Parent-child contact and post-separation parenting arrangements

Research Report No. 9 – July 2004

You are in an archived section of the AIFS website. Archived publications may be of interest for historical reasons. Because of their age, they do not reflect current research data or AIFS' current research methodologies.

2. Research design

This report draws on qualitative data derived from a series of focus groups conducted by the Australian Institute of Family Studies Parent-Child Contact Study. These data were collected as part of a larger study of parent-child contact after separation - the Institute's Caring for Children After Parental Separation Project. The study was designed to obtain a different methodological viewpoint from survey findings into post-separation patterns of parenting, especially in relation to context, diversity, transitions, and process.

Ten focus groups were conducted (see Figure 2). Each group comprised four to seven respondents, and related to one of five patterns of post-separation care: (1) 50/50 shared care (for example, week-about); (2) little or no contact; (3) holiday-only contact; (4) daytime-only contact; and (5) 'standard' contact (every-weekend or every-other-weekend).

Figure 2. Focus groups and the respective group numbers

Figure 2. Focus groups and the respective group numbers, described in text"

All five groups (except the male holiday-only contact group) 10 comprised samesex respondents to minimise the risk of conflict or gender power differentials, and to ensure that members within each group were as similar as possible on certain key attributes. It is this similarity within groups that allows for differences across the groups to be explored.

Participants

Participants were recruited through a range of non-probability sampling techniques. Specifically, a story appeared in the Melbourne Herald-Sun newspaper about the Caring for Children after Parental Separation Project. The story mentioned the focus groups and called for volunteers. Snowball sampling - in which participants first contacted are asked to name acquaintances in similar circumstances, who are then approached, and asked for additional names and so on until the required number of participants is obtained - was also used.

Focus group participants, Institute staff, and the Institute's email alert service were used to invite separated parents who fell into one of the five contact categories to participate. Various organisations were also contacted by letter to recruit participants.

All potential participants were screened for eligibility using an intake protocol, and a quota of six participants for each group was filled where possible.

The final sample for the focus group component of the study comprised 54 separated parents (27 mothers; 27 fathers), each of whom had at least one child under 18 years at interview. Apart from one mother and father, parents were not matched pairs of ex-couples.

The age range of the youngest biological child of each respondent was between two and 17 years. Parents ranged in age from 26-58 years (with an average age of 42) and had been separated for around six years (ranging between three to ten years). Just over half (54 per cent) were single (that is, not living with a partner), 64 per cent of them women; 44 per cent of them men. Virtually all participants were Australian-born of English-speaking background. Forty-four per cent of participants reported getting along pretty well with their former partner; 27 per cent reported not getting along too well; almost one quarter of the sample avoided any contact with their former partner; and the remaining 5 per cent had contact but fought and did not get along.

Question guide

A structured group interview guide was used comprising around a dozen questions (see Appendix 1). Focus group interviews typically have a particular logic, where questions are guided by a funnel design (see Figure 3). Relatively broad, easy, nonthreatening questions are initially asked to promote group cohesion, rapport and trust. Introductory questions are then followed by transition questions, which help participants move their thinking towards the key issues, and finally key questions (targeted and more personal in nature) are asked. As far as practical, the same questions are asked of all of the groups so that points of contact and disparity can be explored both across and within the groups. By and large, all questions were asked of the group (rather than of individuals).

Figure 3. Focus group question structure

Figure 3. Focus group question structure, described in text.

Examples of the questions are:

  • Could you tell me a bit about how your particular arrangement works? [Introductory question]
  • Why this pattern of contact? [Transition question]
  • How is the overall pattern of contact working for everyone - for you, your former partner, and the children? [Key question]
  • What advice would you give to other separated parents in making arrangements for children after separation? [Closing question]

Logistics

All of the focus groups were conducted at the Australian Institute of Family Studies in Melbourne. The groups comprised four to seven members. Focus group sessions, held across a two-week period in February 2003, lasted about 90 minutes. They were audio taped with the participants' permission, and subsequently transcribed verbatim. Participants were paid a small amount of money to cover any expenses incurred in attending the sessions.

Each group had a moderator, and a moderator's assistant. The project team (two females, one male) filled these roles. Where possible, female groups were moderated by female team members while the male team member and a male colleague (not part of the team) moderated the male groups.

Caveats

Given the small-scale non-probability design, observations are framed as 'insights' rather than findings generalisable to the wider population. These insights, presented throughout this report as boxed insets, may help to generate hypotheses for subsequent larger scale study.

Five caveats should be noted. First, the richness and depth of respondents' individual stories can never be fully represented by any form of data display.

Second, the groups of separated and divorced men and women are very small, and they were not drawn randomly from the general population of separated or divorced parents. Thus no claim is made that the responses are representative of parents with similar arrangements or generalise to that population. Qualitative data are not used to draw inferences about a particular population at large. Rather, they provide in-depth information about context, diversity and process, and can help to generate hypotheses that may be tested in subsequent empirical work with representative samples.

Third, collecting information from both parents from the same relationship is extremely difficult (Smyth 2002). In the Parent-Child Contact Study, only two of the 54 focus group participants were from the same former union. Without access to both sides of each story, making sense of some of the family dynamics at play is a formidable task.

Fourth, some issues requiring great sensitivity (such as domestic violence or personal depression) were not explored or tackled if they were raised. Such issues could not be explored adequately and ethically within the confines of a group situation. Where sensitive or potentially serious disclosures were made, the wellbeing of individuals who made these was monitored closely in line with predeveloped ethical protocols, and information on support, counselling and referral services was made available to participants.

Finally, one potential disadvantage of focus groups is that they discourage individual members to speak against the group norm. This means that individual group members may be reluctant to disclose attitudes or behaviours that are inconsistent with 'group disclosures' (Iiydén and Bülow 2003). Some things may therefore remain hidden.

Nonetheless, few studies of post-separation parenting in Australia have been able to yield the depth and breadth afforded by mixed-model designs, in which qualitative and quantitative data are combined.

10. The holiday-only group of non-resident fathers included one non-resident mother ('Julie'). This group was extremely difficult to fill, and many of the issues with which this non-resident mother was wrestling were similar to those raised by fathers in the screening interview.