Difference between revisions of "Secondary Data Sources"
| (3 intermediate revisions by the same user not shown) | |||
| Line 3: | Line 3: | ||
== Read First == | == Read First == | ||
* [[Impact Evaluation Team|Research teams]] usually rely on two broad categories of data - [[Primary Data Collection|primary data]], and '''secondary data'''. | * [[Impact Evaluation Team|Research teams]] usually rely on two broad categories of data - [[Primary Data Collection|primary data]], and '''secondary data'''. | ||
* '''Impact evaluations''' rely on many different sources of '''secondary data''', such as: [[Administrative|administrative]], [[Geo Spatial Data|geospatial]], [[Remote Sensing|sensor]], [[Telecom Data|telecom]], and [[Crowd-sourced Data|crowd-sourcing]]. | * '''Impact evaluations''' rely on many different sources of '''secondary data''', such as: [[Administrative and Monitoring Data|administrative]], [[Geo Spatial Data|geospatial]], [[Remote Sensing|sensor]], [[Telecom Data|telecom]], and [[Crowd-sourced Data|crowd-sourcing]]. | ||
* '''Research teams''' should decide on the kind of data they want to use, based on context and project needs. | * '''Research teams''' should decide on the kind of data they want to use, based on context and project needs. | ||
== Types of Secondary Data == | == Types of Secondary Data == | ||
=== Administrative and Monitoring Data === | === Administrative and Monitoring Data === | ||
[[Administrative Data|Administrative data]] includes all data collected through existing government ministries, programs and projects. It is a potentially rich source of data for an impact evaluation. Some of the key challenges with administrative data include: | [[Administrative and Monitoring Data|Administrative data]] includes all data collected through existing government ministries, programs and projects. It is a potentially rich source of data for an impact evaluation. Some of the key challenges with administrative data include: | ||
* '''Digitization | * '''Digitization''': in a lot of cases, the data is in paper format only. | ||
* '''Restricted access | * '''Restricted access''': it is also difficult to get access to certain data because it contains sensitive information. | ||
* '''Lack of unique ID | * '''Lack of unique ID''': in some cases, administrative datasets might be missing a numeric [[ID Variable Properties|ID variable]]. | ||
=== National Survey Data === | === National Survey Data === | ||
Existing survey data may be of use depending on the sampling frame for the impact evaluation, level of representativity of the existing data, and availability of disaggregated data. National Statistics Office typically collect a wide array of nationally-representative data, such as Living Standards Measurement Surveys and censuses. International survey efforts such as the Demographic and Health Surveys [https://dhsprogram.com/] and Enterprise Surveys [http://www.enterprisesurveys.org] are also good sources. | Existing [[Survey Pilot|survey]] data may be of use depending on the [[Sampling#Establish the Sampling Frame and Master Dataset|sampling frame]] for the '''impact evaluation''', level of representativity of the existing data, and availability of disaggregated data. National Statistics Office typically collect a wide array of nationally-representative data, such as Living Standards Measurement Surveys and censuses. International '''survey''' efforts such as the Demographic and Health Surveys [https://dhsprogram.com/] and Enterprise Surveys [http://www.enterprisesurveys.org] are also good sources. | ||
=== [[Geo Spatial Data]] === | === [[Geo Spatial Data]] === | ||
This includes data from traditional satellites, micro- and nano-satellites | This includes data from traditional satellites, micro- and nano-satellites and unaccompanied aerial vehicles (UAVs, e.g. drones). | ||
=== [[Remote Sensing]] === | === [[Remote Sensing]] === | ||
| Line 23: | Line 23: | ||
=== [[Telecom Data]] === | === [[Telecom Data]] === | ||
This includes call detail records, social media data, web scraping. | This includes [[Innovative Data Sources#Mobile Big Data|call detail records]], [[Innovative Data sources#Types of Secondary Data|social media data]], and web scraping. | ||
=== [[Crowd-sourced Data]] === | === [[Crowd-sourced Data]] === | ||
Latest revision as of 15:42, 16 August 2023
Secondary data is data collected by any party other than the researcher, including administrative data from programs, geodata from specialized sources, and census or other population data from governments. Secondary data provides important context for any investigation, and in some cases (such as administrative program data), it is the only source which covers the full population needed to conduct a research project.
Read First
- Research teams usually rely on two broad categories of data - primary data, and secondary data.
- Impact evaluations rely on many different sources of secondary data, such as: administrative, geospatial, sensor, telecom, and crowd-sourcing.
- Research teams should decide on the kind of data they want to use, based on context and project needs.
Types of Secondary Data
Administrative and Monitoring Data
Administrative data includes all data collected through existing government ministries, programs and projects. It is a potentially rich source of data for an impact evaluation. Some of the key challenges with administrative data include:
- Digitization: in a lot of cases, the data is in paper format only.
- Restricted access: it is also difficult to get access to certain data because it contains sensitive information.
- Lack of unique ID: in some cases, administrative datasets might be missing a numeric ID variable.
National Survey Data
Existing survey data may be of use depending on the sampling frame for the impact evaluation, level of representativity of the existing data, and availability of disaggregated data. National Statistics Office typically collect a wide array of nationally-representative data, such as Living Standards Measurement Surveys and censuses. International survey efforts such as the Demographic and Health Surveys [1] and Enterprise Surveys [2] are also good sources.
Geo Spatial Data
This includes data from traditional satellites, micro- and nano-satellites and unaccompanied aerial vehicles (UAVs, e.g. drones).
Remote Sensing
This includes all data collected by sensors, and through the Internet of Things (IoT).
Telecom Data
This includes call detail records, social media data, and web scraping.
Crowd-sourced Data
This includes all data collected by crowd-sourcing, often through social media or mobile apps.
Related Pages
Click here for pages that link to this topic.
Additional Resources
- JPAL, Handbook on Using Administrative Data
- DIME Analytics (World Bank), Secondary Data Sources