Skip Navigation Bar

National Information Center on Health Services Research and Health Care Technology (NICHSR)

Page 6 of 16
Term Definition
Benchmark Results that are used as a standard against which an observation can be calibrated. For example, a physician may compare the height and weight of a patient to the averages for their age group as reported by the National Health and Nutrition Examination Survey.
Census An enumeration of an entire population. For example, the decennial census of population and housing includes all persons living in the United States as of the census date.
Coding Assigning categories to observations. In health statistics standard coding conventions like the International Classification of Diseases is used as the basis for coding.
Confidence Interval In sampling, the range around an estimate defining the likely occurrence of the true value. For example, we might say that a health condition is present in 10% of a population with a confidence interval of plus or minus 2% meaning that the true number lies between 8% and 12%.
Controlled study A study that uses control groups to measure the effect of an interventions. For example, one type of controlled study design assigns subjects to experimental and control groups randomly so that the effect of extraneous influences can be taken into account.
Cross sectional observation A study used cross sectional observation to support inferences. This type of study relies on comparisons across groups without intervening in their behavior. [Contrast this with a controlled study.]
Cross validation Comparing the results from studies in order to judge the outcome and effects of factors controlling for the artifacts of research procedures.
Page 6 of 16