Fundamentals of
longitudinal research
Helen Baldwin
SPSW, University of York
Explores social phenomena through time
Collects data from samples at more than one point in time – “waves”
Examines temporal processes, e.g. social mobility
May be prospective or retrospective: “looking forward” vs. “looking back”
Has “time frames” and “tempos”
Can provide a picture of society across the generations: infancy, childhood,
adolescence, adulthood and older age
Has a wide scope
Utilised in a range of disciplines including the social sciences, public
health, medicine and developmental psychology
Covers a range of health and social issues: lifestyle behaviours, health,
family, socioeconomic factors
1983: Cebu Longitudinal Health and Nutrition Survey – Philippines
1989: China Health and Nutrition Survey (CHNS)
1990: Birth to Twenty (BT20) – South Africa
1991: Avon Longitudinal Study of Parents and Children (ALSPAC)
1991: Understanding Society (formally British Household Panel Survey)
2000: Millennium Cohort Study (MCS) – UK
2002: English Longitudinal Study of Ageing (ELSA)
2011: Chinese Longitudinal Aging Study
Types of longitudinal research design
Design Key features
Total population design Total population is surveyed/measured at each time
point, e.g. census. There is some change in the sample
between time points due to births and deaths.
Longitudinal panel design The same group of participants (a “panel” or “cohort”) is
followed over time. Some participants are likely to drop
out with each wave of data collection.
Revolving panel design A panel is followed over time. Participants who drop out
at each wave are replaced with new participants.
Repeated cross-sectional A different sample is drawn from the population at each
design wave of data collection. The same measures are repeated
at each wave. Does not examine individual change.
(Menard, 2007)
Strengths and limitations
Strengths
Provides insights into causality:
• helps us to understand the factors/exposures that influence future outcomes
• allows us to identify potential intervention points
Generates data that may be archived and re-used by other researchers
Limitations
Expensive
Time consuming
Panel maintenance and attrition
Ethical issues – changing ethical requirements, consent to share data, avoiding
harm
Quantitative vs. Qualitative
Quantitative longitudinal enquiry links time to trends: generates the
long shot, birds eye view, the broad vista, the epic movie. Time is linear.
Qualitative longitudinal enquiry links time to textures: generates close
ups of individuals and groups, the twists and turns in the story lines, the
intricacies of human lives, the personal movie. Time is fluid.
(Neale, 2018)