Validity
The truth of, correctness of, or degree of support for an inference
Power
The probability of correctly rejecting a false null hypothesis; in an experiment, usually interpreted as the probability of finding an effect when an effect exists
Sampling error
The part of the difference between a population parameter and its sample estimate that is due to the fact that only a sample of observations from the population are observed.
Correspondence theory of truth
An epistemological theory that says a knowledge claim is true if it corresponds to the world.
Coherence theory of Truth
An epistemological theory that says a claim is true if it belongs to a coherent set of claims
Pragmatic Theory of Truth
An epistemological theory that says a claim is true if it is useful to believe that claim.
Internal validity
The validity of inferences about whether the relationship between two variables is causal. (Can you make a causal claim from this data?)
External validity
The validity of interferences about whether the casual relationship holds over variations in persons, settings, treatment variables, and measurement variables. (Can you make a generalization from this data?)
Statistical conclusion validity
The validity of inferences about covariation between two variables (Can you make a statistical conclusion from this data?)
Construct validity
The degree to which inferences are warranted from the observed persons, settings, and cause-and-effect operations sampled within a study to the constructs that these samples represent. (What are the constructs of this data?)
Null Hypothesis Significance Testing
The practice of testing the hypothesis and then declaring that an effect exists only if p < .05
Null hypothesis
The hypothesis being tested, traditionally that there is no relationship between variables.
Unreliability
A lack of consistency
Reliability
Consistency
Floor effects
Responses on a variable approach the minimum possible score so that further decreases are difficult to obtain
Ceiling effects
Responses on a variable closely approach the maximum possible responses so that further increases are difficult to obtain.
Interactions
In experiments, when the effects of treatment vary over levels of another variable.
Attrition
Loss of units; in randomized experiments, refer to loss that occurs after random assignment has taken place.