SOCECOL 10 Lecture Notes - Lecture 12: Null Hypothesis, Standard Deviation, Condom
Document Summary
Extending the logic of experimentation: within-subjects and matched-subject approaches. Designs in which the same participants are exposed to different experimental conditions. Every participant serves in every group: receives every level of your independent variable. Compare the same people after different treatments: following different treatments, compare measures/results from the same group of people instead of assigning groups to different treatments (e. g. ) condom studies: brands a, b, and c. The total number of participants is reduced: more efficient because you can get more information out of participants, yet disadvantage because external validity is reduced. Fewer types of people in your study means less generalization to the. Error variance is reduced general public: reduces your within-groups variance, reduce standard deviation of your groups. Standard deviation within the group is smaller. Inferential test is more sensitive to changes in the treatment. More likely to be able to reject the null hypothesis: more likely to get p<0. 05 with within-subjects.