Stat 217 – Day 2 Topic 3: Drawing ConclusionsLast Time – Topic 1 Wrap UpActivity 1-2 (p. 6)Activity 1-4Activity 1-6Preliminaries (p. 32)Activity 3-1Slide 8Slide 9Slide 10Def (p. 37)Brief Summary of new terms todaySlide 13Stat 217 – Day 2Topic 3: Drawing ConclusionsPreliminaries (p. 32)• Do you believe Elvis is still alive?• Guess the percentage of adult Americans who believe Elvis faked his death.• Would you say you consume candy rarely, sometimes, or often?Last Time – Topic 1 Wrap UpA variable is a characteristic that varies from observational unit to observational unitVariables can be classified as categorical (a group designation) or quantitative (numerical value).The research question generally summarizes the variable or compares the variable across different groups.Can I say something about a larger group?Is there a genuine difference between the groups?Activity 1-2 (p. 6)Obs units = 50 statesa) Gender of state’s governorb) Number of states that have a female governorc) Percentage of the state’s residents older than 65d) Highest speed limit in the statee) Whether or not the state’s name contains one wordf) Average income of the adult residents of the stateg) How many states were settled before 18653Activity 1-4(b) Psychologist interviews 150 counselors to predict outcome for couple to see if they were right more than half the timeObservational unit = 150 counselorsVariable = was prediction correct (binary)Summary: How many counselors correct?Research Question: Correct more than half the time?Activity 1-6Of 257 shifts that Gilbert worked, 15.6% had at least one patient dieOf the 1384 shifts she did not work, 2.5% had at least one patient dieObservational units: shiftsVariable 1: did she work the shift? (binary)Variable 2: did a patient die?Preliminaries (p. 32)Do you believe Elvis is still alive?Guess the percentage of adult Americans who believe Elvis faked his death.Activity 3-1Goal: Estimate the proportion of adult Americans who believe Elvis is alivePopulation = the entire group of observational units of interestSample = the part of the population we gather data fromSample size = number of observational units in sampleActivity 3-1(a) Population and samplePopulation = all adult AmericansSample = those who called in(b) Agree with the 56%?Voluntary response = those who choose to respond (especially at a cost) probably tend to feel differently than those who do notBad sampling frame: not all adult Americans list to the radio, radio listeners probably differ in opinions about ElvisActivity 3-1Literary Digest poll(c) Population, samplepop = all voterssample = 2.4 million respondents(d) Agree with 57%?voluntary response: those unhappy with current president probably more likely to replybad sampling frame: those with vehicles and phones in 1936 tended to be wealthier => overrepresented republicansFinish Activity 3-1 at homeDef: parameter vs. statisticDef (p. 37)The variable whose effect you want to study is the explanatory variable. The variable that you suspect is affected by the other variable is the response variable.Explanatory → ResponseExamples:Eating cheerios lowers your cholesterol (EV: whether eat cheerios, RV: cholesterol)Running barefoot reduces impact (EV: barefoot vs. shoes; RV: impact strength)Brief Summary of new terms todaySample vs. populationSample may not be representative of populationExplanatory vs. Response variableWhen can I attribute changes in the response variable to the explanatory variable?CheeriosNo cheerioscholesterolcallersadultsTo Turn in, with Partner, on a separate piece of paper, with both namesActivity 3-2 (p. 36-7) parts (a) and (e)Activity 3-4 (p. 40) parts (a) and (c) “Give an argument why we can’t attribute the lower academic performance to the sports examples for this study” (suggest an alternative explanation for why that group did worse)For Wednesday (classroom)Finish Activities 3-1, 3-2 (parameter vs. statistic)Self-check Activity
View Full Document