reported by cohabiting than married parents.

Compared to married parents, cohabiting parents were more likely to have had:

  • a major financial crisis;
  • either parent lose their job, not from choice; and
  • someone in the household with an alcohol or drug problem.

Compared to married parents, single parents were more likely to have had:

  • a major financial crisis;
  • something of value lost or stolen; and
  • moved house.

These various findings, combined, were also reflected in married couples experiencing, on average, a lesser number of life events than cohabiting couples and single parents (Table 24).

Table 24: Mean number of life events by parental relationship status
Mean number of life events in previous 12 months Married Cohabiting Single parent Total
Mean
Overall mean 1.27 1.87 2.06 1.49 ***
Excluding new baby, new partner, other new household member 1.15 1.66 1.77 1.33 ***
Also excluding drought, storms/floods 1.01 1.52 1.64 1.19 ***
Sample size ( N) 6,244 847 975 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in finval column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Primary carer’s age

Age of primary carer was explored to examine, in particular, whether younger parents were at greater risk than other parents of experiencing life events. Primary carers were classified into one of three groups: “youngest”, “mid-age” and “oldest”. This corresponded to different ages in each of the cohorts: B-cohort: youngest were aged under 32 years old, mid-age 32–40 years, oldest 40 or more years; K-cohort: youngest were aged under 36 years, mid-age 36–43 years, oldest 44 or more years. These age groups were designed to place approximately 20% of primary carers in each sample in the “youngest” group and 20% in the oldest group. The actual percentages vary, due to the distribution of ages in the sample (see Table E1).

The overall results are shown in Table 25, which show that experiences of most of the life events examined varied significantly according to the age of the primary carer, with the youngest parents being the most likely to experience all except one of these events (outlined below).

Table 25: Prevalence of parents’ life events, by age of primary carer
Parental life events in previous 12 months Youngest primary carers (%) Mid-age primary carers (%) Oldest primary carers (%) Total (%)
Health concerns or the death of someone close
Sufered a serious illness, injury or assault 12.6 11.6 11.8 12.1
Had a serious illness, injury or assault happen to a close relative 20.1 19.4 19.6 19.7
Had a parent, partner or child die 3.2 5.7 7.3 5.4 ***
Had a close family friend or another relative die 29.8 23.4 22.9 25.4 ***
Family and household composition
Was pregnant or had a baby 15.2 5.4 1.8 7.5 ***
Had someone new (other than a new baby or partner) move into the household 9.5 5.9 5.8 7.0 ***
Started living with a new partner/spouse 3.7 1.3 0.9 2.0 ***
Had a separation due to relationship or marital difficulties 8.9 4.2 3.4 5.5 ***
Broke off a steady romantic relationship 6.8 2.7 2.3 3.9 ***
Financial and social matters
Had a major financial crisis 14.8 10.3 9.2 11.7 ***
Lost job, but not from choice 9.7 8.4 7.3 8.4
Had something of value lost or stolen 8.8 5.6 6.6 6.8 ***
Had someone in the household with an alcohol or drug problem 4.7 3.0 2.7 3.5 **
Residential matters
Moved house 23.8 14.7 11.2 16.9 ***
Lived in a drought-affected area 7.0 7.2 6.9 7.1
Had home or local area affected by bushfire, flooding or severe storm 5.5 7.1 5.2 6.5
Sample size ( N) 1,858 4,792 1,483 8,313

Note: Within each cohort of LSAC, age of the primary carer was grouped according to the distribution of ages of primary carers in each sample. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

The multivariate analyses allow us to see whether these differences can be, to some extent, explained by systematic differences in other characteristics of parents in the different age groups. When the various other characteristics were controlled, age of the primary carer continued to be quite a strong predictor of several life events. In other words, age of the primary carer is an important factor predicting families that are particularly like go on to experience certain life events. The most notable finding from these analyses were that the youngest parents were more likely than the other two age groups of parents to experience the following:

  • having a close family friend or relative die (other than a parent, partner or child);
  • breaking off a romantic relationship or having a separation due to relationship or marital difficulties;
  • being pregnant or having a baby;
  • moving house; and
  • having someone new move into the household.

However, the youngest parents were the least likely to have experienced the death of a parent, partner or child. Given the above trends, it is not surprising that young parents also experienced the highest average number of the life events assessed (Table 26).

Table 26: Mean number of life events, by age of primary carer
Mean number of life events in previous 12 months Youngest primary carers Mid-age primary carers Oldest primary carers Total
Mean
Overall mean 1.84 1.36 1.25 1.49 ***
Excluding new baby, new partner, other new household member 1.56 1.23 1.16 1.33 ***
Also excluding drought, storms/floods 1.43 1.09 1.04 1.19 ***
Sample size ( N) 1,858 4,792 1,483 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Age of youngest child in the family

At Wave 3 of LSAC, study children were aged 4–5 years (the B cohort) and 8–9 years (the K cohort). These study children, however, can have younger or older siblings. When the data from the two cohorts are combined, in 56% of families the youngest child was aged 0–4 years at Wave 3, in 21% the youngest child was aged 5–7 years and in 23% the youngest child was aged 8–11 years.

As all families in LSAC had relatively young children, we would not expect a great deal of variation in the experience of life events according to the age of the youngest child in the family (see Table 27). The only life events that varied significantly according to this factor in the multivariate analyses were:

  • being pregnant or having a baby (more likely in families with relatively young children);
  • having someone other than a new baby or partner move into the household (the likelihood of which increased with increasing age of the youngest child); and
  • starting to live with a new partner/spouse (more likely in families with older children).
Table 27: Prevalence of parents’ life events by age of youngest child
Parents’ life events in previous 12 months 0–4 years (%) 5–7 years (%) 8–9 years (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 11.4 11.8 13.2 12.1
Had a serious illness, injury or assault happen to a close relative 19.8 17.7 20.8 19.7
Had a parent, partner or child die 4.9 4.8 6.9 5.4 **
Had a close family friend or another relative die 25.7 23.0 24.7 25.4
Family and household composition
Was pregnant or had a baby 10.5 4.2 2.4 7.5 ***
Had someone new (other than a new baby or partner) move into the household 6.2 7.5 7.7 7.0
Started living with a new partner/spouse 1.5 2.3 2.5 2.0 **
Had a separation due to relationship or marital difficulties 5.9 3.9 5.1 5.5
Broke off a steady romantic relationship 4.0 3.1 3.6 3.9
Financial and social matters
Had a major financial crisis 11.3 10.3 12.0 11.7
Lost job, but not from choice 8.4 7.9 9.5 8.4
Had something of value lost or stolen 6.7 6.1 6.7 6.8
Had someone in the household with an alcohol or drug problem 3.6 2.7 3.5 3.5
Residential matters
Moved house 17.6 15.0 14.7 16.9 **
Lived in a drought-affected area 6.8 7.0 7.6 7.1
Had home or local area affected by bushfire, flooding or severe storm 6.2 6.5 6.4 6.5
Sample size ( N) 4,512 1,708 1,846 8,313

Note: Total includes families who had missing information about age of youngest child at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Reflecting these findings, the mean number of life events experienced was higher for those whose youngest child was aged 0–4 years or 8–9 years, than for those whose youngest child was aged 5-7 years. This trend applied when all life events were included, and when specified sets were omitted (see Table 28).

Table 28: Mean number of life events, by age of youngest child
Mean number of life events in previous 12 months 0–4 years 5–7 years 8–9 years Total
Mean
Overall mean 1.51 1.34 1.47 1.49 *
Excluding new baby, new partner, other new household member 1.32 1.20 1.35 1.33 *
Also excluding drought, storms/floods 1.19 1.06 1.21 1.19 *
Sample size ( N) 4,512 1,708 1,846 8,313

Note: Total includes families who had missing information about age of youngest child at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Family background: Parents’ country of birth and main language

The only variable explored in respect to family background represented concerned main language spoken at home and parents’ country of birth (Table 29). Families were classified as follows:

  • only Australian-born English speakers;
  • overseas-born English speakers; and
  • one or both parents mainly speak a language other than English.

This information was based on the country of birth and main language spoken at home by either parent. The categories are mutually exclusive such that if either (or both) of the parents mainly speak a language other than English, they were represented in the last of these groups. If neither mainly speaks a language other than English but at least one is overseas-born, they were included in the second category. This leaves all families in which parents are Australian-born and mainly speak English in the first category. Just over half (53%) of the families in the Wave 3 sample in these analyses included parents who were only Australian-born and mainly spoke English. Another 31% included a parent who was overseas-born, but the parents mainly spoke English. There were 16% in the final category, in which either or both parents mainly spoke a language other than English.

According to the multivariate analyses, the likelihood of experiencing the different life events did not vary much according to this factor. Overall, families in which one or both parents mainly spoke a language other than English had a slightly lower likelihood of experiencing the life events examined. This was most apparent for the regionally related matters (living in a drought-affected area, having had home or local area affected by bushfires/flooding/severe storms), possibly because migrant families are more likely to live in major cities of Australia, rather than in regional and remote areas, where these events are more likely to occur. In addition, compared with the Australian-born English speakers, the overseas-born English-speakers were more likely than the Australian-born to have had someone (other than spouse/partner or child) join their household, started living with a new partner/spouse, and broken off a steady romantic relationship.

Other differences according to parental country of birth and main language that are evident in Table 29, such as moving house, were not statistically significant in the multivariate analyses.

Table 29: Prevalence of parents’ life events by country of birth and language
Parents’ life events in previous 12 months Both parents Australian born, English-speaking (%) Either or both parent overseas born, main language English (%) Either or both parents main language not English (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 11.9 13.1 9.4 12.1 *
Had a serious illness, injury or assault happen to a close relative 19.4 21.3 16.9 19.7
Had a parent, partner or child die 5.5 5.3 4.8 5.4
Had a close family friend or another relative die 26.4 23.1 23.7 25.4
Family and household composition
Was pregnant or had a baby 5.9 8.2 5.7 7.5
Had someone new (other than a new baby or partner) move into the household 5.3 9.4 6.5 7.0 ***
Started living with a new partner/spouse 0.9 4.1 0.8 2.0 ***
Had a separation due to relationship or marital difficulties 5.7 5.6 3.4 5.5
Broke off a steady romantic relationship 2.2 7.2 1.6 3.9 ***
Financial and social matters
Had a major financial crisis 9.7 14.4 10.0 11.7 ***
Lost job, but not from choice 8.6 8.2 9.1 8.4 **
Had something of value lost or stolen 5.9 8.2 5.7 6.8 **
Had someone in the household with an alcohol or drug problem 3.0 4.3 2.8 3.5
Residential matters
Moved house 15.9 19.5 11.9 16.9 ***
Lived in a drought-affected area 9.4 5.7 2.2 7.1 ***
Had home or local area affected by bushfire, flooding or severe storm 7.7 6.1 2.3 6.5 ***
Sample size ( N) 4,604 2,458 997 8,313

Note: Total includes families who had missing information about parental country of birth. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

The mean number of life events varies significantly with country of birth and language, with the overseas-born English-speakers indicating that the greatest number of life events and those who mainly spoke (or whose partner mainly spoke) a language other than English reporting the smallest number of life events (Table 30).

Table 30: Mean number of life events by country of birth and language
Mean number of life events in previous 12 months Both parents Australian born, English-speaking Either or both parent overseas born, main language English Either or both parents main language not English Total
Mean
Overall mean 1.45 1.63 1.18 1.49 ***
Excluding new baby, new partner, other new household member 1.31 1.42 1.04 1.33 ***
Also excluding drought, storms/floods 1.14 1.30 0.99 1.19 ***
Sample size ( N) 4,604 2,458 997 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Family socio-economic circumstances

Five indicators of socio-economic circumstances were included in the analyses: household income, main source of income, parental employment, the primary carer’s education, and housing tenure. While the results for each of these indicators are discussed sequentially, the indicators are, of course, inter-related.

The quite consistent finding across these analyses is that families living in poorer socio-economic circumstances were more likely to:

  • experience a pregnancy or have a baby;
  • experience a major financial crisis; and
  • move house.
Household income

Families were classified according to the total household income, with three groups created to identify families with the lowest incomes (up to $996 gross income per week), mid-range incomes ($996–2,550 gross income per week), and highest incomes (> $2,550 gross income per week). These categories placed 20% of the sample in the lowest income category, 54% in the mid-range income category and 16% in the highest income category.

Table 31 shows that the experience of several life events varied somewhat with household income, but some these differences were not statistically significant in the multivariate analyses. This suggests that these relationships could be explained by other characteristics examined in the analyses (such as housing tenure, ages of parents). Table 20 shows that families with the lowest incomes were significantly more likely to have experienced a major financial crisis.

Table 31: Prevalence of parents’ life events by household income
Parents’ life events in previous 12 months Lowest household income (%) Mid-range household income (%) Highest household income (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 14.4 11.6 9.9 12.1 *
Had a serious illness, injury or assault happen to a close relative 20.9 19.0 18.9 19.7
Had a parent, partner or child die 5.0 5.1 5.5 5.4
Had a close family friend or another relative die 27.5 24.6 21.8 25.4 **
Family and household composition
Was pregnant or had a baby 9.6 6.7 5.8 7.5 *
Had someone new (other than a new baby or partner) move into the household 8.5 6.5 5.9 7.0 *
Started living with a new partner/spouse 4.4 1.1 0.8 2.0 ***
Had a separation due to relationship or marital difficulties 7.5 4.7 3.3 5.5 ***
Broke off a steady romantic relationship 8.2 2.4 1.1 3.9 ***
Financial and social matters
Had a major financial crisis 18.3 9.8 6.4 11.7 ***
Lost job, but not from choice 8.0 9.0 7.9 8.4
Had something of value lost or stolen 9.7 5.5 5.7 6.8 ***
Had someone in the household with an alcohol or drug problem 6.2 2.6 1.6 3.5 ***
Residential matters
Moved house 17.9 15.6 16.4 16.9 *
Lived in a drought-affected area 7.1 8.0 4.4 7.1 ***
Had home or local area affected by bushfire, flooding or severe storm 5.4 6.8 6.3 6.5
Sample size ( N) 1,383 4,476 1,503 8,313

Note: Total includes families who had missing information about parental income at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 32 shows that the average number of life events decreased with increasing income. However, these differences were not statistically significant in the multivariate analyses (Table 22), suggesting that this trend could be explained by other characteristics distinguishing these families (for example, main source of income, which is explored in the next subsection).

Table 32: Mean number of life events by household income
Mean number of life events in previous 12 months Lowest household income Mid-range household income Highest household income Total
Mean
Overall mean 1.79 1.39 1.22 1.49 ***
Excluding new baby, new partner, other new household member 1.56 1.25 1.09 1.33 ***
Also excluding drought, storms/floods 1.44 1.10 0.99 1.19 ***
Sample size ( N) 1,383 4,476 1,503 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Main source of household income

For these analyses, families were classified according to whether their main source of income was from:

  • wages/salary;
  • government support; or
  • other income.

If wages or salary represented the main source of income of either or both parents, the family’s main income source was deemed to be wages/salary (80% of the sample at Wave 3). Where this did not apply, and where either or both parents had income other than government support, they were classified as having “other” income as their main source (9% of the Wave 3 sample). The latter group includes families with income from self-employment.. The government support category, then, includes single parents who nominated government support as their main income, and couple families in which both parents nominated government support as their main source of income (11% of the Wave 3 sample).

According to Table 33, compared with other families, those that relied on government support as their main source of income were at a greatest risk of experiencing most of life events listed (though they were the least likely to have lived in drought-affected areas or to have had their home or area affected by bushfire, flooding or a severe storm). To some extent these trends were related to factors that co-occur with reliance on government support (e.g., having a low income, being jobless and, in many cases, being a single parent), since a more limited sets of life events varied significantly with main source of income in the multivariate analyses (Table 18 to Table 21). According to the multivariate analyses, These analyses found that, compared with their wage-earning counterparts, parents who relied on government support as their main source of income were more likely to have:

  • experienced a pregnancy or had a baby; and
  • had someone other than a baby or partner move into the household.
Table 33: Prevalence of parents’ life events by main source of household income
Parents’ life events in previous 12 months Wages (%) Government payments (%) Other (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 11.0 17.9 11.5 12.1 ***
Had a serious illness, injury or assault happen to a close relative 19.1 23.3 18.7 19.7
Had a parent, partner or child die 5.1 6.0 6.1 5.4
Had a close family friend or another relative die 24.4 28.8 24.0 25.4
Family and household composition
Was pregnant or had a baby 6.4 12.3 8.8 7.5 ***
Had someone new (other than a new baby or partner) move into the household 5.8 13.3 6.9 7.0 ***
Started living with a new partner/spouse 1.4 6.1 1.0 2.0 ***
Had a separation due to relationship or marital difficulties 4.8 10.2 3.4 5.5 ***
Broke off a steady romantic relationship 2.7 11.3 2.7 3.9 ***
Financial and social matters
Had a major financial crisis 9.5 20.1 14.2 11.7 ***
Lost job, but not from choice 9.1 7.1 5.0 8.4 ***
Had something of value lost or stolen 5.9 11.8 5.8 6.8 ***
Had someone in the household with an alcohol or drug problem 2.7 8.3 2.8 3.5 ***
Residential matters
Moved house 15.8 22.1 13.9 16.9 ***
Lived in a drought-affected area 6.6 6.2 12.3 7.1 ***
Had home or local area affected by bushfire, flooding or severe storm 6.2 5.7 8.8 6.5 *
Sample size ( N) 6,679 649 712 8,313

Note: Total includes families who had missing information about main source of income at Wave 3. Other includes those with income from a business. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No star indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

These parents also experienced a greater number of the life events examined, compared with wage-earning parents (Table 34).

Table 34: Mean number of life events by main source of household income
Mean number of life events in previous 12 months Wages/salary Government payments Other Total
Mean
Overall mean 1.36 2.10 1.46 1.49 ***
Excluding new baby, new partner, other new household member 1.23 1.79 1.29 1.33 ***
Also excluding drought, storms/floods 1.10 1.67 1.08 1.19 ***
Sample size ( N) 6,679 649 712 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Parents whose main source of income was classified as “other” (i.e., neither wages nor government payments) were less likely than those who were relying on wages/income to have:

  • lost their jobs,
  • but more than this other group to have:
  • had a major financial crisis;
  • been living in a drought-affected area; and
  • had their home or local area affected by bushfire, flooding or severe storm.

The latter two events probably applied disproportionately to parents in this category, because many of them would have been primary producers.

Parental employment

Families were classified according to the employment status of the parent(s), using the following categories:

  • jobless family;
  • part-time employment only; and
  • full-time employment.

Jobless families included couple families in which neither parent was in paid work and single parents without paid work. Overall, 8% of families in the Wave 3 sample were jobless. The second category includes those in which one or both parents have a part-time job but no parent has a full-time job (14% of the Wave 3 sample). The final category includes those families in which one or both parents had a full-time job (79% of the Wave 3 sample).

Table 35 suggests that families in which neither parent had full-time employment were at greater risk than other families of experiencing a range of life events. However, in the multivariate analyses, which included related factors such as main source of income, most of these results were not statistically significant.

Table 35: Prevalence of life events by parental employment
Parents’ life events in previous 12 months Jobless parent(s) (%) Parents only employed part-time hours (%) Full-time employed parent(s) (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 16.1 16.2 10.6 12.1 ***
Had a serious illness, injury or assault happen to a close relative 22.4 22.2 18.8 19.7
Had a parent, partner or child die 6.7 5.7 5.1 5.4
Had a close family friend or another relative die 25.7 26.9 24.5 25.4
Family and household composition
Was pregnant or had a baby 11.4 8.5 6.6 7.5 **
Had someone new (other than a new baby or partner) move into the household 12.9 7.7 5.9 7.0 ***
Started living with a new partner/spouse 6.2 3.0 1.2 2.0 ***
Had a separation due to relationship or marital difficulties 9.7 6.5 4.6 5.5 ***
Broke off a steady romantic relationship 11.5 6.9 2.3 3.9 ***
Financial and social matters
Had a major financial crisis 20.3 15.6 9.5 11.7 ***
Lost job, but not from choice 7.3 8.3 8.7 8.4
Had something of value lost or stolen 12.0 8.5 5.6 6.8 ***
Had someone in the household with an alcohol or drug problem 8.4 5.0 2.5 3.5 ***
Residential matters
Moved house 21.4 20.0 15.2 16.9 ***
Lived in a drought-affected area 4.6 5.8 7.6 7.1 **
Had home or local area affected by bushfire, flooding or severe storm 5.5 6.6 6.4 6.5
Sample size ( N) 482 1,099 6,575 8,313

Note: Total includes families who had missing information about parental employment at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No star indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

On average, jobless families indicated that a significantly greater number of life events than families with full-time work (Table 36)—a trend that remained statistically significant when the effects of other characteristics were controlled (Table 22).

Table 36: Mean number of life events by parental employment
Mean number of life events in previous 12 months Jobless parent(s) Parents only employed part-time hours Full-time employed parent(s) Total
Mean
Overall mean 2.02 1.74 1.35 1.49 ***
Excluding new baby, new partner, other new household member 1.72 1.54 1.21 1.33 ***
Also excluding drought, storms/floods 1.62 1.42 1.07 1.19 ***
Sample size ( N) 482 1,099 6,575 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Primary carer’s education

The primary carer’s level of educational attainment was classified as:

  • incomplete secondary only;
  • secondary, certificate or diploma; or
  • degree or higher.

At Wave 3, 21% of primary carers had left school early and had no post-school qualification (here called “incomplete secondary only”), 56% had completed their secondary education, or achieved a certificate or a diploma, and 23% had a degree or higher qualification.

As with other indicators of socio-economic circumstances, Table 37 shows that families in which the primary carer had a degree or higher qualification were the least likely to experience a range of life events, but some of these differences could be explained by other socio-demographic factors examined.

Table 37: Prevalence of parents’ life events by primary carer’s education
Parents’ life events in previous 12 months Incomplete secondary (%) Secondary, certificate or diploma (%) Degree or higher (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 13.8 12.1 9.6 12.1 **
Had a serious illness, injury or assault happen to a close relative 18.7 19.8 19.9 19.7
Had a parent, partner or child die 6.1 5.1 5.2 5.4
Had a close family friend or another relative die 27.9 24.7 22.5 25.4 *
Family and household composition
Was pregnant or had a baby 7.5 7.9 5.8 7.5 *
Had someone new (other than a new baby or partner) move into the household 8.8 7.1 4.1 7.0 ***
Started living with a new partner/spouse 2.3 2.3 0.5 2.0 ***
Had a separation due to relationship or marital difficulties 6.6 5.9 2.5 5.5 ***
Broke off a steady romantic relationship 5.3 4.2 0.8 3.9 ***
Financial and social matters
Had a major financial crisis 14.2 12.2 6.0 11.7 ***
Lost job, but not from choice 8.9 8.8 7.5 8.4
Had something of value lost or stolen 6.3 6.9 5.9 6.8
Had someone in the household with an alcohol or drug problem 5.3 3.2 1.9 3.5 ***
Residential matters
Moved house 18.9 16.9 12.9 16.9 ***
Lived in a drought-affected area 6.9 7.2 6.7 7.1
Had home or local area affected by bushfire, flooding or severe storm 7.0 6.1 6.2 6.5
Sample size ( N) 1,230 4,536 2,300 8,313

Note: Total includes families who had missing information about parental education at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

The multivariate analyses suggest that the primary carers in the two lowest educational groups (those without a degree) were more likely than those with a degree or higher qualification to have experienced the following events:

  • a separation due to marital or relationship difficulties;
  • a major financial crisis;
  • a new household member (other than baby or partner).

In addition, primary carers who had left school early and had not achieved any post-school qualification were less likely than those with a degree or higher qualification to have:

  • experienced a serious illness, injury or assault; and
  • been living in a drought-affected area.

Finally, primary carers who had completed their secondary education or had achieved a certificate or diploma were more likely than those with a degree or higher qualification to report having:

  • broken off a steady romantic relationship.

While Table 38 indicated that the average number of life events experienced decreased with increasing educational attainment level, the multivariate analyses (Table 22) suggest that educational attainment level is not a particularly important variable in explaining the variation in the number of life events. Again, this would be explained by the fact that parental education would be strongly related to other variables included in the analyses, such as family type, housing tenure and main source of income.

Table 38: Mean number of life events by primary carer’s education
Mean number of life events in previous 12 months Incomplete secondary Secondary, certificate or diploma Degree or higher Total
Mean
Overall mean 1.64 1.51 1.18 1.49 ***
Excluding new baby, new partner, other new household member 1.46 1.33 1.08 1.33 ***
Also excluding drought, storms/floods 1.32 1.20 0.95 1.19 ***
Sample size ( N) 1,230 4,536 2,300 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Housing tenure

The final socio-economic indicator explored in these analyses is housing tenure. Housing tenure provides some indication of the assets held by the families (i.e., family wealth), and is therefore a useful addition to the other indicators of socio economic circumstances. For these analyses, families were classified as:

  • owning or buying their home;
  • renting or boarding); and
  • other.

Because only a very small proportion of families were in “other” housing arrangements (2% at Wave 3), the discussion below focuses on the comparison of owners/buyers (71% of families at Wave 3) and renters/boarders (27% of families at Wave 3).

In the multivariate analyses, housing tenure was an important predictor of a number of the life events (Table 18 to Table 21). Compared to owners/buyers (Table 39), those who were renting/boarding were more likely to:

  • suffer a serious illness, injury or assault;
  • have a close friend or other relative die;
  • become pregnant or have a baby;
  • have someone new move into the household;
  • have a separation due to relationship or marital difficulties;
  • break off a steady romantic relationship
  • have a major financial crisis;
  • lose a job;
  • have something valuable lost or stolen;
  • have a household member with an alcohol or drug problem; and
  • move house.
Table 39: Prevalence of parents’ life events by housing tenure
Parents’ life events in previous 12 months Owns or buying own home (%) Renting or boarding (%) Other (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 10.0 16.5 15.1 12.1 ***
Had a serious illness, injury or assault happen to a close relative 18.5 21.7 26.1 19.7 *
Had a parent, partner or child die 5.2 5.5 5.5 5.4
Had a close family friend or another relative die 23.2 29.2 26.6 25.4 **
Family and household composition
Was pregnant or had a baby 5.5 11.9 6.5 7.5
Had someone new (other than a new baby or partner) move into the household 5.2 10.7 6.4 7.0 ***
Started living with a new partner/spouse 1.0 4.1 2.1 2.0 ***
Had a separation due to relationship or marital difficulties 4.0 8.4 6.8 5.5 ***
Broke off a steady romantic relationship 1.8 8.3 4.3 3.9 ***
Financial and social matters
Had a major financial crisis 8.4 18.7 8.6 11.7 ***
Lost job, but not from choice 7.8 10.7 6.1 8.4 ***
Had something of value lost or stolen 5.3 9.9 4.4 6.8 ***
Had someone in the household with an alcohol or drug problem 2.2 5.9 6.3 3.5 ***
Residential matters
Moved house 10.9 29.5 24.6 16.9 ***
Lived in a drought-affected area 7.2 6.0 14.5 7.1 ***
Had home or local area affected by bushfire, flooding or severe storm 6.4 6.0 8.2 6.5
Sample size ( N) 6,082 1,781 203 8,313

Note: Total includes families who had missing information about housing tenure at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 2) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No stars indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

It is therefore not surprising that those who were renting/boarding experienced a significantly greater number of the life events examined than those who were owners/buyers (Table 40)—a trend that remained significant when the effects of the other characteristics were controlled (Table 22).

Table 40: Mean number of life events by housing tenure
Mean number of life events in previous 12 months Owns or buying own home Renting or boarding Other Total
Mean
Overall mean 1.23 2.03 1.72 1.49 ***
Excluding new baby, new partner, other new household member 1.11 1.76 1.57 1.33 ***
Also excluding drought, storms/floods 0.97 1.64 1.34 1.19 ***
Sample size ( N) 6,082 1,781 203 8,313

Note: Total includes families who had missing information about relationship status at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; **p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Residential location: Remoteness of region

The remoteness of the area in which families lived represents the only other characteristic examined in these analyses. The measure of remoteness is based upon an underlying Accessibility/Remoteness Index of Australia Plus (ARIA+) score, which is derived from information about road distances from major service centres (Glover & Tennant, 2003). Note that while we included a focus on families who were living in remote or very remote parts of Australia, caution should be used in evaluating these results, as LSAC was not designed to be representative of families living in these regions of Australia. Families were classified as living in one of the following areas:

  • major city;
  • inner regional;
  • outer regional; and
  • remote/very remote.

The key finding emerging from this analysis was that the more remote the area in which the families lived (as assessed in Wave 3), the greater was their likelihood of reporting in Wave 4 that, during the previous 12 months, they had been living in a drought-affected area, or that their home or local area had been affected by bushfire, flooding or severe storm (see Table 41). The first of these circumstances continued to hold when multivariate analyses were applied to the data. That is, the more remote the area, the more likely were they to have been living in a drought-affected area. However, the results based on multivariate analyses suggested that, while those in major cities were significantly less likely than each of the other groups to subsequently report that their home or local area had been affected by bushfire, flooding or severe storm, those in outer regional areas were the most likely to report this (Table 21).

Table 41: Prevalence of parents’ life events by remoteness
Parents’ life events in previous 12 months Major cities (%) Inner regional (%) Outer regional (%) Remote or very remote (%) Total (%)
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 10.9 14.6 12.7 9.7 12.1 *
Had a serious illness, injury or assault happen to a close relative 19.1 20.5 20.7 18.4 19.7
Had a parent, partner or child die 5.2 5.7 5.5 4.6 5.4
Had a close family friend or another relative die 24.2 24.4 29.1 23.9 25.4 *
Family and household composition
Was pregnant or had a baby 7.7 5.9 7.2 9.5 7.5
Had someone new (other than a new baby or partner) move into the household 6.5 6.9 7.4 10.1 7.0 *
Started living with a new partner/spouse 1.6 2.1 2.6 4.6 2.0 *
Had a separation due to relationship or marital difficulties 4.3 8.1 5.3 5.9 5.5 ***
Broke off a steady romantic relationship 3.3 4.7 3.9 4.7 3.9 ***
Financial and social matters
Had a major financial crisis 10.9 12.7 11.2 8.2 11.7
Lost job, but not from choice 8.8 8.8 7.1 7.9 8.4
Had something of value lost or stolen 6.8 5.6 6.7 8.5 6.8
Had someone in the household with an alcohol or drug problem 3.2 4.0 3.5 2.6 3.5
Residential matters
Moved house 15.4 17.8 18.4 20.1 16.9
Lived in a drought-affected area 2.5 10.7 18.9 22.9 7.1 ***
Had home or local area affected by bushfire, flooding or severe storm 4.4 7.5 12.1 13.0 6.5 ***
Sample size ( N) 4,819 1,702 1,300 233 8,313

Note: Total includes families who had missing information about region at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Chi-square tests (df = 3) were used to test whether prevalence varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001. No star indicates prevalence did not vary according to this characteristic.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Table 41 shows that the likelihood of experiencing a number of other events varied significantly with residential location. Multivariate analyses indicated that the following trends remained statistically significant, net of the effects of all other characteristics examined, those living in major cities differed from one or more of the other groups in the following ways:

  • those in inner regions were more likely to subsequently report that they experienced a serious illness, injury or assault and that they had a separation due to relationship or marital difficulties (Table 18 and Table 19);
  • those in inner and outer regions were less likely to subsequently indicate that they had become pregnant or had a baby (Table 19);
  • those in outer regions were also less likely to indicate that they had lost their job (Table 20); and
  • those in remote areas were more likely to indicate that someone other than their partner or baby had moved into their household (Table 19).

Compared with those living in major cities:

  • those in inner regions were more likely to subsequently report that they experienced a serious illness, injury or assault and that they had a separation due to relationship or marital difficulties (Table 18 and Table 19);
  • those in inner and outer regions were less likely to subsequently indicate that they had become pregnant or had a baby (Table 19);
  • those in outer regions were also less likely to indicate that they had lost their job (Table 20); and
  • those in remote areas were more likely to indicate that someone other than their partner or baby had moved into their household (Table 19).

The mean number of life events, with all life events included in the count, increased with the remoteness of the area. That is, families living in major cities experienced, on average, the smallest number of life events, and those living in remote areas experienced, on average, the largest number of life events (Table 42). This is partly driven by the increasing likelihood of these families experiencing drought, storms or floods. When these life events, along with those pertaining to family composition changes, were excluded from the count, families in inner regional and outer regional parts of Australia experienced the largest number of life events.

Table 42: Mean number of life events by remoteness
Mean number of life events in previous 12 months Major cities Inner regional Outer regional Remote or very remote Total
Mean
Overall mean 1.35 1.60 1.72 1.75 1.49 ***
Excluding new baby, new partner, other new household member 1.19 1.45 1.55 1.50 1.33 ***
Also excluding drought, storms/floods 1.12 1.27 1.24 1.15 1.19 **
Sample size ( N) 4,819 1,702 1,300 233 8313

Note: Total includes families who had missing information about region at Wave 3. Life events were reported in Wave 4, and have been tabulated against characteristics in Wave 3. Analysis of Variance tests were used to test whether the mean number of life events varied across groups. Statistical significance is shown in final column. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Waves 3 and 4, B and K cohorts combined

The related multivariate analyses (summarised in Table 22) indicate that, when all life events were included in the count, the average number of events experienced by those who were living in major cities was significantly lower than the averages derived for all other groups. When the count excludes the addition of a new partner, baby or other person into the household, then the average number of events experienced by those living in major cities remained significantly lower than the averages derived for those in inner and outer regional areas. However, no significant differences between the averages were apparent when drought/storms/flooding experiences were also removed from the count.

3.6 Socio-demographic characteristics, life events and distress

The analyses based on the HILDA dataset (outlined in Section 2) suggested that experiences of certain life events that would generally appear to be unpleasant were associated with falls in wellbeing, often from a relatively low base (relative to those who did not go on to experience the event in question). However, in some cases, wellbeing was already so low before the event, that the event itself was not associated with much change in wellbeing. This was apparent, for example, for women who subsequently separated from their spouse/partner. Many may have chosen separation to end their unhappy relationship, but they would typically face many new problems as a result of separation (e.g., financial loss; need to negotiate property division and perhaps parenting arrangements).

Whereas we focused on several indicators of wellbeing in the HILDA-based analyses, we restricted attention to one indicator of (low) wellbeing in the related LSAC-based analyses: non-specific psychological distress, as measured by the Kessler 6 (Kessler et al., 2002). This six-item measure has been widely used and validated in many epidemiological studies. LSAC respondents were asked to indicate how often in the previous four weeks they had experienced the six symptoms of distress (e.g., felt nervous, hopeless, worthless). Respondents used a 5-point scale, ranging from 0 “none of the time” to 4 “all of the time”. Responses were summed to produce a score between 0 and 24. A higher score therefore indicates a higher level of psychological distress. The analyses here focus on mothers’ level of distress.

Consistent with the approach adopted in the relevant HILDA-based analyses, the following steps were taken in the LSAC-based analyses:

  • First, before assessing the association between life events and distress, we considered how levels of distress measured in 2010 (Wave 4) varied according to the socio-demographic characteristics of parents apparent in the same survey wave. The set of characteristics examined were the same as those discussed in the previous section, but of course the characteristics of some respondents had changed by Wave 4, partly as a result of life events (such as separation from a spouse/partner). Furthermore, the previous analysis focused on the characteristics of all primary carers (most of whom were mothers) while the analyses outlined in this section restrict attention to mothers.
  • Second, for an initial look at how distress was related to life events, net of the relationships between these various socio-demographic characteristics and distress, this model was then extended to include each of the life event indicators. This approach allowed us to see the nature and strength of any association between mothers’ experiences of life events and their levels of distress, when the effects of all other events experienced and mothers’ various socio-demographic characteristics were controlled. As a variant of this approach, a model was then introduced which included the number of life events rather than each of the specific life events. Again, Wave 4 of LSAC is used for these analyses.
  • Finally, we examined mothers’ levels of distress apparent in Wave 3 (in 2008), and changes in distress from Waves 3 to Wave 4, according to whether or not they had experienced each of the events examined. These analyses are described more fully when we present them in the following subsection. The results of some more detailed analyses of these data are presented in Appendix F.

An important point about these various sets of analyses is that the LSAC life events questions ask about whether the primary carer (in virtually all cases the mother) or partner experienced the various life events. Thus, unless the primary carer was single, we assessed the extent to which the experience of a life event that occurred to one or both parents was associated with mothers’ levels of distress in Waves 3 and 4, and with any change in distress levels. While some of the events were clearly family-level events (such as experiencing a financial crisis), and events such as job loss would have financial ramifications for both partners, it seems reasonable to suggest that disturbing events occurring to one partner would be likely to affect the other partner.

Cross-sectional analyses

Socio-demographic characteristics

Ordinary least squares was applied to the data to identify the nature and strength of any links between mothers’ levels of distress and their socio-demographic characteristics, as assessed in Wave 4. As noted above, the characteristics examined in the analyses are the same as those focused upon in the analyses of prevalence of life events. The coefficients presented in Table 43 show the extent to which mothers’ scores on the distress scale (ranging from 0 to 24) varied with each of the characteristics listed, net of the effects of all other characteristics in the model.

Table 43: Multivariate analyses of mothers’ distress, by socio-demographic characteristics and life events (OLS)
Characteristics as measured Wave 4 Socio-demographic characteristics (OLS) Socio-demographic characteristics and specific life events (OLS) Socio-demographic characteristics and number of life events (OLS)
Family type (ref. = couple, married)
Couple, cohabiting 0.47*** 0.28* 0.29*
Single parent 0.92*** 0.58*** 0.74***
Main source income (ref. = wages)
Government support 1.33*** 1.06*** 1.23***
Other 0.04 –0.02 0.01
Parental employment (ref. = full-time)
Jobless 0.54* 0.48* 0.51*
Part-time only 0.06 –0.01 0.00
Housing tenure (ref. = own/buy)
Renting 0.37*** 0.11 0.12
Other –0.04 –0.21 –0.24
Parental income (ref. = middle)
Lowest quintile 0.20 0.00 0.12
Highest quintile –0.30** –0.22* –0.28**
Education level of mother (ref. = highest)
Lowest 0.01 –0.01 0.02
Middle 0.01 –0.02 –0.01
Age of mother (ref. = middle)
Younger 0.20* 0.16 0.13
Older –0.29 –0.17 –0.18
Country of birth/English language (ref. = English-speaking, Australia-born)
English-speaking, overseas-born –0.20 –0.21* –0.18
Main language other than English (Australia- or overseas-born) 0.76*** 0.85*** 0.85***
Region (ref. = Major cities)
Inner regional –0.18 –0.22* –0.25*
Outer regional –0.13 –0.13 –0.22*
Remote/very remote –0.25 –0.24 –0.36
Age of youngest child (years, continuous) 0.01 0.01 0.03
Life events
Suffered a serious illness, injury or assault   0.92***  
Had a serious illness, injury or assault happen to a close relative   0.55***  
Had a parent, partner or child die   0.15  
Had a close family friend or another relative die   0.09  
Was pregnant or had a baby   –0.20  
Had someone new move into the household   0.51**  
Started living with a new partner/spouse   –0.14  
Had a separation due to relationship or marital difficulties   0.28  
Broke off a steady romantic relationship   1.04***  
Had a major financial crisis   2.29***  
Lost job, but not from choice   0.08  
Had something of value lost or stolen   0.68***  
Had someone in the household with an alcohol or drug problem   0.87***  
Moved house   0.06  
Lived in a drought-affected area   0.22  
Had home or local area affected by bushfire, flooding or severe storm   0.03  
Number of life events (ref. = none)
One     0.47***
Two     0.89***
Three     1.13***
Four or more     2.62***
Constant 2.68*** 2.32*** 2.13***
Sample size 8084 8084 8084

Note: Kessler 6 (scale 0 to 24, with higher value = more distressed). Models also include (a) an indicator for K compared to B cohort; and (b) indicators for having missing income. * p < .05; ** p < .01; *** p < .001.

Source: LSAC Wave 4, B and K cohorts combined

The key findings are that levels of distress are higher for the following mothers:

  • those who were single parents or cohabiting, rather than married;
  • those in families in which either or both parents mainly spoke a language other than English;
  • those in families in which the main source of income was government support, or those in jobless families;
  • those with relatively low or mid-range, rather than higher, income; and
  • those who were renting or boarding, rather than those who were owners/buyers.

These mothers were also among those who had a higher risk of experiencing adverse life events. The relatively high distress of these mothers may have been influenced more by their difficult living conditions, than by the life events per se.

Life events and distress

The specific life events were then added to the models in order to identify the nature and strength of any association between mothers’ experience of each life event and her level of distress, when the effects of the various above-mentioned socio-demographic characteristics of the mothers and families were controlled (see the second column of results in Table 43). Compared with other mothers, distress appeared to be higher among mothers who reported that they had experienced the following life events in the previous 12 months:

  • had a major financial crisis;
  • broke off a steady romantic relationship;
  • suffered illness, injury or assault;
  • had someone in the household with an alcohol or drug problem;
  • had something of value lost or stolen;
  • had a serious illness, injury or assault happen to a close relative; and
  • had someone new (other than a new baby or partner) move into the household.

The order in which these life events are listed above reflects the relative strength of the association with the measure of mothers’ psychological distress, with having had a major financial crisis having the strongest association with distress.

Note that life events not appearing on this list may not have shown up as being statistically significant because of the rarity of the event (e.g., having a parent, partner or child die). Absence of a significant trend, therefore, would not indicate that the experience of such events would not be associated with heightened levels of distress.

The third set of results in Table 43 refer to links between the number of life events experienced by mothers and their level of distress. Compared with mothers who did not experience any of the life events examined, those who experienced any of the events indicated significantly high distress, with mothers’ levels of distress increasing with increases in the number of events (i.e., two, three or four or more) experienced. These results are consistent with those that emerged in the relevant analyses of the HILDA dataset.

Longitudinal analyses of life events and distress

Following the approach adopted in the analyses of the HILDA dataset, we examined whether those who experienced life events already had higher distress before the event occurred. Separate analyses were then undertaken to examine the extent to which levels of distress changed for those who experienced an event. First, the mean Wave 3 distress score of mothers who subsequently experienced each event (taken separately) was compared with the mean score derived for mothers who did not go on to experience the event. Second, the mean Wave 4 distress scores of these two groups of mothers were compared, and third, their mean level of change in distress scores (from Wave 3 to Wave 4) were compared.

It is important to appreciate that the way people interpret events will vary according to their circumstances and personalities, and their interpretations can change as the repercussions of events unfold.16 Some events may be seen has entailing mixed blessings, some may be seen as largely positive (such as being pregnant/having a baby), and others, as negative. These analyses allow for the fact that distress levels may diminish.

The results of these analyses are shown in Table 44 and are generally consistent with those that emerged in analyses of the HILDA dataset. The results pertaining to Wave 3 levels of distress show that, for most life events, the mothers who subsequently experienced the event already had higher average distress than other mothers. There were some exceptions to this trend. No significant differences were apparent in the average Wave 3 distress scores of mothers who did and did not subsequently experience the following events:

  • had a parent, partner or child die;
  • was pregnant or had a baby;
  • lived in a drought-affected area; and
  • had home or local area affected by bushfire, flooding or severe storm.
Table 44: Mothers’ mean level of distress at Waves 3 and 4, and mean change across waves, by those who did and did not experience each life event
Wave 4: Parents’ life events in previous 12 months Experienced event Did not experience event Tests
Wave 3 Wave 4 Change Wave 3 Wave 4 Change
Health concerns or the death of someone close
Suffered a serious illness, injury or assault 4.25 4.75 +0.50 3.37 3.17 –0.21 a,b,c
Had a serious illness, injury or assault happen to a close relative 3.97 4.17 +0.19 3.35 3.15 –0.20 a,b,c
Had a parent, partner or child die 3.43 3.74 +0.31 3.48 3.33 –0.15  
Had a close family friend or another relative die 3.74 3.64 –0.10 3.39 3.26 –0.13 a,b
Family and household composition
Was pregnant or had a baby 3.89 3.82 –0.08 3.44 3.32 –0.13  
Had someone new (other than a new baby or partner) move into the household 4.20 4.57 +0.37 3.43 3.27 –0.16 a,b,c
Started living with a new partner/spouse 4.95 4.76 –0.19 3.45 3.33 –0.12 a,b
Had a separation due to relationship or marital difficulties 4.90 5.38 +0.48 3.40 3.25 –0.16 a,b,c
Broke off a steady romantic relationship 5.46 6.35 +0.90 3.40 3.24 –0.16 a,b,c
Financial and social matters
Had a major financial crisis 5.12 6.25 +1.13 3.28 3.01 –0.27 a,b,c
Lost job, but not from choice 3.75 4.00 +0.25 3.45 3.29 –0.16 a,b,c
Had something of value lost or stolen 4.44 4.87 +0.43 3.41 3.25 –0.16 a,b,c
Had someone in the household with an alcohol or drug problem 5.63 5.55 –0.08 3.40 3.28 –0.13 a,b
Residential matters
Moved house 4.01 4.00 –0.01 3.37 3.23 –0.15 a,b
Lived in a drought-affected area 3.52 3.52 0.00 3.47 3.34 –0.13  
Had home or local area affected by bushfire, flooding or severe storm 3.19 3.42 +0.23 3.50 3.35 –0.15 c

Note: Life events were reported in Wave 4. The sample is restricted to those who reported a distress measure at Wave 3 and Wave 4, and reported on the life events questions at Wave 4. a The Wave 3 score differs according to whether the life event occurred—significant at least at the p < .01 level. b The Wave 4 score differs according to whether the life event occurred—significant at least at the p < .01 level. c The change in the distress score differs according to whether the life event occurred—significant at least at the p < .01 level.

Source: LSAC Waves 3 and 4, B and K cohorts combined

The general pattern of results pertaining to Wave 4 levels of distress is the same as that outlined in the previous section. That is, those who experienced many of the life events had higher levels of distress at Wave 4, when compared to those who had not experienced those events. The exceptions were the same as those listed above.

Table 44 also shows the distress of mothers who experienced the following events increased to a significantly greater extent from Wave 3 to Wave 4 than was the case for mothers who did not experience the event:

  • suffered a serious illness, injury or assault;
  • had a serious illness, injury or assault happen to a close relative;
  • had someone new (other than a new baby or partner) move into the household;
  • had a separation due to relationship or marital difficulties;
  • broke off a steady romantic relationship;
  • had a major financial crisis;
  • lost job, but not from choice;
  • had something of value lost or stolen; and
  • had home or local area affected by bushfire, flooding or severe storm.

We also extended these analyses through multivariate analyses to examine changes in distress over three waves of LSAC, and including reports about the experience of life events at each survey wave. This was done through a method that specifically seeks to explain how changes in predictor variables (in these analyses, the life events indicators) are associated with changes in an outcome variable (in these analyses, the K6 measure of psychological distress). These analyses could only include life events that were asked about in each of Waves 2, 3 and 4, and so Table F3 shows that the list of life events is shorter than in previous tables (see table footnote for details). One advantage of this method is that it takes account of the reporting of life events at each wave, not just at Wave 4. More information about the method used is given in Appendix F.

In addition to the findings described above, this analysis suggested that levels of distress increased significantly if:

  • someone in the household had an alcohol or drug problem; or
  • a parent, partner or child died.

The fact that these findings were statistically significant over three, but not two, waves of analyses may indicate that some events have longer term effects on distress than others. Another possible explanation is that because some events are rare, it is more difficult to detect any association between those events and levels of distress, and incorporating data for more survey waves makes it more likely that significant differences will be detected.

To illustrate the above analyses of Wave 3 and Wave 4 associations between life events and psychological distress, those with relatively high Kessler scores (at least 10), were classified as being “highly distressed”. For each life event, Figure 5 shows the percentage of mothers who were thereby classified as highly distressed at Wave 3 (before the life event occurred, if it was experienced) and at Wave 4 (after the life event occurred, if it was experienced). At each wave, the sample is divided into those who did and did not report experiencing that life event in the 12 months prior to Wave 4. (Note, though, that these analyses do not take account of respondents’ experiences of life events in the 12 months prior to Wave 3.)

Figure 5: Longitudinal analyses of mothers’ level of distress, by life events and whether experienced or not in previous 12 months

Note: The percentage shows those with a K6 score of 10 or higher, here referred to as “distressed”. Only those mothers with a distress score at Wave 3 and Wave 4 are included.

Source: LSAC Waves 3 and 4, B and K cohorts combined

These analyses show, for example, that among those who, in Wave 4, reported that they had suffered a serious illness, injury or assault during the previous 12 months, 10% already appeared to be highly distressed (in Wave 3) and 15% indicated high distress after the event took place (Wave 4). Among those who had not experienced such events, only 7% indicated high distress in each survey wave (taken separately).

Among respondents who reported in Wave 4 that they or their partner had experienced the following events, the proportion of mothers with high distress at Wave 3 (scores of 10 or more) was significantly higher than it was for those respondents who, at Wave 4, indicated that they had not experienced these events:

  • suffered a serious illness, injury or assault (10% distressed at Wave 3 and 15% at Wave 4);
  • had someone new move into the household (12% distressed at Wave 3 and 15% at Wave 4);
  • started living with a new partner/spouse (14% distressed at Wave 3 and 13% at Wave 4);
  • had a separation due to relationship or marital difficulties (13% distressed at Wave 3 and 20% at Wave 4);
  • broke off a steady romantic relationship (16% at Wave 3 and 25% at Wave 4);
  • had a major financial crisis (16% distressed at Wave 3 and 23% at Wave 4);
  • had something of value lost or stolen (10% distressed at Wave 3 and 12% at Wave 4);
  • had someone in the household with an alcohol or drug problem (12% distressed at Wave 3 and 15% at Wave 4);
  • lost job, but not from choice (19% distressed at Wave 3 and at Wave 4); and
  • moved house (10% distressed at Wave 3 and 12% at Wave 4).

Note that for some of the life events, distress levels were already relatively high among mothers before experiencing the event (i.e., in Wave 3), compared to those who did not experience it. For example, of the respondents who said they had not experienced a financial crisis, 6% were classified as highly distressed at each of Wave 3 and Wave 4. Among those who said they had experienced a financial crisis, the proportions distressed were 16% at Wave 3 and 25% at Wave 4. That is, a higher proportion of mothers who subsequently experienced a financial crisis indicated high distress compared with mothers who did not go on to experience such a crisis, and there was some indication of mothers becoming more likely to be distressed following this time of financial difficulties.

Multiple life events and distress

To explore how the experience of multiple life events was associated with distress, the above illustrative analysis was repeated, with comparisons made according to the number of life events reported in Wave 4, instead of the specific life events.

Figure 6 shows, not surprisingly, that the proportions of mothers classified as highly distressed at Wave 4 increased with increases in the number of life events that they (or their partner, where relevant) experienced. Given the findings presented above for specific life events, it is also not surprising to see that even at Wave 3 the percentage of mothers classified as highly distressed increased with the number of life events reported at Wave 4. For those who reported experiencing 3 life events or 4 or more life events at Wave 4, the percentage of mothers classified as highly distressed increased between Wave 3 and Wave 4. This was particularly marked for those who reported experiencing 4 or more life events.

Figure 6: Longitudinal analyses of mothers’ distress, by number of life events in previous 12 months

 Figure 6: Longitudinal analyses of mothers’ distress, by number of life events in previous 12 months - as described in text

Note: The percentage shows those with a K6 score of 10 or higher, here referred to as “distressed”. The grouping of life events is based on life events reported at Wave 4. Only those mothers with a distress score at Wave 3 and Wave 4 are included.

Source: LSAC Waves 3 and 4, B and K cohorts combined

Limitations of analyses

As noted earlier, in relation to the HILDA results, one limitation of these analyses is that we have not taken account of the fact that some of the events examined may have been chronic, or may represent stages in a long-term process. That is, the reports of experiencing life events in the 12 months prior to Wave 4 may follow a pattern of similar experiences during preceding years, are represent an outcome of such experiences (e.g., relationship separation following increasing disenchantment with the relationship). It is possible that findings regarding distress being higher even before the life event occurred (as reported in Wave 4) resulted from such longer term processes. Further, the measure of distress captures relatively recent indicators of distress (over the previous 6 weeks), and we do not have information about when specific events occurred. Presumably, those that occurred close to the time of the survey would have had a greater effect on wellbeing than those that may have occurred closer to one year before. Also, these analyses only use one measure of wellbeing (a negative measure). Life events would affect families in ways other than that examined in this set of analyses.

This concludes the analyses of the LSAC data. Section 4 draws together the key findings from this section, and the analyses of the HILDA data from Section 3.

Footnotes

13 LSAC also collects information directly from children, and from others, such as teachers. This information is not used in this report.

14 In newly formed couple families, it remains possible that such events occurred to one partner only.

15 Note that these percentages were calculated using the Wave 3 sample, restricted to those who answered the life events questions at Wave 4, and weighted using the Wave 3 sample weights.

16 Of course, the extent to which repercussions of an event can be influenced by the individuals who are directly affected would vary according to the nature of the event and relevant personal circumstances (including social networks and access to effective services).