Since the overall multivariate test indicated a significant multivariate main effect for diabetes (Wilks' Lambda =.96, F = 2.82, df = 8, 535, p =.005), univariate probes were conducted across the eight SF-36 domains (see Table 3).
Multivariate tests indicate statistical values across all components.
The multivariate Lagrange Multiplier test indicated a significant omitted relationship between Task Persistence and Coping Self-Statements.
Maulchy's test indicated that the assumption of sphericity had been violated: χ (2)=20.56, p<0.001; therefore, multivariate tests are reported (ɛ=0.93).
Although multivariate analysis failed to show CFR < 2.5 as an independent predictor of cardiac hard events (Table 4), a chi-square test (Fisher's exact probability test) indicated that CFR < 2.5 was significantly related to cardiac hard events (p < 0.01).
Statistical test indicated.
The multivariate tests of significance indicated that there was a statistically significant multivariate effect for success levels (F = 4.118, p = .021, η2 =.126) though its contribution to the model was very modest (Pillai's Trace = .126, Wilks Lambda = .874).874
The multivariate tests of significance indicated that there was a statistically significant multivariate effect for success levels (F = 8.283, p = .000, η2 =.307) though its contribution to the model is very modest (Pillai's Trace = .307, Wilks Lambda = .693).693
Both multivariate logistic regression analysis and z-tests indicated that the occurrence of EPI was closely associated with shock, sepsis, diabetes, cardiac arrest, hyperlactacidemia, invasive mechanical ventilation and haemodialysis.
A multivariate test of significance including all three variables resulted in F = 21.0, P < 0.001, indicating significant overall reduction in hyperglycemia, which was consistent with the observed reduction in HbA1c.
