Homework ReviewIntroductionControl GroupsRandomized DesignBlinded ExperimentsAssignmentUnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentUnderstanding ExperimentsLecture 10Section 3.5Robb T. KoetherHampden-Sydney CollegeFri, Jan 30, 2008UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentOutline1Homework Review2Introduction3Control Groups4Randomized Design5Blinded Experiments6AssignmentUnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentHomework ReviewExercise 2.32, p. 122A 1-in-14 systematic sample will be taken from a populationof 555 members of an art appreciation club. The 555members are labeled from 1 through 555.(a) What is the chance that the member with label 17 willbe selected?(b) Suppose the sample is selected and the number withlabel 17 is included. What is the sample size for theresulting sample?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentHomework ReviewSolution(a) Because we are using 1-in-14 sampling, the chance is1/14.(b) If 17 was included, then the selected random numbermust have been 3 (3 + 14 = 17).Also, 555/14 = 39914.That means that there are 39 full blocks of size 14 andone partial block of size 9.Since 3 ≤ 9, we will pick up one member from the partialblock.That will make the sample size 40.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentIntroductionSuppose a drug is given to 100 patients suffering froma particular disease.After 2 weeks, 90% of the patients have recovered.The researchers conclude that the drug was effective.What is wrong with this?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentIntroductionSuppose a drug is given to 100 patients suffering froma particular disease.After 2 weeks, 90% of the patients have recovered.The researchers conclude that the drug was effective.What is wrong with this?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentIntroductionSuppose a drug is given to 100 patients suffering froma particular disease.After 2 weeks, 90% of the patients have recovered.The researchers conclude that the drug was effective.What is wrong with this?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentIntroductionSuppose a drug is given to 100 patients suffering froma particular disease.After 2 weeks, 90% of the patients have recovered.The researchers conclude that the drug was effective.What is wrong with this?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentTreatment and Control GroupsDefinition (Treatment group)The treatment group is the group that receives thetreatment.Definition (Control group)The control group is similar to the treatment group in allrespects except that it does not receive the treatment.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentRandomized DesignWhy would it be wrong to allow the individualsthemselves to choose whether to be in the treatmentgroup or the control group?Why would it be wrong for the researchers to decide,subject by subject, who goes into which group?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentRandomized DesignDefinition (Randomized design)A randomized design is a design in which the subjects arerandomly assigned to either the treatment group or thecontrol group.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentRandomized DesignRandomized DesignSuppose that there are 100 subjects.Number them 1 - 100.Then use a random number generator to obtain 50(distinct) random numbers from 1 - 100.Those 50 subjects would be assigned to the treatmentgroup.The rest would be assigned to the control group.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentPossible BiasAre the subjects in the treatment group aware of thepurpose of the experiment?Are the subjects in the control group aware that theyare not receiving the drug?Will it make a difference?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentTwo Types of BiasDefinition (Response bias)A sampling method exhibits response bias if the subjectsgive what they perceive to be the desired response ratherthan the true response.Definition (Experimenter bias)A sampling method exhibits experimenter bias if theobserver records the desired values rather than the trueobserved values.How can we avoid these biases?UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentPlacebosDefinition (Placebo)A placebo is a treatment, usually a pill, that is known to haveno effect.Definition (Single-blind experiment)A single-blind experiment is an experiment in which thesubjects do not know who is in the treatment group and whois in the control group.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentPlacebosEverybody in the treatment group is administered thedrug.Everybody in the control group gets the placebo.The researchers look for differences in the groups’recovery rates.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentDouble-Blind ExperimentsDefinition (Double-blind experiment)A double-blind experiment is an experiment in which neitherthe subjects nor the observers know who is in the treatmentgroup and who is in the control group.UnderstandingExperimentsRobb T.KoetherHomeworkReviewIntroductionControlGroupsRandomizedDesignBlindedExperimentsAssignmentAssignmentHomeworkRead Section 3.5, pages 168 - 171, 173 - 182.Let’s Do It! 3.5, 3.6, 3.7, 3.8.Page 171, exercises 17 - 22.Page 182, exercises 25, 26, 28 - 31.Chapter 3 review, p. 196, exercises 39 - 43, 45, 47, 49 -51, 59, 60, 67, 68,
View Full Document
Unlocking...