An Empirical study comparing the learning environments of open and closed computer laboratories

Journal of Information Systems Education, 2002 by Newby, Michael

4.4 Achievement Achievement was measured as the grade obtained in the course, as a mark out of 100. This grade was contributed to by three components, a final examination, assignments and laboratory exercises. Both the examination and the assignments tested knowledge and skills that should have been gained mainly in the laboratory classes, whose main purpose was to give practical experience of material covered in the lectures. Using means and standard deviations obtained for each course, each grade was converted into a z-score. Of the 104 students from Curtin, 77 provided their student number and of the 109 students from Fullerton, 74 did so. This allowed the grades of these students to be determined.

4.5 Research Questions Closed computer laboratory classes require more resources and greater commitment than open laboratories and this provides the rationale for Research Question # 1:

Do students who receive their laboratory experience via open computer laboratories perceive their learning environment differently from those who receive their laboratory experience via closed computer laboratories? Previous research shows an association between classroom environment and student outcomes (Fraser, 1991) and this formed the focus of Research Question #2:

Are the course outcomes in terms of attitude and achievement different for students who receive their laboratory experience via open computer laboratories from those who receive it via closed computer laboratories? 5.RESULTS Table 4 shows the alpha reliabilities and mean correlations with other scales for the scales of the CLEI for both samples. The reliabilities for the Australian sample vary from 0.56 to 0.89, and for the USA sample from 0.61 to 0.80. These are consistent with previous studies and indicate that the reliabilities of the scales are satisfactory.

The mean correlations with other scales vary from 0.08 to 0.23 for the Australian sample and 0.06 to 0.24 for the US sample. These demonstrate that there is little overlap in what the scales are measuring and the results are consistent with previous studies in which factor analysis was used to confirm that there are five distinct scales (Newby, 1999). Table 5 shows the alpha reliabilities and mean correlations with other scales for the scales of the ACCC. The alpha reliabilities vary from 0.64 to 0.90 for the Australian sample and from 0.72 to 0.89, indicating that the scales have a satisfactory internal consistency for these samples. The mean correlations show that the scales measure distinct but overlapping aspects of students' attitudes towards computers and the course. Factor analysis has been used in a previous study to confirm a structure of four factors (Newby, 1998). An independent samples t-test was carried out on the environment variables, the attitudinal variables and on achievement measured by the z-score, using country of study as the grouping variable. The results for the environment variables are given in Table 6, and for the attitudinal variables and achievement in Table 7.