SOCECOL 10 Chapter Notes - Chapter 7: Standard Deviation, Statistical Inference, Final Analysis
Document Summary
Methods toward a science of behavior and experience. If our results are indeed due to some factor other than the iv, the internal validity of the experiment is threatened: we seek to eliminate sources of uncontrolled variance that operate in a systematic manner. Some things vary a little, some things vary a lot. The relationship of the between-groups variance and within-groups variance is the same, if there are no treatment effects. Statistical hypothesis testing: consider how the null hypothesis is related to the ratio of between-groups variance to within-groups variance. If the ratio is close to 1, we will not be able to reject the null hypothesis. The smaller the ratio, the less certain we can be that the results are not due to chance: actually, we are never fully certain that our results are or are not due to chance.