SOAN 2120 Lecture Notes - Lecture 5: Internal Validity, Central Limit Theorem, Random Assignment
Document Summary
Assess whether an independent variable changes the value of a dependent variable. One group gets treatment and the other is the control group. However even with ra the two (or more) groups will not be exactly equal in all aspects (random individual variations) If the assignment is random, we can use tests of statistical significance (manipulation vs. random variation) Central limit theorem allows us to assess whether the variation is due to random chance or from the manipulation (p-values) As n increases, individual differences decline ( aka the law of large numbers) Without ra: findings may be attributable to pre-existing differences between the groups. We could compare pre-test scores (dependent variable measures) before the experiment begins (significantly different?- equivalent - proceed as if random assignment had taken place) What about all other potential confounding variables?- this automatically with random assignment. X represents an experimental or effect (the independent variable)