PSCH 343; Statistical Methods In Behavior Science. Week 12 Notes
PSCH 343; Statistical Methods In Behavior Science. Week 12 Notes PSCH 343
Popular in Statistics Methods In Behavioral Science
Popular in Psychlogy
verified elite notetaker
This 5 page Class Notes was uploaded by Katie on Friday April 8, 2016. The Class Notes belongs to PSCH 343 at University of Illinois at Chicago taught by Liana Peter-Hagene in Spring 2016. Since its upload, it has received 11 views. For similar materials see Statistics Methods In Behavioral Science in Psychlogy at University of Illinois at Chicago.
Reviews for PSCH 343; Statistical Methods In Behavior Science. Week 12 Notes
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: 04/08/16
PSCH 343 Week 12 Notes Independent samples t-test Chapter 8 Instructor: Liana Peter-Hagene Review Dependent samples t-test o Types of designs: within subjects, pre-post, matched pairs o Difference scores for each pair of scores – a new distribution o Comparison distribution: means of score differences, mean= 0 o Estimate variance based on your sample of difference scores o Calculate SE for the comparison distribution o More power than between-subjects comparisons Independent samples t-test Research questions about two separate, independent samples o Experimental manipulations, natural groups, drug vs. placebo studies o We no longer pretend we know population means, we rely ONLY on sample data Hypothesis testing with independent samples Research question: Is there a difference between people’s scores based on group (sample) membership (the IV)? We are interested in between people differences If there is no effect of the IV, there should be no difference between Sample A and Sample B means o If there is no effect, the difference between means should be 0 o H : µ = µ o 1 2 o H : 1 ≠ 1 2 o Intervention to reduce aggression in children Number of aggressive behaviors during play session for control and intervention groups Comparison distribution Comparison distribution Comparison distribution Distribution of differences between means Step 2 – Characteristics of comparison distribution Mean: µ =M0 2. Standard error – several steps to get there Estimate the pooled population variance based on sample data o Estimate σ 2 andσ from sample data 1 2 o σ 21 = SS 1df 1 σ 22 = SS 2df 2 o calculate df Total o df Total df1+ df 2 o Calculate the pooled population variance: 2 2 2 o σ Pooled= df1/dfTotal σ +1df2/dfTotal σ 2 Step 2 – Characteristics of comparison distribution Calculate SE for the comparison distribution 2 o Calculate SE for each population variance: o b. Add the two SE values to get the variance of the distribution of mean differences: 2 2 2 § SE DifferenceSE +1SE 2 o Take square root of the sum to get the standard error: § SE = √(SE + SE ) 2 Difference 1 2 Summary of calculations Think backwards: o You need the SE Difference o You need SE & SE 2 1 2 o You need "σ2Pooled" 2 2 o You need σ & σ 1 2 o You need df Total o Number of aggressive behaviors during play session for control and intervention groups Number of aggressive behaviors during play session for control and intervention groups Step 3 – Determine the cutoff score Use the t-tables to find the t value that corresponds to your df Significance level p < .05, two-tailed Total Step 4 – Calculate the t statistic Step 5 – Decision about the null hypothesis Reject, fail to reject? State the conclusion in plain English –what have you learned about the intervention? How large is the effect? Calculate effect size Factorial (Two-Way) Analysis of Variance PSCH 343 Instructor: Liana Peter-Hagene One-way and Factorial (two-way) ANOVA In one-way ANOVAs, we examine one factor (IV) for two or more populations. o Eg., the effect of other people’s presence on helping behavior For two-way ANOVAs, we examine two factors (IVs) for two or more populations. o We examine whether the means of a two or more populations differ for two factors by examining the estimated between and within variance of the populations o Eg., the effect of other people’s presence and gender on helping behavior (would it matter whether it was a woman or a man who dropped the papers?) § IV1: Number of confederates § IV2: Gender of the person dropping the papers § DV: Time it takes for participant to help Factorial Analysis of Variance (ANOVA) Research questions about the effect of two or more independent variables o Experimental manipulations, natural groups o Categorical IVs (nominal), continuous DV (interval, ratio) Same logic as one-way ANOVA: o Compare within group variance to between groups variance o This is done for each independent variable separately -- MAIN EFFECTS o It is also done for the combined effects of the two independent variables n INTERACTION Hypothesis testing with Factorial ANOVA You should always have at least 3 research questions: o Is there a difference between people’s scores based on group (sample) membership (the IV)? (2 main effects) o Is the effect of one IV different at different levels of the other IV? (interaction) Research questions: o Is there an effect of other people’s presence on helping behavior? o Does the gender of the person who needs help make a difference in helping behavior? o Is the effect of other people’s presence on helping behavior different when the person who needs help is a woman versus a man? o Calculating the factorial ANOVA Calculate F ratios for each main effect separately and for the interaction Calculate MS for each main effect and the interaction Between Always use the same MS Within Compare each F ratio to the critical value based on the degrees of freedom ANOVA Table for Factorial Designs An example from an actual research article Two-Way ANOVAs & interactions two factors for two or more populations, e.g.: Sleep deprived versus not sleep deprived Morning encoding versus evening encoding The ANOVA tells us whether the factor ‘sleep deprivation’ has a main effect and whether the factor ‘encoding time’ has a main effect on false memories It also tells us whether the factors interact, i.e.: Is the effect of one factor changed by the other factor? Does the pattern differ for different levels of the other factor? Two-Way ANOVA & interactions For the next few graphs, let’s figure out if they represent: Main effect of other people’s presence on helping Main effect of “helpee’s” gender An interaction Where are the interactions? Where does the pattern between the results differ?
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'