regression models for ordinal responses a review of methods

Constraints 0 2 and 1 2 1 3 4 7 4 1330 international

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ation 4° α1 α2 α3 α4 β γ2 γ3 γ4 Likelihood ratio test Midline episiotomy α1 α2 α3 α4 β γ Likelihood ratio test Midline episiotomy χ2 P-value –2.8431 ± 0.0336 –3.0543 ± 0.0370 –3.1972 ± 0.0395 –5.1914 ± 0.1034 0.7125 ± 0.0946 0.1471 ± 0.0278 0.2225 ± 0.0393 0.9394 ± 0.1738 56.7 28.0 32.1 29.1 0.0001 0.0001 0.0001 0.0001 89.8 116.0 0.0001 0.0001 57.3 46.6 0.0001 0.0001 85.6 108.1 0.0001 0.0001 χ2 4 χ2 4 b. Constrained modelb Intercept1 Intercept2 Intercept3 Intercept4 Midline episiotomy Constraint parameter Estimate ± SE χ2 2 χ2 2 a –2.8432 ± 0.0336 –3.0455 ± 0.0363 –3.2026 ± 0.0393 –5.1510 ± 0.0939 0.7160 ± 0.0946 0.1126 ± 0.0165 Response variable is degree of laceration: no laceration, 1°, 2°, 3°, and 4°. Model (a): log Pr (Y yj ) = αj – β(Episiotomy) – γ2(Episiotomy:Lacr 2°) – Pr (Y yj ) γ3(Episiotomy:Lacr 3°) – γ4(Episiotomy:Lacr Model (b): log Pr (Y yj ) Pr (Y yj ) = αj – β(Episiotomy) – γ j (Episiotomy) . Constraints: = 0, = 2, and 1 2 = 1...
View Full Document

Ask a homework question - tutors are online