How precise and accurate does the information have to be to meet your needs?
Page 14 of 40
Data are only good or bad, adequate or inadequate, valid or invalid when compared to your information or research goals. Match the quality of your sources to your goals rather than trying to decide if a source is good in the abstract.
Go beyond the subject matter to ask:
How much evidentiary support do I need? [For example, material prepared for use in news reports, the courts or scholarly journals may require different kinds of proof or corroboration.]
Who should the estimates cover? [This is sometimes called the “reference population” because it’s the population to which the estimates “refer.”]
What measures do you need that shows the quality of the results you are going to use?
Even though there are differences in the amount of detailed the information needed for different purposes each researcher wants timely, accurate and complete information. The results must include sufficient facts to assess the indicators strengths and weaknesses. The description of how data were collected and measures of their statistical properties is important for their appropriate use.
Appreciating inherent limits is essential to effectively derive estimates from data.
Web publication makes methods information more available because responsible compilers can now include technical descriptions as they present the results. [In the past, the cost of printing this information discouraged release.]