Class Notes (834,511)
Canada (508,560)
Sociology (1,100)
SOCY 211 (69)
Carl Keane (12)

SOCY211 Week 7, Lecture 2

3 Pages
Unlock Document

SOCY 211
Carl Keane

Nominal measures of association for tabular data - Continuation of Tuesdays’ lecture: - Builds on material from before the break - Looking at a table will do a couple of things: you can compare across columns of the independent variable for changes - Does our dependent variable vary because of our independent variable - Chi Square- Allows us to generalize that the association found in the sample will also be present in the greater population - Measures of association for nominal level variables from 0 to 1.0 - You need to be able to verify which variable is dependent and which one is independent - PRE (proportional reduction in error) is a very important concept!!- tell us how much we are able to minimize errors MEASURES OF ASSOCIATION FOR ORDINAL VARIABLES - These measures are to be used for tabular data when both variables are measured at the ordinal level. - GAMMA (Goodman and Kruskal’s Gamma)- has a wider range, can range from +1.0 to -1.0 - The closer it gets to 1.0 the stronger it is - A positive gamma represents a positive association - Gamma is also a symmetrical measure- no distinction is made between the dependent and the independent variable - Gamma is a very popular measure of association, but it has its limitations - As the number of variable categories is reduced, the value of gamma will increase - Value of gamma is inflated because of the reduction in categories - Proportional Reduction in Error (PRE) measure (of association) - Somer’s d- is an asymmetric measure of association - When looking at the output you have to make sure you put the right value that refers to Somer’s d - Goes from -1.0 to 1.0 - The formula for Somer’s d differs from gamma and the value (for Somer’s d) will always be smaller than the value we get for gamma - You need to know how to interpret the statistical output that you get when you use gamma or Somer’s d - EXAMPLE: research question- relationship between people’s mental health and their physical health - Assuming that mental health will affect physical health - Multivariate Tabular Analysis (should have been at the bottom of the page**) - Elaboration- involves identifying those variables that are able to explain away the original relationship that was found - Control variables- variables that may explain that original relationship 1 State of health * Take medication for depression Crosstabulation Take medication for Total depression Yes No State of Excellent? Count 6 443 449 health % within Take 8.3% 30.6% 29.5% medication for depression Very good? Count 22 513 535 % within Take 30.6% 35.4% 35.2% medication for depression Good? Count 17 375 392 % within Take 23.6% 25.9% 25.8% medication for depression
More Less

Related notes for SOCY 211

Log In


Join OneClass

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

Sign up

Join to view


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.