Page 2 of 6
3.1 DATA SOURCES
3.1.1 PRIMARY DATA SOURCES
Primary source data provide direct evidence about an event. Data collection can take different forms, whether through a population census, national or local research (typically, sample-based or non-sample-based surveys). The creation of an information system to achieve specific objectives will generate primary data. In other words, indicators are said to be based on primary data if the data source was created to achieve a specific purpose.
The objective of most countries' disease reporting systems for tuberculosis is to support surveillance and control of this disease. The indicators generated from that information system are examples of indicators produced using primary data. Similarly, a survey of schoolchildren designed to estimate prevalence rates of behavioral risks in a sample population is considered as a primary source when it is used for this purpose.
3.1.2 SECONDARY DATA SOURCES
Secondary source data are data that was originally collected for other purposes. The data from these existing sources are considered secondary data. Although these sources were not created for the purpose at hand, they facilitate the development of the required indicators. Data from a census, research, information system, etc., are secondary source data.
3.1.3 ADVANTAGES AND DISADVANTAGES OF PRIMARY AND SECONDARY DATA SOURCES
The advantages and disadvantages of using primary and secondary data sources in preparing and monitoring health indicators are shown in Figure 3.
Figure 3. Advantages and disadvantages of primary and secondary data sources for preparing and monitoring health indicators.
| SOURCES AND DATA |
|---|
| Characteristics | Primary | Secondary |
|---|
| Advantages |
- Greater control over quality, better collection with standardized procedures, better definition of variables and target population, etc.
- Less difficulty in stratifying the indicators according to population subgroups (since the variables have been collected with a view to stratifying the objectives).
- Greater temporal and spatial comparability of indicators, due to the feasibility of instituting more standardized definitions and procedures used for the different groups, while extending the period of analysis (which better serves the proposed objectives).
|
- Lower cost, and greater timeliness of data and indicators.
|
| Disadvantages |
- Greater cost in time and money to obtain the data, which can affect the timeliness of the indicator (particularly in the area of health management) and limit its potential for use.
|
- More analytical effort required to extract, define, and interpret the indicators, due to:
- the possibility that they will prove inadequate (with regard to the relevant objectives and indicators) to define the case and target population;
- doubts as to data quality (due to lack of standardization of procedures and training of those generating the data, etc.);
- Greater difficulty stratifying the indicators according to population subgroups, due to the potential absence of relevant variables that allow for stratification; and
- Greater difficulty making temporal and spatial comparisons of indicators, due to possible changes in definitions and procedures used.
|