Study Guides (248,581)
Canada (121,629)
Psychology (776)
PSYC 2001 (29)
all (12)

psyc 2001 d 2013 notes res methods.docx

14 Pages
67 Views
Unlock Document

Department
Psychology
Course
PSYC 2001
Professor
All Professors
Semester
Winter

Description
October 15 2013Psyc 2001 DChapter 8 Experimental Designs BetweenSubjects DesignHow can we compare groupsWithin Subjects Designsto be discussed later Use the same group of individuals and subject them to the different treatmentsBetween Subjects Designsdifferent individuals in different groupsalso known as an Independent Measures Experimental DesignCan you think of advantages and disadvantages for both approachesoAdvantages independent score protects against practice experience fatigueboredom contrast effects oDisadvantages Need many participantsIndividual Differences can be confounding variables ie group A does better than group Bwhat possible confoundsIndividual Differences can also produce high variability in scoresIndividual Differences and ConfoundingConfounding from individual differences is also called assignment bias ie age SESConfounding can occur because of environmental variables ie large room vs small room Solution Equivalent GroupsoCreated equally same selection processoTreated equallyoComposed of equivalent individualsEquivalencyRandom assignment Matching oMust identify correct cariable first oMust measure it oTime consumingcostlyoCant match everythingFinally holding groups constant can create equivalency ie use all femalesIndividual Differences and VariabilityIn order to compare group scores we must understand if a difference in average scores between groups is significant Comaring averages is not enough Ex weight and heightGroup A avg height 56Group B avg height 66Group A avg weight 160Group B avg weight 170More on VarianceIf there is not much variability in scores on a given variable in a population then it is much easier to detect differences between groups If there is a lot of variability in scores on a given variable inthe population then it is much more difficult to detect differences between groups WhyGoalminimize variance of scores within reatementsgroupsSolve byoStandardized procedures and treatment settingsoLimit individual differences so that all groups are equivalent ex use same ageoRandom assignemtn and matching oUse a large sample size it helps statistically to minimize high variance in your sample Other issues with BetweenSubjects DesignAttritionunequal groupsCommunication can lead to rivalry or demoralization Statistical significanceGoal are your results due to chance or is there a significance statistical to your findings Usually set it at 05 if you find significance at 05 it means there is a 95 likelihood that your results are due to your treatment and a 5 likelihood they are due to chance not your treatmentStatistical analysis of Between GroupsSimplest form Twogroup designUse a ttest to test significance of mean group differencesLimitation You can only compare 2 groups 2 levels of your IV More complex form Multiplegroup designUse ANOVA to determine if significant differences exist then follow up with posttests to determine which groups are significantly different from each otherLimitation If you use too many groups they may be similar to the point you cant detect differences too many levels of your IVComparing ProportionsIf the IV is on a nominal or ordinal scale then there is no numerical value for scores and therefore we cannot compute averages for groupsCannot use an independent ttest or ANOVAUse CHISQUARE
More Less

Related notes for PSYC 2001

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit