Reference no: EM132852587

Student's t test was used univariately to compare continuous data between two groups. To evaluate the relative importance of multiple factors for stroke severity, outcome, and mortality, multiple linear regression and multiple logistic regression analyses were performed. In multiple linear regression all variables of interest were tested backwards, to fit the full model with all the potential explanatory variables. Unimportant variables were then removed one at a time until all those remaining in the model contributed with p values <0.20. The two methods of multiple regression analysis, backward and forward regression, often yield the same model, but differences are not uncommon. Neither approach is more correct than the other. To select the "best model", backward linear regression including only variables with a p value <0.2 was followed by forward linear regression including the same variables. The two methods yielded similar models. The explanatory power of the resulting equations was determined by the adjusted R2 statistic. Assumption of linearity was checked by plotting standardised residuals against standardised predicted values as well as against independent variables, and by partial regression plots. The distribution of residuals was examined by histograms for standardised residuals and by normal probability plots. In logistic regression similar procedures to the ones described above for linear regression were conducted by backward and forward Wald analyses. The explanatory power of the resulting equations was determined by the value of correct classification. The required two-tailed significance level for all tests was set at 0.05.

The sentence that begins with "student's t-test was used..." Rewrite the entire paragraph providing a clear explanation of what was done statistically and WHY.