Recently the Federal Government launched its National Families Strategy to promote healthy supportive family relationships via prevention and early intervention programs. The Minister for Family and Community Services, who has primary responsibility for the development of the Strategy, has indicated that the areas of marriage and relationship education across the life cycle are key priorities. As part of the Strategy, the Government is investigating ways to encourage couples to participate in premarriage education. A pilot project involving the provision of vouchers for premarriage education will be undertaken in Perth and Launceston.

These initiatives coincided with plans by the Australian Institute of Family Studies and the Australian Catholic University to conduct a round table discussion on premarriage education. The idea for the round table grew out of the recognition that, while generally supportive, empirical evidence of the efficacy of premarriage programs is thus far unconvincing. To ensure that resources are allocated appropriately, further verification that premarriage preparation does in fact help Australian couples develop stronger and more stable relationships is required.

The round table, held at the Institute on 9 September 1999, was attended by relationship professionals from across the country. Based on the proceedings, this Briefing gives an overview of:

• the purpose of premarriage education;
• current status of premarriage research;
• identifying gaps in the research;
• identifying research problems; and
• a model for future research.

Purpose of premarriage education

In general, premarriage programs seek to engage couples in the processes of reflection and skills training with the aim of promoting and supporting the development of strong and stable relationships. Australian programs have evolved from an instructive, often prescriptive, lecture format to 'learner centred and facilitative programs' (Harris et al. 1992:1). Educators have been eclectic in the design of programs, drawing on educational, psychological and behavioural frameworks (House of Representatives Standing Committee 1998).

There are various ways of preparing couples for marriage:

• Individual partners may complete an inventory (such as PREPARE or FOCCUS) and then come together with a trained administrator to discuss the results. Inventories are a facilitative tool designed to help couples find out more about themselves and their partner, and to work through the issues important to their relationship;
• Couples can participate in group programs comprising up to 20 couples, to engage in discussions and activities designed to raise awareness and impart skills considered important to marital quality.
• Some programs combine both inventory and group formats.
• Some programs take the form of formal or informal discussions with a religious celebrant.

The length and format of programs may vary considerably, from short, one-off sessions (perhaps three hours), to one-day weekend programs, short sessions (of, say, three hours) spread over a number of weeks, or weekend residential programs.

Premarriage programs are often perceived incorrectly by the general public as being steeped in religion, probably because couples typically come into a premarriage program via the church in which they plan to marry. However, providers may be church-based, church-affiliated, or independent. Some programs contain references to Christian marriage, but this is by no means a defining feature of all such programs. (See Harris et al. 1992: 18-20 for an analysis of the extent to which religious dimensions are incorporated into church-based and church-affiliated programs.)

Status of premarriage research

Practitioners have few doubts as to the effectiveness of premarriage education - a belief stemming from their own experiences with couples. And couples who have participated in premarriage programs are also known to regard them favourably, with most who complete programs regarding them as a 'valuable event' (Harris et al, 1992: 114).

While the anecdotal evidence of the benefits of premarriage education is compelling, the empirical research, although generally supportive, is less convincing. Gains have been demonstrated in communication, conflict management skills, and relationship satisfaction, but a number of studies have been unable to demonstrate measurable positive effects (see Bagarozzi and Rauen 1981; Sayers, Kohn and Heavey 1998; and Simons 1999 for reviews). Overall, however, the culmination of the research so far suggests that premarriage education programs are effective in their aims of raising awareness and building relationship skills (Sayers et al. 1998).

There are four dimensions on which premarriage research has been criticised. First, much of the research is fragmented, selective and small-scale. Second, with the notable exception of Markman's (1988) Prevention and Relationship Enhancement Program, follow-up periods tend to be short- often six months to two years. Third, the bulk of the published research is conducted in the United States of America and outcomes may not naturally translate to the Australian population. Fourth, the research is methodologically flawed, most notably in the general absence of control and comparison groups, assumptions about the homogeneity of clients and educators, and the use of unstandardised measures.

Identifying gaps in the research

Most studies of premarriage education have focused on validating one specific program or another. This has tended to leave unanswered a number of broad and fundamental questions:

• Do participants in premarriage education have measurably better quality relationships than non-participants? In which aspects of their relationship do they differ?
• How is effectiveness best defined and measured?
• Why are some couples not impressed with the programs they attend?
• Why is premarriage education ineffective for some couples?
• What is the best combination of program type and format for particular types of couples?
• What alternative forms of participation should be made available (for example, Distance Education, web-based material)? Should these be evaluated in the same way as other programs?
• How long do any positive effects last?
• Is premarriage the best time to intervene?
• How do the known risk factors and indicators influence the success or otherwise of a premarriage program?
• How do couples, as distinct from individuals, learn and how do they apply their skills and knowledge in their daily lives?
• Which couples would gain more from individual sessions with an educator than a group setting? How could they be identified?
• What socio-demographic and relationship factors mediate or moderate the effectiveness of programs?

Identifying research problems

Various problems that can undermine the efficacy of research into premarriage education need to be addressed. These include: practical and methodological problems; issues of comparability; and issues of measurement.

Practical and methodological problems

Participants' perceptions of the helpfulness of premarriage education, their preference for a particular type of program, and the manner in which they came to be enrolled, may influence their willingness to participate in program activities. Reluctant participants are less likely than willing ones to become fully engaged in the process, and so gain less from their attendance.

Group composition is often determined prior to the researcher's involvement and thus control groups are difficult to obtain. The recruiting of couples into a research program can be quite successful, but the random allocation of couples to the premarriage group is problematic: allocated couples must still accept the offer of premarriage education and those who do accept may have different relationship characteristics from those who do not.

Couples who seek premarriage preparation are by definition different from those who do not; therefore the quality of their subsequent marital relationship may have less to do with attending a program than the possible absence of recognised marital distress risk factors.

Simply participating in a research project via their involvement in a premarriage program may have a positive effect on the behaviour of some participants that may ultimately lead to an overestimation of the effects of the program.

The effects of participating in premarriage education are typically assessed over very short periods. A full understanding of how programs contribute to the stability and duration of marriages requires repeatedly returning to participants and non-participants beyond the 'honeymoon' and 'seven-year-itch' periods.

Longitudinal studies, while considered essential, are susceptible to participant attrition. Procedures need to be put in place to minimise the loss of participants over extended periods.

Few studies of premarriage education programs have focused on the same outcomes or employed the same measures, resulting in a lack of cohesion in the literature.

Differences between couples, and between participants and non-participants, may be masked because researchers often rely on participant self-reports. (However, observational and behavioural data are also problematic.)

Issues of comparability

A comprehensive study of whether premarriage education is effective is likely to encounter difficulties in comparing like with like- for example in comparing programs, educators and participants.

Comparing programs

Although programs offered by different agencies or providers can appear similar, individual course content may vary both across programs and across groups.

The four most common content areas identified by Harris et al. (1992) were communication, sexuality, conflict resolution, and family of origin. Eleven other topics were also named, including finance, expectations, and change and growth.

Some programs have a degree of flexibility that allows the group to decide, outside the established core content areas, which of a set of optional topics are covered in a particular course. This facilitates learning via a sense of input into, or ownership of, the process, but complicates research design by further reducing the comparability of programs.

Comparing educators

Just as it should not be assumed that all programs are alike, it would be a mistake to ignore the variations in those who conduct programs. Educators may be individual celebrants, members of a team of educators working under the auspice of a church-sponsored or independent agency, or individuals who provide pre-marital counselling for individual couples.

It should also be recognised that the training received by educators can also vary. While competency standards are available to all, they may not be used by all service providers.

Comparing participants

Variations in personal and relationship characteristics provide possible alternative explanations for any observed differences between participants and non-participants.

For instance, newly-wed couples vary in the extent to which they are committed to, and satisfied with, their relationship. Individual partners differ in the extent to which they are affected by the range of recognised risk factors (age at marriage, experience of parental divorce, and so on), they enter a program with different levels of relationship skills, and with varying relationship experience.

Issues of measurement

The effectiveness of premarriage programs can be defined and measured in several ways:

• overall relationship satisfaction;
• intimacy and commitment;
• individual and couple wellbeing;
• knowledge of self, partner and resources available to the relationship;
• relationship skills and behaviours;
• personality variables;
• relationship expectations;
• fewer divorces;
• longer relationships.

Further concerns stem from the differences in the particular aims and objectives of individual programs. Applying the same measure of effectiveness to all programs would produce misleading and meaningless conclusions. Thus gathering different types of data is desirable.

However, data collected by observational and self-report methods do not always produce consistent findings, and concerns have been raised in previous research about whether behaviour produced in artificial settings is an accurate reflection of a couple's everyday interaction (Simons 1999).

Gathering large amounts of observational and behavioural data may also be impractical. However, reliance on quantitative data methods will mean the loss of valuable information about the subtleties of relationship dynamics that contribute to the perceptions and behaviours that may be recorded.

A model for future research

As can be seen from the above discussion, there are many aspects to be considered in establishing a program of research to evaluate premarriage education.

Research design

In a community and service-driven setting, the experimental approach to investigating group differences is impractical. Studies in which participants are randomly assigned to premarriage programs would be extremely difficult and costly to establish, as would constructing matched samples.

Alternative ways of establishing differences between participants and non participants must be found. Having groups of couples receive a less interactive form of premarriage preparation (such as a set of readings) offers a basis for comparison but, because any form of intervention is likely to produce some effect, it does not answer the essential question of whether premarriage participants and non-participants differ.

An alternative approach may be to obtain a large random sample of engaged and de facto couples prepared to participate in a longitudinal study of marriage and, as part of the set of independent variables, factor in their participation in a program, and the type of program attended. A correlational design such as this would also allow for pre-participation differences between those who attend (whether voluntarily or at the behest of their celebrant) and those who do not attend to be incorporated into the statistical analysis.

This design would eliminate the need to disrupt programs in order to recruit participants. It would provide additional data on the pathways and processes of marital relationships that would inform both researchers and practitioners. It would also provide a mechanism for further investigation of those couples who show no benefits or who are negatively affected by their participation.

The Australian Family Panel Survey, currently being designed by the Australian Institute of Family Studies, provides a vehicle for such a design, and will include the collection of data on premarriage education attendance. As a longitudinal study, the Panel Survey will enable the Institute to monitor across the life cycle the relationships of those who do and do not attend premarriage preparation.

However, neither the experimental nor the correlational design can clearly demonstrate how premarriage education programs work, that is, how couples apply what they learn in a program to their daily lives. This kind of information, beyond the resources of most service providers, will help program designers and educators improve the quality and effectiveness of programs.

Thus a two-pronged approach is required. The first, an investigation of the differences in relationship quality of those who have received premarriage education compared to those who have not. The second, a parallel examination, primarily qualitative, of the process of how couples transfer what they learn in the artificial setting of the program into the real world of their everyday lives.

Data

Given the dynamic nature of the educative process, purely quantitative techniques are unlikely to capture adequately the role of premarriage education in the development of the couple relationship. Qualitative techniques would need to be employed to probe the intricacies of the relationship dynamics of participant and non-participant couples. Individual or group interviews would provide access to that information.

Observations of couple interactions are highly desirable to complement self reported communication and conflict resolution patterns. However, the cost could be prohibitive in a large study, and behaviour reproduced in the laboratory may not be indicative of how the couple interact in their daily lives.

Studying the interactions of an initial sub-sample of couples, and comparing that data with their self-reported communication strategies, would clarify whether observation of the entire sample was warranted.

Program typology

To overcome the problems of making comparisons across different programs, there is a need for a detailed typology of programs. Harris et al. (1992) surveyed the field of premarriage education agencies/providers, educators and participants, and developed a model to distinguish programs on the basis of structural flexibility. This model could be developed further, to incorporate other relevant dimensions (such as educative style, content, religious emphasis, cost and duration).

Skills and competence of individual educators

Optimal control of the variation in educators might be achieved by restricting analysis to only those educators who have been certified as attaining the established competency standards. However, it would not be desirable to eliminate programs on this basis because of the reduction in useable data. Instead, variability in the training and competence of educators could be controlled adequately by statistical methods, in much the same way as data about the type of program a couple attended.

Couple characteristics

Needless to say, measurement of personality and couple characteristics should take place prior to participation in premarriage education. The relative risk of future marital distress could be calculated for each couple, and their pathway into the premarriage course, their relationship history, satisfaction and quality, be included as baseline measures.

Measuring effectiveness

Simply evaluating whether the specific goals of a particular program are achieved does not help to determine whether there are any long-term benefits, or what those benefits might be. The simplest, but not particularly informative, measure may just involve comparing the divorce rate of those who participate with that of the general population.

However, greater depth would be obtained by measuring a range of relationship characteristics. Multiple standardised assessment tools administered both pre- and post-marriage, and then at suitable intervals (yearly or twice yearly) would give a comprehensive evaluation of whether differences existed between participants and non-participants. This would also point to how a range of relationship dimensions differ over time, and identify when any improvement or deterioration begins to occur

Conclusion

The round table on premarriage education identified a number of questions that are yet to be answered by research, and it highlighted several problems awaiting researchers who attempt to identify differences between premarriage education participants and non-participants. Although much of the premarriage curriculum is founded on a range of theoretical perspectives and educational models, the empirical investigation of its effectiveness has largely been short-term, piecemeal and, not surprisingly therefore, inconclusive.

The methodological flaws in the prior research, the unanswered questions, and the inherent practical problems suggest the need for a large, integrated, and longitudinal design.

This Briefing has offered potential solutions to some of the problems identified, and suggested a correlational design as an alternative to the more scientifically rigorous but difficult to implement experimental design. The design would accommodate many of the problems associated with premarriage education research. It would also generate a comprehensive view of the long term trajectory of the quality, stability and duration of marital relationships.

A further study to determine how those who have undertaken premarriage education actually apply that knowledge is necessary for the ongoing improvement of programs and their ability to make a difference in the lives of those who participate.

This is a necessarily brief outline of the way forward for premarriage education research. The Australian Institute of Family Studies' MARE (Marriage and Relationship Education) email discussion list has been established to facilitate dialogue between premarriage education practitioners and researchers. It offers a forum for debate about this paper, and other issues relevant to the development of marriage and relationship programs.

References

• Bagarozzi, D.A. & Rauen, P. (1981), 'Premarital counselling: appraisal and status', The American Journal of Family Therapy, vol. 9, pp. 13-30.
• Harris, R., Simons, M., Willis, P. & Barrie, A. (1992), Love, Sex and Waterskiing: The Experience of Pre-marriage Education in Australia, Centre for Human Resource Studies, University of South Australia, Adelaide.
• Markman, H.J., Floyd, F.J., Stanley, S.M. & Storaasli, R.D. (1988), 'Prevention of marital distress: a longitudinal investigation', Journal of Consulting and Clinical Psychology, vol. 56, pp. 210-217.
• Sayers, S.L., Kohn, C.S. & Heavey, C. (1998), 'Prevention of marital dysfunction: behavioural approaches and beyond', Clinical Psychology Review, vol. 18, no. 6, pp. 713-744.
• Simons, J (1999), 'Can marriage preparation courses influence the quality and stability of marriage?', J. Simons (ed.) High Divorce Rates: The State of the Evidence on Reasons and Remedies. Volume 2, Lord Chancellor's Department, London.
• House of Representatives Standing Committee on Legal and Constitutional Affairs (1998),To Have and To Hold: Strategies to Strengthen Marriage and Relationships, AGPS, Canberra