What is macro data?
The terms micro and macro data are often used to denote data used in social science research. The distinction between them is, however, not always obvious.
Micro data can generally be described as individual level data. These data have often been collected from each individual through a survey or interview. In such a dataset, each row typically represents an individual person and each column an attribute such as age, gender or job-type. Some well-known surveys that collect this type of data include the European Social Survey (ESS), the General Social Survey (GSS), the World Values Survey (WVS) etc. 'Micro data' would also denote data on individuals collected from governmental administrative systems and registers.
While the main distinction is most often drawn between micro and macro data, the term 'meso data' is also sometimes used. Meso data generally refers to data on collective and cooperative actors such as commercial companies, organizations or political parties.
'Macro data' is generally a term used to describe mainly two subtypes of data;
- aggregated data
- system-level data
Aggregated macro data provide information constructed by combining information on the lower level units, which the higher level unit is composed of (Diez-Roux 2002). Examples of aggregate data include summaries of the properties of individuals, unemployment statistics, demographics, GDP etc. Most often, aggregated macro data imply that the variables are summaries of the properties of lower level units and not measures of inherent higher level properties.
System level macro data yield information about properties of the state or the political system and cannot be disaggregated to lower level units. This type of data form political indicators, such as institutional variables and regime indices, and is not based on summaries of the properties of lower-level units, but measures characteristics of the higher-level units themselves.
The MacroDataGuide provides links and qualitative information on a wide range of both aggragated and system-level macro data sources.
Diez-Roux, Ana V. 2002. “A glossary for multilevel analysis”. Journal of Epidemiology and Community Health 56 (August): 588-594.