chapter 13
DESCRIPTION
Chapter 13. Association Between Variables Measured at the Nominal Level. Chapter Outline. Introduction Chi Square-Based Measures of Association Proportional Reduction in Error (PRE). Chapter Outline. A PRE Measure for Nominal-Level Variables: Lambda The Computation of Lambda - PowerPoint PPT PresentationTRANSCRIPT
Chapter 13
Association Between Variables Measured at the Nominal Level
Chapter Outline
Introduction Chi Square-Based Measures of
Association Proportional Reduction in Error (PRE)
Chapter Outline
A PRE Measure for Nominal-Level Variables: Lambda
The Computation of Lambda The Limitations of Lambda
Nominal Level Measures of Association
It is always useful to compute column percentages for bivariate tables.
But, it is also useful to have a summary measure – a single number – to indicate the strength of the relationship.
Nominal Level Measures of Association
For nominal level variables, there are two commonly used measures of association: Phi or Cramer’s V Lambda
Nominal Measures: Phi Phi is used for 2x2 tables. The formula for Phi:
Nominal Measures: Cramer’s V Cramer’s V is used for tables larger than
2x2. Formula for Cramer’s V:
Nominal Measures: Phi The phi for Problem
12.1 is 0.33. This is a strong
association.
Value Strength
Between 0.0 and
0.10Weak
Between 0.10 and
0.30Moderate
Greater than 0.30
Strong
Limitations of Phi
Phi is used for 2x2 tables only. For larger tables, use V.
Phi (or V) is an index of the strength of the relationship only. It does not identify the pattern.
To analyze the pattern of the relationship, see the column %s in the bivariate table.
Nominal Measures: Lambda Like Phi, Lambda is used to measure the
strength of the relationship between nominal variables in bivariate tables.
Unlike Phi, Lambda is a PRE measure and its value has a more direct interpretation.
While Phi is only an index of strength, the value of Lambda tells us the improvement in predicting Y while taking X into account.
Association and Bivariate Tables To compute λ, we must first find E1 and E2:
E1 = N – largest row total = 44 – 22 = 22
E2 = For each column, subtract the largest cell
frequency from the col. total = (27 – 17) + (17 – 12) = 10 + 5 = 15
Low High
Low 10 12 22
High 17 5 22
27 17 44
Nominal Measures: Lambda Formula for Lambda:
Nominal Measures: Lambda
Lambda is a PRE measure. A Lambda of .32 means that
authoritarianism (X) increases our ability to predict efficiency (Y) by 32%.
The Limitations of Lambda Lambda gives an indication of the strength
of the relationship only. It does not give information about pattern.
To analyze the pattern of the relationship, use the column %s in the bivariate table.
When row totals are very unequal, lambda can be zero even when there is an association between the variables.