Thanks to the richness of the data they collect, longitudinal studies provide valuable information on several complex aspects of life. There are many birth cohorts worldwide offering powerful insight to researchers and policy makers. However, a study is only as powerful as its design. As discussed previously on the ScotCen blog, fathers have often been overlooked in research about children and families - including cohort studies - often because of the challenges and costs of gathering data from multiple sources. This is problematic because gathering these data provides an important perspective on significant aspects of children’s day-to-day lives and relationships. A common compromise is to collect data on the father-child relationship by proxy, from mothers or older children, but this introduces potential issues with accuracy, bias and missing data.
Such challenges apply to fathers who live with the mother and cohort child full time. Fathers who live in a different household - whom we call ‘own-household-fathers’ - present additional challenges. Some of these challenges are obvious for example, how might these fathers be identified? And how should they be contacted? But there are also more detailed, sensitive and complex challenges such as how different types of fathers may be defined, how the issue of requesting the father’s contact details is handled sensitively, and in a way that minimises any risk of the mother withdrawing from the study and how obtaining a representative sample is better assured? These are just a few of the questions the ‘Recruitment and retention of birth fathers in split-off households’ project - a collaborative project between ScotCen and the Fatherhood Institute - aimed to explore.
Step 1: Typology
To start answering these questions, we first wanted to illustrate the many different types of own-household-fathers. After extensive desk research and conceptualisation exercises - including analysis of data on patterns of contact between children and own household fathers in the Growing Up in Scotland (GUS) study - ten different categories of own-household-father were identified.
These were then grouped into two main categories: ‘part-time resident fathers’ included sub-categories of fathers such as ‘equal (or near equal) overnight care’ father and ‘part-time away’ fathers; ‘no regular overnight stays fathers (i.e. non-resident)’ included sub-categories of fathers such as frequent ‘virtual contact’ fathers and ‘no contact fathers (who are still alive)’.
Residence refers to the co-residence of the father and child. There is a common misconception that own-household-fathers are all of the ‘non-resident’ type, when in fact these fathers can be differentiated by the many different contact patterns and living arrangements they have with their children. This ranges from fathers who provide regular and considerable overnight or daytime care to those who have no contact at all.
Today, fathers who live far away from their child or children, who previously would only have had intermittent contact such as during school holidays, can maintain daily virtual contact via messages, social media and video calls. These variations mean different fathers may have different types and levels of impact on their children’s lives. They also mean that research seeking to involve and understand own-household-fathers may need to engage different fathers in different ways, which requires a valid way of identifying the different categories.
Step 2: Question design and a cognitive interviewing pilot
Having conceptualised all the different types of own-household-fathers, and scoped the identifying questions used on recent UK and international studies, we set out to formulate a complete set of questions for interviews with mothers. These questions were designed to be used in longitudinal studies across the globe to adequately differentiate sub-categories defined by patterns of contact and living arrangements. A small-scale cognitive interviewing pilot was implemented to test and refine the set of questions proposed. The pilot sample was recruited from amongst participants in Birth Cohort 2 of GUS and was comprised of birth mothers who were separated from the cohort child’s birth father.
The complete set of questions can be found in our report. They are designed to cover a wide range of information about the father-child relationship, from the type of contact - face-to-face, overnight stay, virtual - to the geographical distance between households, which can account for less frequent face-to-face contact.
Step 3: Applying the findings
A key insight from this project is how easy it is to overlook fathers - especially own-household-fathers - in the original conception and design of a study. Our review of methodological evidence on recruiting and retaining own household fathers in longitudinal studies, along with our experience of delivering GUS, demonstrates that if a specific sample population is not built into the original design, it can be extremely difficult to recruit and retain them later on. As such, it is key for studies involving children and families to consider carefully the need to recruit all biological fathers from the outset, collect their contact details and create a direct and independent channel of communication with them. A range of suggested strategies to enable this, based on an extensive scope of methodological literature, can be found in our report. This information is already being utilised in the design and development of the new ESRC-funded Early Life Cohort Feasibility Study, the precursor to the next UK birth cohort.
Konstantina is a Senior Researcher in NatCen’s Longitudinal Surveys team, She has been working predominantly on ‘Growing Up in Scotland’ and the ‘English Longitudinal Study of Ageing (ELSA)’. Her research interests include life trajectories as well as the criminal justice system.
As part of the National Centre for Social Research (NatCen), ScotCen have been designing and delivering applied social research projects in Scotland, utilising both qualitative and quantitative methods, since 1998. ScotCen are not-for-profit and focused entirely on delivering high quality social research that can improve people’s lives. Clients span central government, the third sector, universities and local government. Since its launch in 2004, ScotCen has led the Growing Up in Scotland study (GUS) - a large-scale, longitudinal study, commissioned by the Scottish Government, following multiple cohorts of children living in Scotland. Through involvement in GUS, ScotCen have developed interest and expertise in the design and delivery of longitudinal studies and continually seek to identify, test and promote methodological innovations and improvements that will enhance the utility and quality of these important social science endeavours.
This project was funded under the Economic and Social Research Council’s (ESRC) 'UK Population Lab' programme which commissioned a range of research to help to inform thinking about the feasibility and design of innovations and methods in data collection in new and existing longitudinal studies. Amongst the ESRC’s areas of interest, the project focuses on innovative means to recruit and retain ‘hard to reach groups’ and those at greatest risk of dropping out. The full list of outputs from the programme is available from this page on the ESRC website.