Class Notes (835,067)
Sociology (1,100)
SOCY 211 (69)
Carl Keane (12)
Lecture

# SOCY211 Week 8, Lecture 2

7 Pages
49 Views

School
Department
Sociology
Course
SOCY 211
Professor
Carl Keane
Semester
Winter

Description
ANALYSIS OF VARIANCE - Often referred to as a “Comparison of Means” test - Can be used when: Independent variable is measured at the nominal or ordinal level, and the Dependent Variable is measured at the interval/ratio level. - ANOVA - A lot of our independent variables vary- such as happiness - “Total Variation” = two components: (1) the distance, or deviation, of raw scores from their group mean - this is known as “Variation Within Groups,” and, (2) the distance, or deviation, of group means from one another - this is known as “Variation Between Groups.” - Education and income example - All have some overlap and within each group there will be some variations - When you compare between groups you always use the mean  Sum of Squares S =√ ∑(X – X)² N - We use all the values so that it is easier to compare - When we have groups we have different sums of squares but each type represents the sum of sum deviations from the mean  Total Sum of Squares  Within-Groups Sum of Squares  Between Groups Sum of Squares Happiness Measured on a Scale from 1 to 10 (example) Group A Group B Group C 8 x x 7 x x 5 x x 4 x x 24 X X N(A)=4 N(B)=4 N(C)=4 Total N=12 Mean(A) = 24/4 = 6 Total Mean (Hypothetically) = 5 - WITHIN-GROUP VARIATION/DEVIATION (Raw scores for each group member minus their group mean) e.g., Group A 8 – 6 = +2 1 7 – 6 = +1 5 – 6 = -1 4 – 6 = -2 - Then do the same for individuals in Groups B and C – subtracting their individual scores from their group means. - BETWEEN-GROUP VARIATION/DEVIATION (Means for each group minus the total mean) e.g., Group A: 6 – 5 = +1 - Then do the same for Groups B and C - TOTAL VARIATION/DEVIATION (Raw scores for each group member minus the total mean) e.g., Group A 8 – 5 = +3 7 – 5 = +2 5 – 5 = 0 4 – 5 = -1 - Then do the same for every member of Groups B and C. Mean Squares  Total Sum of Squares - measures variation of individual scores about the total mean  Within-Groups Sum of Squares - measures variation of individual scores about their group mean  Between-Groups Sum of Squares - measures variation of group means about the total mean F RATIO Between-Groups Mean Squares F = -------------------------------------- Within-Groups Mean Squares - Shows if our data is statistically significant - The stronger the relationship between the two variables, the larger the value of f (independent and dependent) - If there is a big difference between the variance then there is a relationship - If there is a weak association, the between group variance will be small SUMMARY Ratio of: Relationship B-G Variance/Mean Squares W-G Variance/Mean Squares Large B-G Mean Squares -------------------------- = Large F Ratio Strong Small W-G Mean Squares 2 Small B-G Mean Squares ------------------------- = Small F Ratio Weak Large W-G Mean Squares Medium B-G Mean Squares --------------------------- Moderate F Ratio Moderate Medium W-G Mean Squares Total Sum Within-Groups Between Groups = + of Squares Sum of Squares Sum of Squares So, Between-Groups Total Sum Within-Groups = - Sum of Squares of Squares Sum of Squares 2 - PRE measures of association with ANOVA is called ETA square (E ) - Makes use of the total sum of squares and group totals - Total sum of squares is total variation in dependent variable - Total sum of square is a measure of errors
More Less

Related notes for SOCY 211
Me

OR

Join OneClass

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

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.