 Limited time offer 20% OFF StudySoup Subscription details

# NYU - PSY 001 - Study Guide - Final

### Created by: Julia_K Elite Notetaker

> > > > NYU - PSY 001 - Study Guide - Final

NYU - PSY 001 - Study Guide - Final

##### Description: Chi Square, One and Two Way Anovas, Post Hoc Tests, Simple Effects, One and Two Group T-tests, deciding which test-statistic to use, Matched T-tests, Counterbalancing and Designs, Correlation, Interactions, Confidence Interval, Bonferroni, Power, Type 1 and 2 Errors, Coefficients of Det and Non Det, Bivariate Outliers and Truncated Ranges.
1 5 3 16 Reviews
This preview shows pages 1 - 3 of a 9 page document. to view the rest of the content Statistics for the Behavioral Sciences FINAL Study guide  Topics: Chi­square: Chi Squares tests are non­parametric. Non­parametric work with categorical (nominal/ordinal) data ; use frequencies per
category
Make few assumptions about the population distribution Less powerful Chi­square distribution (X 2 ) are  always positively skewed, and their shape depends on degrees of freedom Chi critical value. (df = k ­1 = 2).  Note: as df increases, cv also increases because you’re measuring participants Values that fall inside the cv are saying that the observed and expected frequencies are
relatively similar.
Values that fall past the cv are saying that the observed and expected frequencies are
different.
Goodness of Fit – used to determine how well observed data falls in with expected values. Independent Hypothesis Testing ­ used to determine whether there is a significant association
between the two categorical variables (ex: gender and voting preference)
One Way ANOVA (Analysis of Variance) components:  Used to compare more than 2 means  in a way that reduces Type 1 Error MSbetween shows how far means are spread out from each other.
The variability of group means.
MSwithin – the average of these variances. Shows how far scores are
generically spread out from means. Known as the error term.
The variability of scores around their group means. F Ratio: F = ( MS Between / MS Within )
Degrees of Freedom:
Df bet = k -1 (where k=number of groups) Df w = NT – k  Df total = NT – 1 (NT = total number of participants)   One Group T-Test Components 1. Null (mu = 0) and Alternative (mu ≠ 0) Hypotheses
2. Find Standard Error
3. Deciding on Z or T (use T if N<40)
4. Degrees of freedom = n-1   If t-calc > t-crit, then we reject the null. If t-calc  < t-crit then we fail to reject the null. 5. Confidence Intervals are used to estimate how confident you are that your  results reflect the true population mean scores. Confidence Intervals should
affirm the null. If you rejected the null, you want the null (0) to fall outside our
confidence interval.
Two Group T-Test Components  1. Null (mu1 – mu2  = 0) and Alternative (mu1 – mu2 ≠ 0) Hypotheses
2. Find Variances (SS/df) and Standard Error
a. Note: df= (n1 + n2) - 2 3. Decide on the appropriate test used to calculate t Questions to ask:  a. Are both sample sizes large? (each sample size must be > 100)  Yes – use large sample test for independent means (use z test!)
No – Go to Question # 2
b. Are the sample sizes equal?  Yes – use pooled variances test for equal sample sizes
No – Go to question #3 to check for Homogeneity of Variance [for there to
be HoV, one variance has to be
no more than twice as big as the other  variance.]  c.  Can the population variances be assumed equal?  Yes – use pooled variances test
No – use separate variances t-test
NOTE:  (mu1 – mu2) in the formulas will be replaced with 0  4. If t-calc > t-crit, then we reject the null. If t-calc < t-crit then we fail to reject the  null.  5. Confidence Interval should affirm the null. If you rejected the null, you want  the null (0) to fall outside our confidence interval.   Matched T-Tests  Matched pairs take correlation between 2 items into consideration. (“Items” can
mean either 2 different subjects, or 1 subject tested twice)
The point is to decrease variability between 2 items. Doing so decreases error. Matched T-test PROCESS:
-Let’s say we use an independent t-test, and end up failing to reject our null based
on our results.
-BUT this conclusion can still be wrong – if there was a lot of variability between the
individual items, then this gets in the way of finding the true pattern and making
accurate results.
-So we turn this into a matched-t test (a test combining both items)  -The null: muD = 0

This is the end of the preview. Please to view the rest of the content Join more than 18,000+ college students at New York University who use StudySoup to get ahead
9 Pages 59 Views 47 Unlocks
• Notes, Study Guides, Flashcards + More! Join more than 18,000+ college students at New York University who use StudySoup to get ahead
##### Description: Chi Square, One and Two Way Anovas, Post Hoc Tests, Simple Effects, One and Two Group T-tests, deciding which test-statistic to use, Matched T-tests, Counterbalancing and Designs, Correlation, Interactions, Confidence Interval, Bonferroni, Power, Type 1 and 2 Errors, Coefficients of Det and Non Det, Bivariate Outliers and Truncated Ranges.
9 Pages 59 Views 47 Unlocks
• Notes, Study Guides, Flashcards + More!
Get Full Access to NYU - PSY 001 - Study Guide - Final
×
Get Full Access to NYU - PSY 001 - Study Guide - Final

I don't want to reset my password

Need help? Contact support

Need an Account? Is not associated with an account
We're here to help