Glossary terms for 'V'
|Validity||The degree to which a measurement represents the phenomenon of interest. For example, the score on a quality of life questionnaire is valid to the extent that it really measures quality of life.|
|Variability||The amount of spread in a measurement, usually calculated as either the standard deviation or the standard error of the mean. For example, if change in body weight produced by a diet ranges from substantial weight gain to substantial weight loss, the change is highly variable.|
|Variable||A measurement that can have different values. For example, sex is a variable because it can take two different values?male or female. See also categorical, continuous, dichotomous, discrete, nominal, ordinal, outcome, predictor, and confounding variable. |
|Verification bias (also called work-up bias or referral bias)||A bias in the assessment of the accuracy of a test that occurs when participants selectively undergo disease verification by gold standard testing based partly on the results of the study test itself. For example, if a study of the accuracy of chest percussion to diagnose pneumonia, included only patients who had a chest x-ray, and if those with dullness to percussion were more likely to get an x-ray, the sensitivity of percussion would be falsely increased, and specificity falsely decreased. |
|Visual analog scale||A scale (usually a line) that represents a continuous spectrum of answers, from one extreme to the other. Typically, the line is 10 cm long and the score is measured as the distance in centimeters from the lowest extreme. For example, a visual analog scale for the severity of pain might present a straight line with “no pain” on one end and “unbearable pain” on the other end; the study participant marks an “X” at the spot that best describes the severity of his pain. |
|Vulnerable persons||Potential study participants who are at greater risk for being used in ethically inappropriate ways in research. For example, people with cognitive impairments or communication problems may be unable to give fully informed consent to participate in research. Other examples include children, prisoners, fetuses, and persons of low socioeconomic status.|
Glossary material from Hulley SB et al. Designing Clinical Research, 4th ed. Philadelphia, Lippincott Williams & Wilkins, 2013.