Survey methodology

We mainly use face-to-face interviewing to collect data from cohort members. We do this to help maximise response rates, and because some of our studies involve physical measurements, biological samples and direct assessments of cognitive function. We have also used telephone interviewing, and the web, to collect data, sometimes in combination with each other and with face-to-face interviewing.

Web data collection is a particular focus for us. We have used this mode in major sweeps of the National Child Development Study (NCDS), Next Steps and the Millennium Cohort Study (MCS), and we are considering whether to make greater use of the web in our studies, including between major sweeps.

Our research interests are around the implications of mode choice in longitudinal surveys, In particular, we focus on how to make the most effective use of the web, whilst ensuring both data quality and high-participation rates and understanding the impact of mode on measurement.

We have used new technologies for data collection in the MCS age 14 sweep for the time-use diary. This was mixed-mode using web, Smartphone app and paper.

For the MCS age 7 and age 14 sweeps, and the age 46 sweep of the 1970 British Cohort Study (BCS70), we used a portable activity monitor (or accelerometer) to collect direct measurements of physical activity.

We’re exploring opportunities for embedding new technologies and innovations within planned upcoming sweeps, and enhancing our studies with data collected, actively or passively, between sweeps using remote modes.

While it is certainly the case that there are many opportunities in this area, they are not without their challenges, such as cost, measurement issues and take-up. Our research interests are around how to make optimal use of these new opportunities to provide data for research, while understanding their limitations and overcoming their challenges.

There are a number of areas of questionnaire design which are of specific relevance to longitudinal surveys, including event histories, retrospective data collection, and dependent interviewing.

All of our study sweeps make extensive use of dependent interviewing, and recent sweeps of BCS70 (age 42), NCDS (age 55) and Next Steps (age 25) have used visual event-history calendars embedded into the questionnaire instrument. The upcoming sweep of NCDS (age 62) will include a retrospective Life History Questionnaire.

Our research interests are around how to optimally design questionnaires in longitudinal surveys, and the impact of these design choices on measurement.

All of our studies have collected a wide range of consents to link to administrative records.

Our research interests are around how to optimally collect record linkage consents, and which factors determine whether cohort members agree to these permissions.

Response maximisation is crucial for delivering accurate and robust data, and making sure our studies remain representative over time. We use a range of best practice approaches to response maximisation, and make use of para-data for case prioritisation during fieldwork.

Our research interests are around how to optimally manage the trade-offs between survey costs, response rates and sample representativeness, and how to improve survey design to maximise response rates.

Keeping track of cohort members who move, and keeping them in engaged over time, is a crucial aim for all of our studies. We use a range of best practice approaches, including using administrative data and social media for participant tracking and engagement.

Our research interests are around how to optimally track and engage participants in our studies.

Collecting objective measurements of health, often referred to as ‘biomeasures’, is an important feature of our studies. This data enables researchers to generate new insights into our understanding of the interplay between social and biological factors in explaining human behaviour.

There are different models for combining social and biomedical data in large-scale surveys, including clinic and home visit based approaches. Our studies have used interviewers to collect biomeasures on MCS, and nurse interviewers on BCS70 and NCDS.

Our research interests are around how to optimally include biomedical data collection in longitudinal surveys, and the impact of these different approaches on cost, response rates and measurement.

Our work on the MCS has pushed forward best practice in surveying children and young people. This covers questionnaire design, engagement and ethical approaches.

Our research interests are around how to optimally design surveys involving children and young people