Knowing background information
(go to Outline)
Be familiar with the accuracy and precision of each of the indicators measured in a survey.
No measurements are perfect. All have measurable biases which lower accuracy and variability which lowers the precision. To properly assess the validity of the survey results, you must know how good its measurements are. Survey reports should describe the measurement methods in enough detail that the reader can easily do such assessment.
Evaluate your survey results relative to known scientific fact.
If the results of your data collection seem to contradict known scientific fact, your methods and results may need some re-evaluation. For example, if a survey shows a very low age-specific mortality rate for children less than 5 years of age in the midst of a measles outbreak in a population with a high prevalence of acute protein-energy malnutrition, perhaps there is something wrong with how you measured mortality.
Of course, interpreting your survey results in light of known scientific fact means you need to know a great deal about the outcomes you include in your survey. As a result, you need to be very knowledgeable and experienced in measuring the outcomes of interest.