Download Stats packet for important methods and more Cheat Sheet Economic statistics in PDF only on Docsity!
SLIDESMANIA
- Fall Packet - Z-ScoreโโโโโโโโโโโโโโโโโโโโโโโโโโSlide Table of contents - Z-TestโโโโโโโโโโโโโโโโโโโโโโโโโโโSlide - T-Test (Independent)โโโโโโโโโโโโโโโSlide - T-Test (Repeated) โโโโโโโโโโโโโโโโโSlide
- ANOVA (Independent)โโโโโโโโโโโโโโSlide - ANOVA (Repeated)โโโโโโโโโโโโโโโโSlide - ANOVA (Factorial)โโโโโโโโโโโโโโโโโSlide - CorrelationโโโโโโโโโโโโโโโโโโโโโโโSlide
- RegressionโโโโโโโโโโโโโโโโโโโโโโโโSlide
- Chi-Squared (Independence)โโโโโโโโโโSlide
SLIDESMANIA
Z score
When it's used: when a researcher wants to determine how far away a value is from from the sample mean (in standard deviations) and knows the population parameters (standard deviation (SD) and mean) Knowledge Claim: Tells us how many SDs away from the mean a data point is or the percentage of scores that fall between the z-score and the mean. Statistical notation: โ (^) z= obtained value โ (^) ฯ= standard deviation โ (^) ฮผ= population mean โ (^) p > or < 0.
SLIDESMANIA
Independent t test
When it's used: when a researcher wants to compare mean differences between two groups that are independent of each other. Knowledge Claim: Allows for rejection or failure of rejection of the null hypothesis by telling if a two samples have similar means. Must also complete Cohenโs D test to determine if the sample has a large effect size. Statistical notation: โ (^) Xฬ
= group means โ (^) ss= sum of squares โ (^) SD or s= standard deviation โ (^) df=(n-2)
SLIDESMANIA
T test- repeated
When it's used: when a researcher wants to test the same participants twice to assess how changes to the experimental design will affect their responses. Knowledge Claim: each participant must be measured twice. The independent variable measures the participant in a โbeforeโ and โafterโ state. Statistical notation: โ (^) D bar= difference means โ (^) ss= sum of squares โ (^) SD or s= standard deviation โ (^) df=(n-1)
SLIDESMANIA
ANOVA repeated
When it's used: when a researcher wants to compare results from one group either taken at different times or under different circumstances. Knowledge Claim: allows for the comparison of data from related but not independent groups. This test requires one independent and one dependent variable. It utilizes t-tests Statistical notation: โ Xฬ
or (p bar)= row or column means โ (^) G bar= grand mean โ (^) k= levels โ b/t df= k-1 w/in df=N-k โ (^) ss= sum of squares โ (^) MS= mean of squares (ss/df)
SLIDESMANIA
ANOVA Factorial
Main Effects Vs Interaction The main effect is the effect of just one of the independent variables on the dependent variable, while the interaction effect is the effect that considers both independent variables. Levels vs Factors The factor of the experiment is the independent variable. Levels are within the factors. If the independent variable is temperature, different levels would be 0 degrees, 20 degrees and 100 degrees (three levels). When to use each type of ANOVA: โ (^) ANOVA: comparing two or more means โ (^) ANCOVA: comparing two or more covariances- โ (^) MANOVA: comparing two or more dependent variables
SLIDESMANIA
Regression (simple)
When it's used: when a researcher wants to compare more than two variables. It can be used to assess correlation, why an event happens, or to measure the impact Knowledge Claim: Allows for the researcher to compare two or more variables. The field of study impacts which predictors are used in the study. Should avoid studies where two variables are highly correlated (multicollinearity)
SLIDESMANIA
Chi squared- goodness of fit
When it's used: when a researcher wants to compare the distribution of a sample to the distribution that would be expected by chance. Knowledge Claim: Allows for rejection or failure of rejection of the null hypothesis by telling if a the observed distribution of a sample is what would be expected by chance. Statistical notation: โ (^) xยฒ is the test statistic/obtained value โ The obtained value is compared to the critical value and the alpha level to determine if the null hypothesis occurs merely by chance (not significantly significant) โ Levels: the different types of the dependent variable tested (K) โ
