Deliverable D 1.3 RESULTS OF META-ANALYSIS ... - cast-eu.org
Deliverable D 1.3 RESULTS OF META-ANALYSIS ... - cast-eu.org
Deliverable D 1.3 RESULTS OF META-ANALYSIS ... - cast-eu.org
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Campaigns and awareness raising strategies in traffic safety — <strong>Deliverable</strong> D-1.1missing cases. These were [emotional and rational], [non-shockingconsequences], [social norm], [target group identified] and [risk of harm].Although we considered these important, there simply was not enoughinformation across the evaluation studies to score these campaign variablessufficiently.The eight remaining variables met statistical recommendations for the ratio ofvariables studied to outcome effects in the sample (Tabachnikk & Fidell,2007).A weighted, simultaneous standard multiple regression was therefore runusing the variables [duration-0 to 29 days], [decade-after 2000], [theme-drinkdriving],[scale-national], [onroad], [personal influence], [enforcement] and[combined mass-media], using ln odds ratios calculated using accident effects(see chapter 7). An inspection of the contributions to outcome variance thenled to the exclusion of the variable [scale-national] to give a final model ofseven variables (Table 8.2).Table 8.2. Results from standard multivariate regression analysis of the final accident model.Variables, their standardised (beta) coefficients, t-values and probabilities are shown.VariableBetacoefficientt-value p-value Semi-partialcorrelation(constant) -- 1,18 .241 --[duration - 0 to 29 days] -.21 2.06 .0042 -.17[after 2000] +.21 4.27