13th European Conference on eGovernment – ECEG 2013 1 | Page 388

Fattah Nazem and Anahita Madankar

eliminated except eleven subjects. Moreover, the given situation shows that there is no need to omit some of the items and it is possible to follow the process of factor analysis while having all the items. The second factor analysis assumption denotes enough sample size. In this research, Kaiser‐ Meyer‐ Olkin( KMO) equals 0.97; consequently, the sample size is sufficient. The third factor analysis assumption is the normality of multivariation distribution known as sphericity. As the approximate chi square equalled 27100.666 with the 861 degrees of freedom, it can be stated that the amount of the approximate chi square is statistically significant and the given statistics is significant at least at the 0.999 level of confidence( α = 0.001).

According to component matrix of items, we can determine both the specific factor of each item and its position in the related factor based on loading factor. Subsequent to studying the table of component matrix precisely, the researcher used rotation method so that loading factor of each item can be determined stressing at recognition of each item in one of the 10 factors. Reiterating that in this research, the researchers have followed exploratory factor analysis, and used principal component methods from extraction of factors, varimax method was applied( Table 1). According to varimax, the researchers were able to determine both the factor to which the item belongs after rotation, and the position of each item in the related factor with reference to loading factor. This table shows each item has been located in which factor after the rotation. For instance, Items 24, 25, 26, 27, 28, 29, and 30 have been located in the first factor( strategy). To fulfill the purposes of the study to determine the indexes of knowledge management and its components, the underlying items, and the index with the highest contribution, eventually, 8 factors have been extracted from rotation of factor analysis; in fact, knowledge management consists of 8 factors respectively as follows: strategy, learning organization, team work and learning communities, sharing knowledge, organizational culture, digital sophistication, intellectual capital and leadership and management. The table also indicates that strategy has the highest level of contribution to the formation of knowledge management at universities. The reason is that, as the first column, that is, strategy factor shows, 7 items with more than 0.5 have been located in this column.

Table 1: Rotated component matrix factors