Limitations of surveys
(go to Outline)
When deciding whether or not to carry out or commission a survey, you must keep in mind the limitations of surveys:
Individual surveys are not good at following trends in real time or over short periods of time.
Because surveys collect data at a single point in time, it is difficult to measure changes in the population unless two or more surveys are done at different points in time. Such repetition is often expensive and time-consuming, making frequent periodic surveys impractical.
Individual surveys generally cannot provide strong evidence of cause and effect.
Because surveys collect data on disease and risk factors at the same time, you often cannot tell which came first, the risk factor or the disease. Without this temporal association, it is very difficult to prove that the reputed risk factor actually causes the disease. For example, a survey in a refugee population may find high incidence rates of diarrhoea, a low prevalence of access to clean water, and a high prevalence of malnutrition. However, a single cross-sectional survey cannot disentangle the different contributions of each of these factors to the others; in fact, it is likely that malnutrition and limited access to clean water contribute to high diarrhoea incidence, while the high diarrhoea incidence also contributes to malnutrition.
Other constraints to using surveys to gather data:
- Insecurity limiting access to the population of concern
- The lack of time to carry out a survey
- The lack of funding necessary to carry out a survey
- The lower priority for carrying out a survey because of competing urgent tasks