The following table indicates the average computing hours per individual for each type of activity as well as the percent of total computing time in that activity. Also included are the total average hours per week spent on computing. Statistics are collected for all students, students living on campus, and students living off campus.
Results of Student Network Computing Survey
Surveys Course MU Personal Total
Returned Specific General Activities Ave. Hrs.
On and Off Campus
Ave Hours/wk 1259 11.99 9.0 22.95 43.9
Percentages 27.30% 20.41% 52.28%
Ave Hours/wk 834 11.49 8.6 25.82 45.9
Percentages 25.05% 18.67% 56.28%
Ave Hours/wk 425 12.96 9.7 17.33 40.0
Percentages 32.37% 24.34% 43.29%
Upon inspecting the hours for each individual, it was found that there were outliers that tended to skew the data to the right. For example one student reported spending a total of 135 hours per week on computing. The overall median total number of hours spent on computing was 37; so, when compared with the mean of 43.9, there is definitely skewness to the right. The large values reported by some of the students may be due to double counting across categories.
The greatest percentage of computing time is being spent on personal activities followed by course specific activities, and finally MU General activities. For Course Specific activities, off campus students spent more hours per week (12.96 hours) than on campus students (11.49 hours). Using a one way analysis of variance, this difference was found to be very significant with an observed significance level of p=.01. The same was true with MU-general tasks with off campus students spending an average of 9.7 hours per week where on-campus students spending 8.6 hours per week. This difference was found to be significant at p=.03. On the other hand, on campus students spent an average of 25.82 hours per week on the computer for personal tasks, whereas off-campus students averaged only 17.33 hours. The significance of this difference was extremely high with an observed significant level less than .001.
The following table presents the average hours per week that 1st year through 4th year spend in the three types of computer activities:
To determine if time spent on personal activities significantly declines from 1st year students through 4th year students, a linear regression was performed using average hours per week as the dependent variable and the year of the student as the independent variable. The resultant regression prediction equation was
Average hrs/wk on Personal Activities 30.95 – 3.38* year.
The slope of –3.38 was tested to be very significant at a significance level less than .001. The interpretation is that the average computer hours spent on personal activities approximately declines by 3.38 hours for each year that a student stays in school.
Another regression that used average computer hours spent on course specific tasks as the dependent variable was performed with year of student again as the independent variable. The regression prediction equation was:
Average hrs/wk on course specific tasks = 8.66 + 1.36* year.
For each year that a student stays in school, the number of hours spent on course related tasks approximately increases on the average by 1.36. Again the slope of 1.36 was tested to be highly significant at a level less than .001.
Since the above table shows that the average hours per week spent on MU General tasks increase each year from the first to the third year and then appears to decrease from the third to the fourth year, a linear regression was conducted using MU-general tasks as the dependent variable and year of student up through year three as the independent variable. The prediction equation was:
Average hrs/wk on MU-general tasks = 5.925 + 1.417*year
Again the linear trend coefficient (1.417) was extremely significant at a level less than .001. To determine if a significant drop in usage of MU-general tasks occurs from years 3 to 4, A one way analysis of variance was conducted with year as the classification variable. The test resulted in no significant drop in MU-general task usage between years 3 and 4. The high variation in the amount of usage among 3rd and 4th year students was largely responsible for this nonsignificance.
3. Comparison between the group of SEAS and IDS students with the group of A&S, EAP, SBA, and SFA Students on
Hours Spent on Personal Tasks
The group consisting of students enrolled in either the SEAS and IDS was compared with the group of A&S,EAP,SBA, and SFA students on the average hours per week of personal usage on computers. The results were amazing. The average hours of the former group was 22.847 hrs/wk and 22.967 hrs/wk for the latter. A one way analysis of variance indicated no significant difference.
A multiple analysis of variance was conducted to determine if the group of three classes of tasks show significant differences for hours spent among the 6 colleges. A Wilk’s lambda of .962 yielded an observed significance level of less than .0001 indicating that there are very strong differences. One way analysis of variances conducted for each category of tasks indicated that there were significant differences in hours spent on course related tasks among the colleges but insignificant differences for MU related and personal tasks (for personal tasks this is due to the high standard deviations).
A further analysis, using a Tukey multiple comparison procedure that controls for Type I experimentwise error rate checked for significant differences among divisions for hours spent on course related tasks. Results indicated that students in the SEAS division spent significantly more hours per week than students in SBA,EAP, and Arts and Sciences on these tasks. Similar analysis conducted on the other two categories (muse and personal) indicated no significant differences among the colleges. However, an independent one way of analysis of variance did indicate a significant difference between IDS and SEAS on hours spent on personal tasks (observed significant level = .02). This result should be treated with skepticism since this test did not employ the Tukey correction.
4. Individual Computer Tasks
The average hours week spent on each computer task are tabulated below.
Average hrs/wk over all 22 tasks 2.00 1.29
Note that the high standard deviation for each task indicates large variations between students in the hours per week spent on each task. The two tasks that students spend the most time are e-mail to family and friends (8.75 hrs/wk) and the use of discipline specific software such as word-processing and spreadsheets (6.02 hrs/wk). The total of the two average times is 14.77 hrs/wk which is more than one-third of the total average hours per week usage of 43.91.
The average hours/wk spent per task is 2.00 with 1.29 hrs/wk as the standard deviation. The above two tasks were the only ones that were at least 2 standard deviations above this average hours/wk per task.
5. Comparisons of Time Spent on Each Computer Task for Each Year Student is in School
The average hours per week spent for each computer task is tabulated below for each year of student.
Question Label 1st Yr 2nd Yr 3rd Yr 4th Yr P-Value
CS1 Word processing,database,spreadsheets 5.46 5.21 6.42 7.29 <.0001
PA8 Surfing the web 4.15 4.03 3.94 3.84 .898
A multiple analysis of variance (MANOVA) indicated very significant differences on hours spent per week on this group of 22 tasks for each class of student. A Wilks’ lambda of .701 yielded an observed significance level less than .0001. Individual ANOVAs were done for each of the 22 questions with the classification variable being the year in school. The P-Value (also referred as observed significance level) for these ANOVAs are presented in the last column of the above table. Individual computer tasks that showed significance differences in hours per week for each class of students will have P-Values less than .05. These are the tasks represented by CS1, CS2, CS3, MU1, MU2, MU3, MU5, MU6, MU8, MU9, PA2, PA5, and PA6.
Looking at each task in detail, the significant differences (at a significance level of .05 or less) in hours per week spent in task CS1 were between 1st and 4th year students and 2nd and 4th year students. The same was true with the task described in CS2. The major significant differences in the task described by CS3 were between 1st year and 3rd year, 1st year and 4th year, and 2nd and 4th year.
Concentrating on MU related tasks, the major difference in hours/wk for MU1 were between 1st and 3rd year students.
MU2 task had the largest hours/wk for 1st year students and this was significantly than the time for 2nd , 3rd , and 4th year students. For the task described by MU3 , the major difference was between 1st and 4th year students. For MU5 more time was spent by 2ndyear students on CBT courses, and this was significant than the time spent by 4th years. 3rd year students
spent more time with MU6 task and this was significantly higher than any of the other classes. Both 3rd and 4th year students spent more time using computers for part-time work (MU8) and these were significantly larger than the time spent by 1st years. With the exception between 1st and 2nd year students, the MU9 task had significant differences in hours per week between each possible pairwise combination of classes.
For personal tasks, PA2 had the most significant comparisons between each pair of classes. The only pair for which the difference in hours/week failed to be significant was between the 3rd and 4th year students. 1st year spent more hours per week on computer games (PA5), and this was significantly higher than the other three classes. The same was true for the task in PA6 with 1st year students spending significantly more time than each of the three other classes in downloading music and movies.
6. Comparisons of Time Spent on Each Computer Task for On Campus and Off Campus Students The table below shows the average hours per week spent by on-campus and off-campus students for each individual computer task. A MANOVA indicated overall significant differences between on and off campus students on hours per week spent on the group of 22 tasks. A Wilks’ lambda of .875 yielded an observed significance level that was less than .0001. Tasks that showed significant differences between the two groups have p-values (observed significant levels) less
Average Hours per Week
Question Label Off Campus On Campus P-value
CS1 Word processing,database,spreadsheets 6.51 5.76 .030
CS2 Library usage 2.75 2.20 .0007
CS3 Publishing web sites, accessing graphics 1.09 .943 .222
CS4 Using Blackboard 1.28 1.42 .220
CS5 Accessing a course web site 1.33 1.17 .252
MU1 Emailing faculty,fellow students 2.75 2.44 .085
MU2 Accessing myMiami 1.45 1.94 .003
MU3 Applying for admission,financial assistance .118 .192 .012