Hany Abdelghaffar and Lobna Sameer
of respondent used the internet daily. 96.7 % of the respondent answered that they see the internet an effective tool of communication with the government and with other citizens.
When it comes to the political awareness and political participation of the surveyed participants, it was found that 94 % of the respondents agreed that they used the internet to follow political news and contributing with their opinions. Furthermore, 96 % of respondents agreed on the impact of the social network as a tool of communication between citizens. 72 % of the respondents have participated in both the referendum on the constitutional amendments in 2011 and the parliamentary elections in 2012.
5. Analysis and results
Validity and Reliability
The composite reliability of each construct was assessed using Cronbach’ s alpha. A reliability coefficient of 0.70 is marked as a lowest acceptable limit for Cronbach’ s Alfa( Robinson et al, 1991). The calculated Cronbach’ s Alpha for this research is equal to 0.862.
Convergent validity has been used to check validity which shows that there is a significant correlation and relation among all dimension and sub factors. The correlation was high as shown in table( 3), which is an evidence of a convergent validity. All the independent factors are significant at level 0.05. Discriminant validity is assessed to measure the extent to which constructs are different. To evaluate discriminant validity, the AVE is used. All constructs have an AVE of at least 0.5( Fornell & Larcker, 1981) and all the square roots of each AVE value are higher than the off‐diagonal correlation elements.
The table 3 presents the correlations between the different factor and each other and decision‐making. As a result of the correlation, all factors are at a positive direction meaning that the decision making is affected by each factor.
Table 3: Correlation analysis
Decision Making |
Campaigning |
Information Provision |
Deliberation |
Consultation |
Awareness Building |
Community Building |
Decision Making 1 |
|
|
|
|
|
|
Campaigning 0.580 1
Information Provision 0.396 0.422 1 Deliberation 0.245 0.343 0.526 1 Consultation 0.612 0.524 0.461 0.318 1 Awareness Building 0.657 0.649 0.547 0.463 0.706 1 Community Building 0.569 0.589 0.440 0.302 0.482 0.601 1
Since multicollinearity might exist in regression analysis and negatively affects the predictive ability, computing the variance inflation factor( VIF) of each variable might help to detect multicollinearity( Myers, 1986). If the VIF of an explanatory variable exceeds 10, the variable is considered to be highly collinear and it can be treated as a candidate for exclusion from the regression model( Kleinbaum, et al., 1988). Findings show that VIF range from 1.98 to 2.56 suggesting that multicollinearity is not an issue with this data set.
The multiple regression analysis has used to test the hypotheses. The R calculated through the regression analysis table( 2) is equal to 0.733, R Square 0.537 as presented in table( 4). The regression test notes that there is a positive strong relationship between the independent and the dependent variables. Any change in the independent variable would affect the dependent variable in the same direction and in a certain degree.
7