PSYC 3380 Chapter Notes - Chapter 3: Internal Validity, Statistical Process Control, Null Hypothesis
Document Summary
Internal validity: approximate truth of inferences about causal relations. External validity: whether inferences hold across variations in cases, treatments, settings, or measures. Mainly a function of sampling and if sampling is representative. Construct validity: the correct measurement of variables that the researcher intends to study. Conclusion validity (statistical conclusion validity): the appropriate use of statistical methods in the analysis to estimate relations between variables of interest. Nuisance (noise) variables: introduce irrelevant or error variance that reduces measurement precision. Ex: testing grade 1 students in a chilly, noisy room may not yield precise reading scores. Confounding variables: the effects of two variables on the dependent variable cannot be distinguish from each other. Independence of observation: the score of one case does not influence the score of another case. 3 general principles must be met before one can reasonably infer a cause-effect relation: