An experiment is defined as a controlled procedure carefully designed to test a hypothesis, about the effect of one or more independent variables on an outcome measure. The scope of the EDA is limited to experiments involving living animals.
The animal characteristics node contains the characteristics of the animals used in the experiment, such as the species, strain, sex, age, weight and so on.
An independent variable is an effect which potentially influences the outcome of an experiment. There are two types of independent variables:
- Independent variable of interest, which the researcher specifically manipulates to test a predefined hypothesis. It is also known as the predictor variable, or the factor of interest. This can be for example a drug or a surgical intervention.
- Nuisance variable, which is of no particular interest in itself but needs to be controlled or accounted for in the statistical analysis, so that it does not conceal the effect of an independent variable of interest. Nuisance variables are defined as other sources of variability or condition which may influence outcome. This can be for example the time of the day animals receive a treatment, or the body weight at the start of the experiment.
The experimental unit is the entity subjected to an intervention independently of all other units. It must be possible to assign any two experimental units to different treatments groups. Generally it is an individual animal, but it could be a cage of animals, a litter, part of an animal such as a limb or an animal for a period of time.
This is the process by which experimental units are assigned to experimental groups. It can be achieved using various strategies, for example a complete randomisation, a randomisation within blocks and/or within factors. In the EDA diagram, an allocation process produces groups.
An intervention is a process which a group of animals (or experimental units) is subjected to, such as a surgical procedure or a drug injection. In the EDA there are several types of interventions: surgical, pharmacological, pathogen infection and euthanasia. For interventions which cannot be assimilated to any of the aforementioned categories, ‘other intervention’ can be used.
A measurement is a process which a group of animals (or experimental units) is subjected to, to collect data; it is recorded as at least one outcome measure. In the EDA, three different types of nodes are used to represent data collection. The outcome measure node and two types of measurement nodes: measurement, for simple measurements, and repeated measurement, when the same measurement is repeated several times.
Also known as the dependent variable, or the response variable. A measurement can be recorded as one or several outcome measures. For example, if body size is measured, one outcome measure could be the length of the animal and the other the body weight. Each can be fully defined in the properties of their respective node.
A data transformation node is used to indicate pre-analysis processing of the data, such as outcome expressed as a percentage of the baseline or a data transformation to enable parametric analysis such as a square root transformation.
In comparative experiments, statistical analysis is used to estimate effect sizes and determine whether the null hypothesis can be rejected. In the EDA, an analysis node receives input from at least one outcome measure and independent variables of interest and nuisance variables are included as factors for the analysis.
There are two types of validity referred to in the EDA:
- Internal Validity, the extent to which the results of a given study can be attributed to the effects of the experimental intervention, rather than some other, unknown factor(s) (e.g. inadequacies in the design, conduct, or analysis of the study introducing bias).
- External Validity, the extent to which the results of a given study enable application or generalisation to other studies, study conditions, animal strains/species, or humans.