The description of factor analysis Essay

Factor analysis extracted one component with an eigenvalue of higher than one. The component (including 5 items) explained 61.57%. According to varimax loading the factor defined subjective norm. The table 7 shows factors analysis on subjective norm.
Factor analysis extracted three components with an eigenvalue of higher than one. The first component (including 7 items), second component (including 5 items) and third component (including 3 items) explained 28.31%, 16.64% and 15.86% respectively. According to varimax loading, the first, second and third factors defined hobby, information and transaction respectively. The table 8 shows factors analysis on internet abuse.
Descriptive statistics gained from SPSS revealed that the mean and standard deviation for internet addiction, job satisfaction, productivity, policy, perceived need, subjective norms, hobby, information and transaction explained 2.347, 3.513, 3.658, 3.298, 3.317, 2.506, 2.408, 2.932, 2.441 and 0.734, 0.632, 0.755, 0.575, 0.640, 0.811, 0.850, 0.792, and 0.972 for standard deviation. The standard deviation indicates dispersion of data from mean. In this case, the dispersion data from mean is considerably low.
The analysis indicated a high correlation between productivity and hobby (r = -0.527), followed internet addiction and hobby (r = 0.460), subjective norm and information (r = 0.393) lastly internet addiction and transaction (r = 0.332). The effect size of this correlation can be classified as medium to large on Cohen's (1988) guidelines.
All items for each construct were tested for reliability. The reliability is measured using Cronbach's Alpha. The Cronbach's Alpha for all variables this study is over 0.5 which is above the required threshold value. Thus, questionnaire used in this study is reliable. The table 9 shows descriptive statistics, correlation and reliability analysis.
Adjusted R square which means that only 42.3 percent,16.9 percent and 17.6 percent of the variance in independent variable can be predicted from the hobby, information and transaction (internet abuse) respectively.
The F-test for first (hobby) regression model was significant (F= 24.07, p <0.001) with four factors of internet abuse entered the resulting equation internet addiction significant at 5% level ( = 0.181, p < 0.05), job satisfaction highly significant at 1% level ( = -0.327, p<0.001), productivity highly significant at 1% level ( = -0.401, p < 0.001) and perceived need significant at 10% level ( = 0.202, p<0.01). The F-test for second (information) regression model was significant (F=13.822, p< 0.001) with two factor of internet abuse entered the resulting equation internet addiction significant at 5% level ( = 0.176, p < 0.05) and subjective norm highly significant at 1% level ( = 3.906, p<0.001). The F-test for third (transaction) regression model was significant (F= 9.875, p <0.001) with three factors of internet abuse entered the resulting equation internet addiction highly significant at 1% level ( = 0.334, p < 0.001), job satisfaction significant at 5% level ( = -0.179, p<0.05) and perceived need significant at 10% level ( = 0.236, p < 0.001). There is no relationship between workplace policy with any indicator for internet media abuse. The table 10 shows multiple regression result.