STAT 110 1nd Edition Lecture 3 Outline of Previous Lecture I. Computing Confidence Intervals for Stratified Random Samples II. Cluster SamplingIII. The Difference between Cluster Sampling and Stratified Random SamplingIV. Experiments Good, and BadOutline of Current LectureI. Lurking/Confounding Variables II. Placebos and the Placebo Effect III. Randomized Comparative Experiment (RCE)IV. Principles of Experimental Design V. Comparing OutcomesCurrent LectureI. Lurking/Confounding Variables a. A lurking variable is a variable that has an important effect on the relationship among the variables in the study, but it is not an explanatory variable. i. A lurking variable is often something we would probably want to measure, and is a possible alternative explanation for the outcome. b. Two variables are confounded when their effects on a response variable cannot be distinguished from each other. An outcome is confounded when we cannot tell what caused the outcome. II. Placebos and the Placebo Effect a. A placebo is a “dummy treatment” in which subjects believe they are receiving a treatment, but they are really not. For example, if the treatment is supposed to be some kind of pill or medication, subjects receiving a placebo will receive a pill with no active ingredients. b. The placebo effect is a subject’s response to a placebo, even though the placebo has no effect on them. III. Randomized Comparative Experiment (RCE)a. The following steps are performed in an RCE:i. First, subjects are randomly assigned to different treatment groups. 1. The groups should be similar in all respects except the treatmentThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.ii. Secondly, two or more treatments are compared (usually to a control group) to ensure that the differences between the groups can be attributed to the treatment, not group differences. IV. Principles of Experimental Design a. A good experimental design should:i. Control the effects of lurking variables on the response variables, usually by comparing two or more treatments. ii. Randomize the subjects to different treatments.iii. Use enough subjects to reduce variation in the outcome due to chance.V. Comparing Outcomes a. We DO NOT expect the same results from different treatments. If the observed differences between the treatments are large enough that we can conclude they are not due to chance, we say the results are statistically