How well do adult cohort members remember their childhoods?

26 July 2012

Asking 50-year-olds to recall certain aspects of their childhood can produce reasonably reliable data, a new working paper shows.

Retrospective questions are a common method of gathering information on earlier life circumstances and events where that data is not available. However, asking people to recall aspects of their past can raise issues of reliability. A new working paper from the Centre for Longitudinal Studies (CLS) examines some of the methodological challenges around retrospective questions, using evidence from the 1958 National Child Development Study (NCDS).

When NCDS cohort members were 11 years old, their parents were asked how many people lived in their household and how many rooms there were in their homes. Thirty-nine years later, the cohort members – then age 50 – were asked to recall the same information from when they were 11.

There was a greater degree of consistency between the age 50 and age 11 answers when it came to the number of people in the household. Sixty-eight per cent provided a response that matched the response provided by their parent exactly, compared to 50 per cent for number of rooms. Where cohort members did provide responses that were inconsistent with their parents’, most were only off by one or two.

Cohort members with less stable family backgrounds, those living in larger households, those with less education and men were more likely to provide inconsistent responses. The author advised researchers to use caution when interpreting responses from these groups to retrospective questions about their childhood homes.

Read the full CLS working paper, Assessing recall of early life circumstances: Evidence from the National Child Development Study, by Matt Brown, Survey Manager for NCDS and the 1970 British Cohort Study.