Un modèle de benefices des données perso mai 2014

1,594 views

Published on

Lire aussi l'analyse sur http://christophe.benavent.free.fr/?p=1063

Published in: Marketing
  • Be the first to comment

  • Be the first to like this

Un modèle de benefices des données perso mai 2014

  1. 1. Données personnelles et bénéfices de la personnalisation : le pivot de la confiance Données MesInfos.fr Vague 1 Mai 2014 Christophe @benavent
  2. 2. L'échelle du privacy concern
  3. 3. Modèle hiérarchique et bifactoriel
  4. 4. Omega Call: omega(m = PC, nfactors = 5, fm = "minres", sl = Bifactorial Alpha: 0.955 G.6: 0.972 Omega Hierarchical: 0.837 Omega H asymptotic: 0.857 Omega Total 0.977 Schmid Leiman Factor loadings greater than 0.2 g F1* F2* F3* F4* F5* h2 u2 Q5_01 0.664 0.518 0.719 0.281 Q5_03 0.773 0.379 0.764 0.236 Q5_04 0.802 0.524 0.921 0.079 Q5_05 0.741 0.318 0.268 0.731 0.269 Q5_06 0.806 0.380 0.826 0.174 Q5_07 0.823 0.494 0.926 0.074 Q5_09 0.616 0.629 0.776 0.224 Q5_10 0.674 0.628 0.858 0.142 Q5_11 0.644 0.722 0.936 0.064 Q5_13 0.534 0.491 0.548 0.452 Q5_14 0.708 0.266 0.447 0.781 0.219 Q5_15 0.719 0.202 0.621 0.379 Q5_16 0.704 0.345 0.640 0.360 Q5_18 0.687 0.418 0.700 0.300 Q5_19 0.719 0.607 0.893 0.107 Q5_20 0.715 0.464 0.737 0.263 With eigenvalues of: g F1* F2* F3* F4* F5* 8.103 1.541 0.940 0.737 0.505 0.551 Explained Common Variance of the general factor = 0.655 The degrees of freedom are 50 and the fit is 0.28 The number of observations was 320 with Chi Square = 86.644 with prob < 0.001 The root mean square of the residuals is 0.012 RMSEA index = 0.0496 and the 90 % confidence intervals are 0.0303 0.0645 BIC = -201.772
  5. 5. Les variables étudiées
  6. 6. Risque perçu Influence sociale Bénéfices perçus Efficacité personnelle Confiance Innovativité 0.297 0.236 Privacy Concern -0.176 0.129 0.158 0.222 0.223 0.136 0.353 Modèle Path Analysis ( With Lavan) Estimator ML Minimum Function Test Statistic 13.211 Degrees of freedom 5 P-value (Chi-square) 0.021 RMSEA 0.068 90 Percent Confidence Interval 0.025 0.112 P-value RMSEA <= 0.05 0.208 0.356
  7. 7. Régression modérée
  8. 8. Coefficients LinearModel.1: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.04162 0.50688 2.055 0.040710 * Data_Social 0.19431 0.06235 3.116 0.002001 ** Data_Innov 0.13693 0.05752 2.380 0.017887 * Data_Risks 0.05715 0.05013 1.140 0.255130 Data_Trust 0.30837 0.05239 5.886 1.01e-08 *** Data_SE 0.21945 0.05698 3.851 0.000142 *** Residual standard error: 1.66 on 314 degrees of freedom Multiple R-squared: 0.3368, Adjusted R-squared: 0.3262 F-statistic: 31.89 on 5 and 314 DF, p-value: < 2.2e-16 Coefficients LinearModel.2: Estimate Std. Error t value Pr(>|t|) (Intercept) 3.037968 0.930174 3.266 0.00121 ** PrivacyConcern 0.006417 0.016486 0.389 0.69736 Data_Social 0.177317 0.062124 2.854 0.00460 ** Data_Innov 0.125172 0.057600 2.173 0.03052 * Data_Risks 0.056534 0.062747 0.901 0.36830 Data_Trust -0.095130 0.157198 -0.605 0.54551 Data_SE -0.092535 0.125477 -0.737 0.46139 Data_Trust:Data_SE 0.059887 0.021814 2.745 0.00639 ** Residual standard error: 1.645 on 312 degrees of freedom Multiple R-squared: 0.3529, Adjusted R-squared: 0.3384 F-statistic: 24.31 on 7 and 312 DF, p-value: < 2.2e-16 anova(LinearModel.1, LinearModel.2) Res.Df RSS Df Sum of Sq F Pr(>F) 1 314 865.36 2 312 844.31 2 21.043 3.8879 0.02149 * Test de la modération

×