PSYB04H3 Lecture Notes - Lecture 11: Measuring Instrument, Declared Rare And Priority Flora List, Observer-Expectancy Effect
Document Summary
Chapter 11: more on experiments: confounding and obscuring variables situation noise, p. 333 power, p. 334. When integrating an experiment, internal validity is the priority. Three of the most common threats to internal validity are: design confounds, selection effects and order effects: all three threats involve alternative explanation for the results. If the pasta served to the large-bowl group had looked more appetizing than the pasta served to the medium-bowl group, that would have been a design confound. It would not be clear whether the bowl size or the quality of the pasta caused the large- bowl group to take more. However, the improvement may have been caused by the therapy, or it may have been caused by greater overall involvement on the apart of the parents who elected to be in the intensive therapy group. These types of threats are just the beginning (oh no!). there are about 12 ways than an experiment might be at risk of confound.
