This preview shows pages 1–2. Sign up to view the full content.
POLI 311: Empirical Methods
Statistics: Nominal level measures of association
Pvalue of Chi square: 0.50 is insignificant
If our variables are related, the way we assumed
covariation
If two variables are perfectly related, they covary perfectly, knowing variables on the
independent level, we would know the values of the dependent variable without any problems
Calculating Lambda
a
Lambda
a
is a prebased measure of association that is appropriate when one or both variables are
measured at the nominal level
It measures how much our predictive ability is improved by knowing the values of cases on the
IV. It ranges in value from 0.00 (no improvement) to 1.00 (perfect predictability)
**Always guess the modal category, if you only know the total size of the sample and you must
ask one person and guess which party they voted for
where: f
i
= the maximum frequency within each category of the IV
f
d
= the maximum frequency in the totals of the DV
N = number of cases
the closer to 0 = the weaker the relationship, the closer to 1 = the stronger the relationship
Lambda
a
will always be zero if the modal value is the same for all categories of the IV
If this happens, then
Cramer’s V
will be an appropriate measure to use for nominallevel
relationships.
Ordinal level measures of association
Ordinal variables: the values of both variables must be listed in their natural order
To interpret a crosstabulation between two ordinal variables: look for a consistent increase in the
percentages in one direction across the top row and the opposite direction across the bottom row
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 Fall '10
 MELANEETHOMAS

Click to edit the document details