From: Basic Methods of Medical Research, II Edition
by A. Indrayan (AITBS Publishers, Delhi)
Glossary of Methodological Terms
(Some of these terms have not occurred in the text)
In the explanation, the terms included in this Glossary are in italics. They can be referred within this Glossary. Some terms are explained in more detail and more succinctly here than in the text. For others, the explanation in the text is better.
Absolute risk — Same as risk. Contrast it with relative risk.
Absolute risk reduction — Risk difference between the control group and the intervention group.
Accuracy — Truthfulness or correctness of a particular value of a measurement. Age recorded as 7 years 4 months and 14 days is more accurate than recorded only as 7 years, although this additional accuracy may be redundant.
Addition rule (of probability) — The probability of occurrence of one of two or more mutually exclusive events is the sum of the probabilities of their individual occurrence.
Additive model — A model stating that combined effect of two or more factors is the sum total of their individual effects – valid when there is no interaction.
Adjusted correlation — Same as partial correlation.
Adjusted odds ratio — The odds ratio obtained after eliminating the effect of other concomitant variables that might be affecting the OR. This is generally obtained by including these concomitants in the logistic regression.
Adjusted rate — The net rate obtained after eliminating the effect of other concomitants that might be affecting the rate. This adjustment increases the comparability between rates in different segments of population.
Adjustment — A procedure by which the effect of structural differences in the two or more groups is minimized—thus improving the comparability. Common methods of adjustment are regression and standardisation.
Aetiology diagram — A diagram that depicts the inter-relationship of various aetiological factors leading to the disease.
Aetiological factor — The characteristic that contributes to the occurrence of disease or a health condition. It may or may not be a causal factor.
Age-adjusted death rate — See age standardisation.
Age standardisation — A procedure of adjustment to remove the effect of differences in age composition of the groups to increase comparability. This adjustment is required when, for example, one group has older subjects than in the other and the outcomeof interest is death. For deaths, this is called age-adjusted death rate. This could be done as direct standardisation or indirect standardisation.
Agreement — When two procedures, two observers, or two sites, etc., tend to give same result in each subject, they are said to be in agreement with one another. For qualitative data the extent of agreement is statistically measured by Cohen’s kappa and for qualitative data by limits of disagreement.
Aleatory uncertainties — Uncertainties arising from variation in the factors internal to the system such as biological, psychological and environmental. Contrast it with epistemic uncertainties.
Allocation bias — The systematic error in results arising from specific allocation of subjects to the test and control group, such as due to nonrandom allocation.
Alpha error — Same as Type I error.
Alpha error rate — The probability of Type I error.
Alpha level — Same as significance level.
Alternative hypothesis — A plausible hypothesis that is accepted by default when the null hypothesis is rejected.