R ESULTS AND D ISCUSSIONS

This study uses factor analysis, regression analysis, and path analysis to verify the research framework to understand the predictive relationship among all the variables. For the analysis of statistical software, personality trait anxiety and information literacy are used as independent variables, mobile computer self-efficacy and mobile computer anxiety are used as intermediate variables, and use intention of mobile computer in teaching is used as the dependent variable. The hypotheses and paths of the research framework are shown in Figure 3-1. The stepwise regression method is used to conduct a path analysis on each variable respectively. The stepwise regression analysis is used to determine the statistical significance and explanatory power. The purpose of path analysis is to explain the predictive model existing among all the variables. The value of path coefficient can be used to find out the influence or effect among variables.

The results of path analysis are shown in Table 5-1. The results of the multiple regression analysis on the influence of personality trait anxiety and information literacy on MCE indicate that there is a statistically significant difference between the variables (p<0.01), and the

2

literacy can explain 50.4% of the variance in mobile computer self-efficacy. It can be inferred that information literacy is an important factor affecting the mobile computer self-efficacy. As for the result of multiple regression analysis, there is an antagonistic effect between personality trait anxiety and information literacy, which reduces the value of β coefficient.

Such an antagonistic effect is also generated on the influence of these two variables on mobile computer anxiety. The finding of this study is consistent with Delcourt (1993) and Heinssen (1987), which suggest that the perceived computer self-efficacy is correlated with personality trait anxiety and behavior confidence, and Wu (2006) on junior high school teachers outside Taiwan, which suggests that the information literacy is significantly positively correlated with the computer self-efficacy.

In addition, the personality trait anxiety of elementary school teachers and their information literacy also have a statistically significant influence on mobile computer anxiety (p<0.01) and the values of Adjusted R² are 36.3% and 28.5%, respectively. Personality trait anxiety is positively correlated with mobile computer anxiety. The research results of Thatcher and Perrewe (2002) verify the reliability of this study. There is a significant difference between mobile computer anxiety and mobile computer self-efficacy, and there is a negative correlation between them. Higher mobile computer self-efficacy indicates less possibility of the generation of mobile computer anxiety. Mobile computer self-efficacy can

explain 40.1% of the variance in mobile computer anxiety. The relationship between two variables, mobile computer self-efficacy and mobile computer anxiety, and use intention of mobile computer in teaching is investigated. Table 5-1 shows that there is a significant positive correlation between mobile computer self-efficacy and use intention of mobile computer in teaching. And then, the statistical explanatory power is 49.5%. There is a significant negative correlation between mobile computer anxiety and use intention of mobile computer in teaching and the statistical explanatory power is 50.4%. Chuang (2007) studies the influence of elementary school administrative support teachers’ computer self-efficacy on the integration of information into teaching, and finds that there is a significant positive correlation between them. Higher mobile computer self-efficacy indicates higher possibility of the generation of a significant positive correlation between it and elementary school teachers’ use intention of mobile computers in teaching.

The results of hypothesis testing are shown in Table 5-2. The results of the testing support seven hypotheses, which can verify the rationality and correctness of the hypotheses in this study. Among them, a positive correlation is found in H1b, H2a, and H4a, suggesting that the person of higher personality trait anxiety has a higher perceived mobile computer anxiety, a person of high information literacy has a higher perceived mobile computer self-efficacy, and a person of higher perceived mobile computer self-efficacy have a higher use intention of

mobile computer in teaching. The path coefficients of H1b, H2a, H4a were 0.28, 0.70, and 0.62, respectively. A negative correlation is found in other hypotheses. In terms of the value of path coefficients, that between information literacy and mobile computer self-efficacy and that between mobile computer self-efficacy and use intention of mobile computer in teaching are higher, suggesting that the degree of predictive relationship is higher.

Table 5-1 Summary of predictive regression analysis Dependent

variable

Independent variable β Adjusted R²

p Sig.

Personality trait anxiety -0.107 0.514 0.000 support Mobile computer

self-efficacy Information literacy 0.697 0.504 0.003 support Personality trait anxiety 0.284 0.363 0.000 support Mobile computer

anxiety Information literacy -0.497 0.285 0.000 support

Mobile computer

Table 5-2 Summary of Research Assumption

Research assumption Sig.

H1a: The effect of elementary school teacher's personality trait anxiety on “mobile computer self-efficacy” is significant negative correlation.

support

H1b: The effect of elementary school teacher's information literacy on “mobile computer self-efficacy” is significant positive correlation. .

support

H2a: The effects of elementary school teacher's personality trait anxiety on “mobile computer anxiety” are significant positive correlation.

support

H2b: The effects of elementary school teacher's information literacy on “mobile computer anxiety” are significant negative correlation.

support

H3: The consciousness of mobile computer self-efficacy will negatively affect the consciousness of mobile computer anxiety for elementary school teachers.

support

H4a: “ mobile computer self-efficacy” generates positive effect on “use intention of applying mobile computer in teaching”.

support

H4b:“mobile computer anxiety” generates negative effect on

“use intention of applying mobile computer in teaching”.

support

The method of PCA is used to conduct the factor analysis, extract the common variance among all the variables, and find the factors of an eigenvalue greater than 1. The results of factor analysis are shown in Table 4-5. The most important factor extracted is mobile

computer anxiety, and its eigenvalue is 5.06 and explained variance is 18.07%, suggesting that the key determinant of this study is mobile computer anxiety, follows by information literacy and use intention of mobile computer in teaching. The top three factors account for 74.73% of the explained variance.