Difference between revisions of "Administrative and Monitoring Data"

Jump to: navigation, search
Line 9: Line 9:
 
=== Administrative Data ===
 
=== Administrative Data ===
  
Administrative data refers to any data that national or local government(s) (ministries, agencies etc.) may already be collecting outside of the impact evaluation. A few examples can be national census data, tax data, school enrollment data etc. A specificity is that this information is not initially collected for research purposes but rather for documentation and tracking of policy beneficiaries, firm owners and general population. The objective is not to replace surveys with administrative data but rather to use various source to complement what is already collected for the research project.  
+
Administrative data refers to any data that national or local governments (i.e. ministries, agencies etc.) collect outside of the impact evaluation. Examples include national census data, tax data, and school enrollment data. Administrative data is generally not initially collected for research purposes but rather to document or track policy beneficiaries, firm owners and the general population. Researchers should aim not to use administrative data in place of survey data but rather in addition to it.
  
Administrative data provides the below advantages:
+
Administrative data offers advantages in quality, cost, and time. It is often considered more accurate than self-reported survey data; consider, for example, that a firm is more likely to accurately report its turnover rate to Financial Administrations than to a research team conducting a firm survey. Furthermore, notwithstanding potential access costs, administrative data doesn't pose additional costs as it is collected independent of the impact evaluation. Finally, administrative data can avail information frequently because it is often collected on a regular basis. This makes administrative data especially advantageous and attractive for research teams retrospectively evaluating interventions for which data collection did not occur.
  
-Quality: this can often be considered more accurate than self-reported survey data (eg. firm turnover is more likely to be correctly reported to Financial Administrations than to a firm survey conducted by a research team)
+
Nonetheless, administrative data doesn't come without a few potential challenges: access, merging, and quality. Accessing administrative data requires strong relationships with national and/or local authorities. In some cases, authorities may not be inclined to share the information. Once accessed, consolidating administrative data with other data often entails merging different databases together: this can be an extensive task when no common unique identifiers exist across the databases. Finally, while in some cases administrative data can provide high accuracy, in others, it may be badly reported, not exhaustive, or not at all existent. Not all governments have the same capacity to collect this information.
-Cost: as this information is collected regardless of the impact evaluation, it usually implies no cost on the research project (notwithstanding potential access costs)
 
-Time: Administrative data is often collected on a regular basis, implying that in some cases information can be available at a rather high frequency level. This is also an advantage when evaluating retrospective interventions, when data collection surveys may not have taken place.
 
 
 
Potential difficulties to working with administrative data:
 
 
 
-Access: this requires good relations with national and/or local authorities. In some cases these may not be inclined to share the information. Developing good working relationships with stakeholders can help address this problem.
 
-Merging: consolidating administrative data often requires to merge different databases together, this can be an extensive task when no common unique identifiers exist across the databases.
 
-Presence and quality: not all government's have the same capacity to collect this information. As such, it is possible that the data is badly reported, not exhaustive or does not exist at all.
 
  
 
=== Monitoring Data ===
 
=== Monitoring Data ===

Revision as of 16:57, 27 March 2019


Read First

Impact Evaluations can involve multiple sources of data. Although we are more commonly familiar with survey generated data, a lot can also be done with secondary data that is, for example, already collected by the government or program team.

Guidelines

Administrative Data

Administrative data refers to any data that national or local governments (i.e. ministries, agencies etc.) collect outside of the impact evaluation. Examples include national census data, tax data, and school enrollment data. Administrative data is generally not initially collected for research purposes but rather to document or track policy beneficiaries, firm owners and the general population. Researchers should aim not to use administrative data in place of survey data but rather in addition to it.

Administrative data offers advantages in quality, cost, and time. It is often considered more accurate than self-reported survey data; consider, for example, that a firm is more likely to accurately report its turnover rate to Financial Administrations than to a research team conducting a firm survey. Furthermore, notwithstanding potential access costs, administrative data doesn't pose additional costs as it is collected independent of the impact evaluation. Finally, administrative data can avail information frequently because it is often collected on a regular basis. This makes administrative data especially advantageous and attractive for research teams retrospectively evaluating interventions for which data collection did not occur.

Nonetheless, administrative data doesn't come without a few potential challenges: access, merging, and quality. Accessing administrative data requires strong relationships with national and/or local authorities. In some cases, authorities may not be inclined to share the information. Once accessed, consolidating administrative data with other data often entails merging different databases together: this can be an extensive task when no common unique identifiers exist across the databases. Finally, while in some cases administrative data can provide high accuracy, in others, it may be badly reported, not exhaustive, or not at all existent. Not all governments have the same capacity to collect this information.

Monitoring Data

Back to Parent

This article is part of the topic Secondary Data Sources


Additional Resources

Please include here links to relevant existing resources outside of the wiki