PSY 1101 Lecture 5: Problems with True Experiments

• School: University of Ottawa
• Department: Psychology
• Course: PSY 1101
• Professor: C. Messier

Document Summary

Problems with true experiments: often the sample size is small. Can we generalize to the population as a whole from such small samples: experiment must be carried out in a controlled setting (often in a laboratory). Let us suppose we are studying changes in personality as a result of alcohol. We hypothesize that alcohol will make participants less inhibited. They will socialize more and perhaps be more aggressive. Lab results indicate, however, that the participants are not more aggressive or sociable. Perhaps this can be explained by the fact that they are in a lab environment and do not act as they would in a more normal, social bar-type situation. The scientist then develops a theory that men are better at maths because of their male hormones. To test this theory, we need to manipulate the independent variable, the quantity of male hormones, in both males and females.