Chi square ppt
Download as PPT, PDF1 like548 views
This document discusses requirements and tests for association between variables, including that bivariate, non-parametric tests require a sample size of at least 50 with 5 frequencies in each cell. It presents a test of independence between car model appeal and age groups, and discusses goodness of fit tests to see if movies or accidents are uniformly distributed. It also covers contingency coefficients, phi-correlations, and lambda coefficients as measures of strength of association between two variables.
Chi square ppt
- 2. Requirements • Bivariate • Non-Parametric • Sample Size should be 50 • 5 frequency in one cell
- 3. Test of Independent • Ho – The car model appeal is independent of the age groups Below 20 20-39 40 and above Like the car 140 80 40 Dislike the car 60 50 30
- 4. Test of Goodness of Fit • The movies appeal equally to all age groups. • The accidents are uniformly distributed over the week. • There is no association between economic condition at home and IQ.
- 5. Strength of Association Between Two Variables or Indexes of Agreement Contingency Coefficient • 0 to 1 value • (2/2) / (3/3) / (4/4)
- 6. Strength of Association Between Two Variables or Indexes of Agreement Phi-Correlation • 0 to 1 value • (2/2)
- 7. Strength of Association Between Two Variables or Indexes of Agreement Lambda Coefficient • Chances of reduction in error in estimating one variable to another variable • E.g. – 30%