7. BEHAVIORAL FACTORS

7.1 Access to Networked Computer

As noted earlier (see section 4.2), in all of the Project's surveys the following question is asked: Is there a computer (in the library or elsewhere) attached to the campus network (directly or by modem) that you can use whenever you want? Our hypothesis is that the easier a Columbia scholar's access to the campus network and materials on CNet and CWeb the more likely he is to adopt online resources, including the collection of online books. In addition, we want to track this measure over time to see how it changes. The responses to this question in the in-class and CWeb online surveys are summarized here.

An overwhelming majority (80 or 94 percent) of the 85 respondents to the CWeb online survey responded to this question in the affirmative.

The students responding to the in-class survey did not see themselves as having such easy access to a networked computer. Of the 239 who responded to this question, 68 percent answered in the affirmative. The percent responding in the affirmative for the different types of classes was:

As in the on-site library survey, undergraduates claim greater access to networked computers than masters students do.

As the above table shows, comparing students' perceived access to networked computers and their preferred book form reveals that such access does not lead students to preferring online books. Given the stated reasons for their preferences, this is logical. Over 66 percent of those responding Yes to this question preferred their own copy of a book while only 61 percent of those responding No did. Photocopy was the preferred form for 13 percent of those responding No, but for only six percent of those responding Yes. This combination of responses suggests that an economic element is at work here. Those who cannot afford their own computers may also prefer not to buy books for classes.

7.2 Time In Online Activities

Based on study of the data, we have settled on collection of information on the amount of time spent per week in various online activities to represent the behavior of the users. (This question was discussed earlier in the context of the on-site survey of library users.) The balance among the various online activities will vary with discipline, and with the position of the user. In the versions of the questionnaire in use since last spring for books in print and online format, the data have been gathered by the following question.

Since these activities are all measured in hours per week, we can sum them to produce a single simple measure of the level of online activity (as we asked the question in the on-site library user survey). The results are instructive. We have prepared tables showing the percent distribution of respondents by number of hours spent online, in all activities, for three different groups: users of online books, users of the OED in paper format, and students surveyed in class. (Table 5 gives equivalent data for the respondents to the March 1997 onsite survey of library users.)

7.2.1 Responses to CWeb Online Survey

As the following table shows, for the 80 users of online books, the mean is 14.8 hours spent online per week and the median is ten hours per week online. The greatest number of hours online reported was 71.

Breaking down the sample into those who claimed easy access to a networked computer (94%) and those who did not, gives means of 15 and 12.3 weekly hours online, respectively. This is not significant because of the small sample size.

7.2.2 Responses to Questionnaire with Paper OED

As the following table shows, for users of the paper format, online activity is lower with a mean of just 3.9 hours.

This set of findings on time in online activities by type of resource being used supports a hypothesis that users of online books will be people who spend significantly more hours per week in online activities than do users of the paper versions. As we reported earlier, onsite library users reporting on their use of online resources in the average week in Winter 1997 had a mean of 5.8 hours.

7.2.3 Responses to In-Class Questionnaire, Fall 1996 and Spring 1997

The in-class questionnaire also asked about weekly hours in online activities. (See Exhibit 4.) Responses were distributed as follows.

Breaking this group down by type of class, we find:

The differences among these groups of students and between the two semesters are not statistically significant.

Breaking the Spring 1997 responses down by their reported access to networked computers gives the following table:

Students claiming easy access to a networked computer spend more time online than those who do not feel that they have easy access. About 31 percent of the former group spend at least six hours a week online while only 18 percent of the latter group do.

In summary, the students surveyed in class have slightly greater uses of online activities than the users of the paper OED, surveyed in the library, but about 10 percent less than library users overall in March 1996 and 1997. Students with easy access to a computer are even greater users of online resources. Together these findings suggest that we are at the beginning edge of the transition to electronic use, and makes us confident that we will be able to map out the complete change in attitudes and behavior as availability and accessibility of online books changes the environment at Columbia University.

One of the more interesting issues to be tracked over the span of this project is whether the center of gravity of user behavior shifts. That is, will we find in two years that the light users of online services are spending more hours per week online than are the average users of today.

[Table of Contents] [Next Page]