Internet Learning Volume 4, Number 1, Spring 2015 | Page 65

Internet Learning Journal – Volume 4, Issue 1 – Spring 2015
two additional questions in terms of e-learning experiences indicated that most participants
had previously taken five to six online courses, and were taking one or two online courses at
the time of the questionnaire’s administration..
Data Analysis Procedures
An independent sample student’s t-test was the statistical technique used to test each
of the research hypotheses. All of the research hypotheses are directional--that is, the
independent variable has a positive effect on each of the dependent variables. Therefore, the
hypothesis testing is one-tailed rather than two-tailed by default. First, a reliability analysis
was conducted to assess the internal consistency of item responses for each dependent
variable. If response reliability is acceptable, the index score of each variable will be
calculated for subsequent t-tests. Also, descriptive statistics including means and standard
deviations from descriptive analyses were reported.
RESULTS
Presented in Table 1 are the results of reliability and descriptive analyses.
Chronbach’s Alpha was the reliability coefficient used in this study to measure the internal
consistency of responses to the items measuring the same variable. If an alpha was greater
than 0.7, this indicated that the measurement error of the items was controlled. In this study,
the measures of all dependent variables exhibited acceptable levels of reliability as they
were around 0.8 or higher. The relatively high reliability of item responses supported the
calculation of index scores for each dependent variable based on the average of item scores.
Table 1 Reliability and Descriptive Analyses
Dependent Variable Alpha Treatment Group Control Group
Learning Objectives 0.87 4.49 (0.54) 4.31 (0.84)
Outcome Assessments 0.85 4.52 (0.64) 4.32 (0.88)
Instructional Materials 0.92 4.31 (0.76) 4.10 (1.02)
Learner Interactions 0.79 4.34 (0.73) 3.99 (1.03)
Course Technologies 0.87 4.25 (0.84) 4.15 (0.88)
The descriptive statistics show that all average responses were above the neutral
point of three, and the standard deviation was around one or lower. Thus, participants
generally had positive perceptions of online courses. As expected, the treatment group
exhibited higher average responses on all of the dependent variables over the control group.
On the other hand, the standard deviations of the responses in the treatment group were
lower than those in the control group. These findings suggest that the faculty training lead to
relatively more positive and consistent perceptions from students with regard to online
63!