Statisticalbiasinexperimentalstudiescan resultfromseveralfactors.Lookbackatyournotes aboutstatistical biasandsee whichsourcesyouthinkcouldapplytoexperimentalstudies.List your ideasbelow.
Experimentalstudiescansufferfromnonrepresentativesamplesandundercoverage,asdiscussedinStudentActivitySheet9.Oftentheseproblemscannotbefixed.
1.Whataresomereasonsfordeliberatelyusinganonrepresentativesampleinan experimentalstudy?Whatproblemscouldresult?
2.Ratherthanspendingagreatdeal of time and money to ensure a representative sample,researchersoften use the techniqueslisted below totryto eliminate any statisticalbiasintroducedbythesamplingmethod.Discusshoweachmethodcan reduce samplingmethod biasandthusincrease theaccuracyof astudy’sresults.
a.Randomassignmentoftreatments
b.Blind/double-blind studies
c.Useofcontrolgroups
d.Replication
3.Someofthesameissues(nonrepresentativesamples,undercoverage,replication)affectobservationalstudies.Otherissuesdiscussed(responsebias,wordingofquestions)can alsospecificallyaffectthesestudies.Whatproblemcouldbeoccurringineachstudydescribed?
a.Jadeissotiredof sales callsthat she refusesto answer thephone if she doesnotrecognizethenameornumberontheCallerID.
b.Serenaknowsthecallerwill probablykeep calling, sosheanswers the phone. Ifthecall’ssubjectis anopinionpoll,shepolitelytells thesurveyor that shedoes notwantto participate.
c.DeShawndoesnotworryaboutsalescallssincehisfamilyusescellphonesandnolongerhasalandline.
d.Atthe mall,anoldermanisinterviewingpeopleabouttheirconsumerspendinghabits.Mariadoesnotwanttotellhimshejustboughtaswimsuit, so she says hershoppingbagcontainsanewshirt.
e.Mrs.Gibbsnormallygetsahamburgerfor lunch.When she notices thatsomeone wholookslikeanurseissittinginthecornerwritingdownwhateachpersonorders,Mrs. Gibbsordersa saladwithfat-free dressing.
f.Thelocaltelevisionstationrunsaviewerpollonthenightlynews:“Callusorlogon toourwebsitetogiveyouropinionaboutthenewskatepark.”
4.Severaltelevisionshowshaveasurveycomponentinwhichviewerscallinandvotefortheirfavoritesinger/dancer/entertainer/competitor.Discusstheissuesyouperceivewith this method.
5.Someshowshaveviewersvoteforthecontestantwhomtheywantto leavetheshow.Discussthe issuesyou perceive with this method.
6.REFLECTION:Whatsuggestionsdoyouhaveforimprovingcall-inoronlinepollsorcontests?
7.EXTENSION:Discusswith yourpartner(s) wherein the researchcycleisstatisticalbiasmostlikelytooccur.Suggestatleasttwostrategiesforavoidingbiasandminimizingvariabilityinyourproject.