Stat 217 – Day 3 Topic 3: Drawing ConclusionsLast Time – Drawing ConclusionsActivity 3-2 (p. 36)Activity 3-2Slide 5Activity 3-4 (p. 40)Other examplesParameter vs. Statistic (p. 35)Slide 9Lab 1: Friend or FoeSlide 11Possible explanations“Simulation”Slide 14Stat 217 – Day 3Topic 3: Drawing ConclusionsLast Time – Drawing ConclusionsIssue #1: Do I believe the sample I have is representative of the population that I am interested in for this issue (generalizable)?Many possible sources of sampling biasVoluntary response, bad sampling frame, nonresponseNOT sample size…population???sampleActivity 3-2 (p. 36)(a) Observational units = studentsVariable = whether or not intentionally injured themselves(b) Population of interest = “college students”Sample = students that responded(c) sample size = 2875Activity 3-2(f) No, they only sampled from two universitiesThat are rather prestigious and have different stress levelsSo students probably more likely to self injureThat have wealthier students, supportive familiesSo students probably less likely to self injureNo, voluntary responseSo those who have experience, strong feelings more likely to respondIf embarrassed, less likely to respondGive main reason why, try to argue a particular direction (over or under estimate), make sure connect to the variable being measuredLast Time – Drawing ConclusionsIssue #2: Can I draw a cause and effect conclusion when comparing groups (causation)?Explanatory variable vs. Response variablesampleExplanatory Group 1Explanatory Group 2Response variableActivity 3-4 (p. 40)A confounding variable (p. 39) changes with the explanatory variable and possibly also affects the response variable, can’t distinguish whichObservational unitsSports SectionVariety of examplesExplanatory variable Response variablePerformance in courseStats studentsEarly time Later timeNon-athletesathletesNot necessarily confounding variables:• Some students study more than others (doesn’t differ between groups)• Instructor (doesn’t differ)• Easier to find parking in the morning (not clearly related to response)Other examplesCEOs are taller than non-CEOsShifts with Kristin Gilbert working saw higher death ratesActivity 3-5 (p. 41):Quebec children with more sleep at night are less likely to be obeseIn the late 1940s, polio cases increased with the consumption of ice cream and soft drinksParameter vs. Statistic (p. 35)Parameter is a number that describes (the variable in) a population63% of all voters that actually voted for Roosevelt (37% that voted for Landon)Average number of hours Cal Poly students slept last nightStatistic is a number that describes (the variable in) a sample57% of voters who indicated they would vote for Alf LandonAverage number of hours of students in this class that slept last nightActivity 3-2 (p. 36)(d) 17% is a statistic because it described the sampleWhat would the parameter be?The proportion of all college students that have injured themselves intentionallynumberpopulationvariableLab 1: Friend or FoeExperiment 1Lab 1: Friend or Foe14 of 16 infants picked the helperDoes this convince you that these infants are generally more likely to pick the helper than the hinderer based on the videos?Discuss with neighborJot down ideasPossible explanationsThese infants genuinely prefer the helper toyThese infants do not genuinely prefer the helper toy but we happened, by chance alone, to get a large majority picking the helper in our sample.We can investigate this second case – If it is the case there is no preference, how often get 14 out of 16 picking the helper“Simulation”Instead of working with infants, we will assume the infants behave like a coin toss. Assuming same probability for each infantToss your coin 16 times, to represent the 16 identical infants, and record the number of headsIs it surprising to get 14 heads when we know heads and tails are equally likely?What conclusion does this point to?For Thursday (Library)Pre-lab for Lab 1 by noonWill email back feedbackDon’t need to bring your textDo bring a USB for saving your work to continue outside of classSit with a partnerFor Monday: Activity 4-1
View Full Document