Introduction to Research in the Behavioral Sciences
Introduction to Research in the Behavioral Sciences PSYC 218
Popular in Course
Popular in Psychlogy
This 5 page Class Notes was uploaded by Dorris Purdy on Friday October 23, 2015. The Class Notes belongs to PSYC 218 at University of Idaho taught by Staff in Fall. Since its upload, it has received 37 views. For similar materials see /class/227927/psyc-218-university-of-idaho in Psychlogy at University of Idaho.
Reviews for Introduction to Research in the Behavioral Sciences
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/23/15
Psych 218 Introduction to Behavioral Research Methods Week 14 Lecture 28 Outline of Today s Lecture Last lecture we discussed Inferential Statistics multi factor ANOVA Parametric vs nonparametric Inferential Statistics Assumptions Posthoc tests vs planned comparisons Today we will discuss Exam Review Outline of Course Topics Eplstemologyi theorlesi nvpotneses dlscol lflrrnatlorl strong inference al vs Relatlorlal Research Designs ni 9x as so in I3 0 3 a I a a o I g a e Measurement rellabllltviscalesiSamplng Research Etnics Noneexperirnental researcn tecnnioues and Survevs Experlmerltal researcn Witnin vs between vs mlxed designs rnalan researcn Descriptive statistics rneasures of central tendencv and dis ersion lnferential statistics 7 relational researcn gtcorrelation and regress on e Experlmerltal researcn gttests for differences arnong rneans Hest ANOVA Rte portlrlg Research results Wl ltll lg papers presentations APA s v e z 3 o gtlt u e a o E Example Definition Questions Define each of the following terms using pictures or examples if necessary Splitplot Research Design Parsimony Example Short Answer Question Condition Ordering in thhin Subiects Designs Joe Schmols designing a Witninsubiects experlmerlt testing tne effects clotning color on attractiveness unique presentat n order eacn Sublectwlll receive Feel free to rnodifv Joe s experlmerltal design if vou feel itWill nelp increase efflclel lcy but make sure vou expllcltly describe and iustifv vo ur modl c l Example Short Answer Question Variables Gerald needs your help in designing a su y 0 ss ss whether morning peoplequot are more productive workers ldentifv and describe examples of lrldeperlderlt dependent intervening extraneous and confounding variables for tnis studv lf anv oftnese variables do not eXlSI forcerald s studv state w v n is cerald using7 E 09 is 1 0 to I a 6m 28 I 01 e 9 10 8 a 2 I Q I e 3 o a E g z 2 Example Essay Question Using a concrete scienti c example generate a set of premises and conclusions based on conditional reasoning properly identifying the 4 poss bilities for premise 2 and whether the accompanying conclusion is logically valid or not Do this in the context ofboth con rmational and discon rmational reasoning Discuss which premises an conclusions are most scienti cally useful and which are e Psych 218 Outline of Today s Lecture Introduction to Last lecture we discussed B h vi r l R r h e a 339 gsea C lnferential Statistics multifactor ANOVA e O S Today we will discuss Parametric vs nonparametric Inferential week 1439 Lecture 27 Statistics Assumptions Posthoe tests vs planned comparisons Example of 2way WithinSS Graph of WithinSS ANOVA ANOVA as uwmun Sul a s ms Estimated Marginal Means of Postural Ease l Postural Ease l l Eves Closed Eves Open Availability ofvisual Information Some Final Points on Multifactor AN OVA Independence of Main Effects and Interactions the general linearmodel 1 Example 2 factors IVs A and B arm DV Score AMEA BMEB ABlAB d error 1321 Where MEAis the main effect of independent variableA MEBis the main effect of independent variable B IABis the interaction between variablesA and B 17229 17229 Parametric vs Nonpa ra metric Statistics Assumptions of Parametric Statistical Tests ttests NOVA Pearson r etc DV has interval or ratio scaling properties DV scores randomly sampled 39om population Sampling distribution ofthe mean is normal Mthingroup variances are homogeneous IV affects means but not variances Homoskedasticity vs heteroskedasticity If any of these assumptions is seriously violated a nonparametric test must be used How do You Know If Your Data Seriousy Violate the Parametric Assumptions Insure DV has interval or ratio scaling properties and that scores are randomly sampled from population 9 assignment of subjects to conditions is ran om Check to see if sampling distribution of the mean is normal e difficult to assess Without replication of experlmerlt however e violations o he norrnality generally have little effect on the validity ofthe ANOVA as long as the distributions of the populations have sirnilar shapes e g i all negatively skewed Mthingroup variances are homogeneous IV affects means ut not variances e cornpute variance for each conditiongroup and cornpare e ForANOVA variances rnust differ by at least 4 times before their inequality invalidates the statistical test Example of 2way Within8s ANOVA Output 2 Factor Univariate Analysis of Variance is used to analyze the data Table of Means and Standard Errors from SPSS 5 vlslnfo nonvls Measure MEAS 95 Confidence lnteival What Do Do If My Data Violate the Parametric Assumptions Despair No Nonparametric statistical tests can be used a e no distri utional assump ions Less restrictive scaling assumptions Caveats Nonparametric are typically not as powerful as parametrictests 9 more likely to commit a Type II error Nonparametric tests do not exist for more complex designs ANOVA Planned and Unplanned posthoc Comparisons ANOVA identi es signi cant main effects an t r ctlons Em but does not identify which g5 speci c conditions actually j s22 Planned and unplanned Emu y comparisons allow you to tease out speci cally which our experimental conditions differ Nd n Use in Use Cell Status ANOVA Planned and Unplanned posthoc Comparisons Planned Comparisons arise from your hypotheses and are plannedbefore the data is analyzed if only a few conducted no correction for familywise error probability pyramiding is required EXAMPLE Using a cellphone while driving will increase breaking response time to a child entering the street when driving in busy traffic but not when driving alone ANOVA Planned and Unplanned posthoe Comparisons Unplanned comparisons are those that you choose to do a er analyzing your data they are not necessarily related to your hypotheses Example compare all poss ble pairs ofmeans to see which pairs signi cantly differ Must adjust uof t test to account for familywise error probability pyramiding W 1 1 DOC or Use posthoc tests that automatically adjust uScheffe test DunnetttestTukey test NewmanKeulstest Duncan test Fisher test
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'