Diet and nutrition: key data sources
Living Costs and Food Survey (formerly the Expenditure and Food Survey)
Strengths: able to measure most of the Scottish dietary targets.
Weaknesses: relatively small sample size (but can be accumulated over a number of years); questions relate to households rather than individuals in households, and exclude children.
Strengths: large sample size improves validity and allows sub-analyses by population group/geographical area; wider scope allows analyses of combinations of risk factors, and relationships with health outcomes.
Weaknesses: eating habits module does not look at all food types, is unable to measure nutrients and has limited data on portion size.
Strengths: large longitudinal and cross-sectional survey which has a unique Scottish focus and is driven primarily by the needs of policy; it covers eating habits, main meals, types of food eaten and sources of advice on children's diets/healthy eating.
Weaknesses: limited questions on diet.
Health Behaviour in School-aged Children (HBSC)
Strengths: allows limited analysis of diet in relation to other health behaviours; international comparisons.
Weaknesses: limited questions on eating behaviour.
Scottish Schools Adolescent Lifestyle and Substance Use Survey (SALSUS)
Strengths: allows limited analysis of diet in relation to other lifestyle factors.
Weaknesses: only a few food questions on eating behaviour.
Scottish Social Attitudes Survey (obesity module)
The Scottish Social Attitudes Survey is a repeated cross-sectional survey that aims to facilitate the study of public opinion and inform the development of public policy in Scotland. The obesity module, carried out in the 2016 survey, included 40 questions on perception of overweight and obesity, recognition of the health impacts and how to maintain a healthy weight.
National Diet and Nutrition Survey (NDNS)
Strengths: allows UK comparisons.
Weaknesses: very small sample size in Scotland (although the current rolling NDNS programme includes a boost to the Scottish sample) and low response rates. Until now survey findings have related to different age groups in different survey years, therefore unable to look at trends over time.
