INTERNAL DATA SOURCES AND PRIMARY DATA COLLECTION
Although a lot of data can be obtained from external sources, it is likely that you will need to collect some of your own data internally.
This page:
- Outlines considerations for use of administrative data in analysis.
- Introduces survey design.
Administrative Data
TLA’AMIN NATION: Using administrative data to track outcomes
Real Case Study
When Tla’amin Nation was negotiating their first five-year Fiscal Financing Agreement (FFA), they gathered information about their fiscal needs. One of the needs they assessed was the number of citizens on social assistance. Their assessment showed that there were 50 employable citizens on social assistance in 2011.
The more data you can obtain through work that is already being done, the better. In a busy government office, this means less work for you and your colleagues.
Your government collects a lot of useful data just by performing day to day administrative functions, such as maintaining a database of citizens, administering funding, and running programs. You can harness this existing data for analysis.
Here are considerations to think about when using administrative data[1]:
- Privacy and confidentiality: Ensure that you can justify the value of using administrative data for statistical purposes. It is not acceptable to use personal data for a substantially different purpose than that for which it was collected without the consent of the individual. Generally, using data for statistical purposes related to the program or service delivery for which the data was collected is considered an acceptable use of data. The best practice is to inform the individual in plain language why you are collecting their information and obtain their consent for the use of the data for statistical purposes. When using any data, care needs to be taken to ensure personal information is protected. For example, your analysis plan should consider if, when combining data points together from multiple sources, personal information is revealed. See Legislation and the Duty of Privacy Protection and Data Security and Privacy for more information.
- Quality: The quality of the data will depend on the needs of the administrative program. Some data fields might not be reliable. For example, if a data field is not essential for the needs of the administrative program, it may often be left blank and less emphasis will be placed on checking for accuracy.
- Coverage: Administrative data will only cover the population relevant to the program, which may not be the entire citizen population.
- Stability over time: Administrative data is likely to change over time as the program changes. Be aware that any changes to the way your data is collected or defined will impact your tracking indicators over time.
[1] “Use of administrative data” https://www150.statcan.gc.ca/n1/pub/12-539-x/2009001/administrative-administratives-eng.htm
Existing Systems
Examples of indicators you might find in different administrative systems include:
| System | Indicators |
| Accounting software | Number of post-secondary students receiving assistance Number of citizens receiving grants |
| Citizenship registrar | Number of citizens Age distribution of citizens |
| Address database | Household size Inter-community mobility |
| Permit database | Level of economic activity |
Modifying Workflows
Collecting data may require modifying workflows. For example, if you want to collect participation rates for a particular program, the program leader must track this information as part of their workflow. See Identifying IT System Requirements for information on how to analyze workflows.
Primary Data Collection
Primary data is data that is used for the same purpose for which it was collected. In this case, it is data you collect yourself for the purposes of tracking indicators. If data cannot be obtained from external data sources or internal administrative data sources, you will need to collect it. A common way to collect population-level indicator data is to use surveys.
This document introduces the main steps and considerations in developing and performing a survey.
Note that there are other data collection methods available, such as focus groups and interviews; however, these are beyond the scope of this Toolkit.
The webinar below is a presentation by Hannes Edinger and Rebecca Wortzman (Big River Analytics) on acquiring and working with data, including primary data collection and working with administrative data. For more webinars covering content in this toolkit click here.
