Video Lecture (from 2019)
Classical Experimental design
Types of experiments:
Classical experimental design
One-shot case study
Validity is the ‘soundness’ or ‘truthfulness’ of a study or a measure.
We show validity by comparing to some accepted external reference.
E.g. We make a two minute test to measure serious mental illness. We validate it by comparing our two minute tests predictions with the medical judgement of a psychiatrist after a one hour interview with each of 60 patients.
Internal validity = ability to state that the relationship between the independent and dependent variable is not spurious. To state that the independent variable is the cause of the change in the dependent variable.
Threats to internal validity
- Selection Bias: e.g milk story
- History: e.g. world event happens
- Maturation: internal changes
- Testing: e.g. pretest could effect DV
- Experimental Mortality: e.g. some type of people drop out
- Contamination or Diffusion of Treatment: e.g. experimental group communicates with control.
- Experimenter or participant expectancy: hence use of double blind experiments and placebos.
External validity = ability to generalise experimental findings to events and settings beyond the experimental setting itself.
Threats to external validity
- Participants are no representative: e.g all first year psych students
- Artificial setting: e.g. you behave different in lab to real world
- Artificial treatment: e.g. you behave differently when experiencing real racism rather than seeing racism in a lab
- Reactivity: Modifying your behavior because you are aware of being studied e.g. when you are tested you do better because of attention.