In an attempt to determine how networking technology is being utilized for education and research at Lehigh University and the attitudinal implications that this technology has toward education, an on-line questionnaire was placed on the Network Server for completion by students, faculty, and staff. This questionnaire included components pertaining to how networking technology is being utilized at Lehigh for education and research and pertaining to the attitudes toward the technology of the users of networking technology in education and research.
The population used for this study included all active users of the campus-wide information system residing on the Network Server systems at Lehigh University during the period from September 23, 1994 through October 7, 1994. Every student, faculty, and staff member at Lehigh University is entitled to an account to use the Network Server with over 80% of those individuals utilizing their accounts in any given month.
Lehigh University enrolls approximately 4,500 undergraduates and approximately 2,000 graduate students and employs approximately 420 faculty members and 640 support staff members. Lehigh offers baccalaureate, masters, and doctoral degrees with programs administered within four colleges: the College of Arts and Science, the College of Engineering and Applied Science, the College of Business and Economics, and the College of Education.
Overall, 497 individuals completed the survey. As shown in Table 1, of these individuals, 340 (68.4%) were male and 157 (31.6%) were female. This represented a 7.5% response rate for all individuals with Network Server accounts and approximately a 9.4% response rate for individuals who actually logged into their accounts (i.e., active users) during that two week period. The response rate for active users ranged from 5.3% for undergraduate seniors to 20.3% for faculty members. For active undergraduate students as a whole, 6.7% responded (5.4% of juniors, 7.8% of sophomores, and 8.0% of freshmen). For active graduate students, 10.7% responded. For active exempt staff, 13.4% responded as did 16.3% of non-exempt staff members.
For undergraduate respondents, 61.0% were in the College of Engineering and Applied Science, 25.7% were in the College of Arts and Science (5.5% in Humanities, 6.4% in Social Sciences, and 13.8% in Sciences), and 9.2% were in the College of Business and Economics. Of all undergraduate respondents, 76.1% were male and 23.9% were female.
Graduate student responses were similar with 49.5% of graduate student respondents being in the College of Engineering, 27.3% in the College of Arts and Science (7.1% in Humanities, 8.1% in Social Sciences, and 12.1% in Sciences), 16.2% in the College of Education, and 6.1% in the College of Business and Economics. Of all graduate respondents, 73.7% were male and 26.3% were female.
For faculty member respondents, 54.4% were associated with the College of Arts and Science (11.8% in Humanities, 11.8% in Social Sciences, and 30.9% in Sciences), 25.0% with the College of Engineering, 13.2% with the College of Business and Economics, and 4.4% with the College of Education. Of all faculty respondents, 80.9% were male and 9.1% were female.
Survey Respondents: Affiliation by Campus Status and Gender
Freshman Sophomore Junior Senior
M F T M F T M F T M F T
A&S-Humanities 6 0 6 1 1 2 1 1 2 2 0 2
A&S-Social Sci 0 1 1 1 4 5 1 1 2 4 2 6
A&S-Sciences 5 6 11 6 4 10 2 3 5 3 1 4
Engineering 39 7 46 30 4 34 25 5 30 19 4 23
Business 3 2 5 5 0 5 3 1 4 5 1 6
Other 2 0 2 3 4 7
Total 55 16 71 46 17 63 32 11 43 33 8 41
% of Active Users 8.0% 7.8% 5.4% 5.3%
Grad Faculty Non-Exmpt Exempt
M F T M F T M F T M F T
A&S-Humanities 4 3 7 4 4 8 0 2 2 0 1 1
A&S-Social Sci 1 7 8 3 5 8 0 2 2 0 1 1
A&S-Sciences 10 2 12 21 0 21 2 3 5 1 1 2
Engineering 45 4 49 17 0 17 2 2 4 3 2 5
Business 5 1 6 6 3 9 0 2 2 2 1 3
Education 7 9 16 3 0 3 0 2 2 2 1 3
Administrative 6 14 20 14 9 23
Academic - Other 0 1 1 1 2 3 3 5 8
Other 1 0 1 1 0 1 5 9 14 5 7 12
Total 73 26 99 55 13 68 16 38 54 30 28 58
% of Active Users 10.7% 20.3% 13.4% 16.3%
____________________________________________________________
Over the two week period from September 23, 1994 through October 7, 1994, a number of system bulletins were placed on the Network Server requesting that all users of the system complete the on-line Networking Technology Questionnaire. Appendix A.01 contains a copy of the system bulletin; Appendix A.02 contains a copy of the Networking Technology Questionnaire.
Networking Technology Questionnaire
For the purpose of this study, the Networking Technology Questionnaire was developed to obtain a profile of the use of networking technology at Lehigh University. Before posting the questionnaire on-line, the survey instrument was tested for both reliability and validity. According to Oppenheim (1966), "reliability refers to consistency, to obtaining the same results again" (p.69) while "validity tells us whether the question or item really measures what it is supposed to measure" (pp.69-70). In discussing reliability, Fowler (1984) stated that "one step toward ensuring consistent measurement is that each respondent in a sample is asked the same set of questions" (p.75) which was the case with this questionnaire. Fowler also stated that "if all respondents are asked exactly the same questions, one step has been taken to ensure that differences in answers can be attributed to differences in respondents" (p.79) and that "the questions should all mean the same thing to all respondents" (p.79). To ensure that this was the case with this questionnaire, all questions were first pilot tested with a number of individuals with diverse computer backgrounds.
One measure of validity as described by Smith and Glass (1987) is that of face validity. In describing face validity, they stated that "researchers attempting to support the interpretation of the measurement and its connection to the construct will seek professional judgment that there is a plausible connection between the surface features of the measure's content and the construct as theoretically defined" (p.107). To ensure face validity, all questions in this questionnaire were reviewed by individuals who are familiar with both networking technology and educational constructs including personnel from both the Lehigh University Computing Center and the Lehigh University Libraries.
The questionnaire items related to the first general research question "What is the relationship between user classification and level of utilization of networking technology for educational and research purposes at Lehigh University?" and to the intellectual skills being utilized through networking technology, were derived from the Taylor Residential College Networking Questionnaire. That questionnaire was created to obtain feedback from participants of a pilot project to network the Taylor Residential College residence hall at Lehigh University. Through that questionnaire and individual follow-up questioning, face validity was established for items which relate to knowledge acquisition, comprehension, and application.
The questionnaire items related to the second research question "What are the attitudes of users toward the utilization of networking technology for education and research at Lehigh University?" were derived from the Computer Attitudes Scale, as described and validated by Gressard and Loyd (1985), for items pertaining to the enjoyment of using networking technology. For items pertaining to anxiety and user confidence levels in using networking technology, questionnaire items were derived from the Attitudes Toward Computer Usage Scale (ATCUS), as described and validated by Popovich, Hyde, Zakrajsek, and Blumer (1987).
The Computer Attitudes Scale, as described by Gressard and Loyd (1985), contains three subscales to measure attitude pertaining to computer anxiety, computer confidence, and computer liking. Three different studies were conducted to determine the validity of the subscales. The first study used a factor analysis and subscale reliability coefficients to determine that the three subscales were sufficiently stable. The second study compared the subscale results to another instrument to "indicate that the Computer Attitudes Scale has reasonable convergent validity" (p.18), while the third study utilized pre-tests and post-tests to determine that "the Computer Attitude Scale is sensitive to attitude changes resulting from computer instruction and experience" (p.18).
The Attitudes Toward Computer Usage Scale (ATCUS), as described by Popovich et al. (1987), was found to have an internal reliability coefficient of .84. A factor analysis was performed on the data collected with this instrument and four interpretable factors were found. These were "Negative reactions to computers" (p.266), "Positive reactions to computers" (p.266), "Computers and children/education" (p.266), and "Reactions to (familiar) computer-related mechanisms" (p.266). In a study of undergraduate students utilizing this instrument, significant sex differences were found with "females responding more negatively to computer usage than males" (p.267). In another study which utilized the Attitudes Toward Computer Usage Scale, Baack, Brown, and Brown (1991) compared the attitudes toward computers of older adults with those of younger adults and found that "the young adults (college students) surveyed in this study revealed more positive attitudes toward computers and computer applications than did the older adults" (p.429).
Conducting surveys in an on-line environment is no longer a unique occurrence. Ferl and Millsap (1992), Snure (1991), and Horner and Thirlwall (1988), have all described on-line surveys conducted within networked information environments. In describing a study which was specifically designed to determine the validity of computer administration of attitude questionnaires, Booth-Kewley, Edwards, and Rosenfeld (1992) found that "the present results of no systematic difference between computer and paper-and-pencil responding on attitude questionnaires are consistent with much of the published research literature" (p.565) and stated that "an argument for the use of computers is that they are typically regarded more favorably by respondents" (p.565).
The Networking Technology Questionnaire was analyzed through various crosstabulations of the questions with respect to demographic information pertaining to the respondents. Where items within these tabulations were mutually exclusive, a chi-square analysis was performed. As stated by Dowdy and Wearden (1983), although the chi-square distribution "is usually not a direct model of a population distribution, it has many uses when we are trying to answer questions about populations" (p.97). One of these uses is to determine whether any relationship is present between the items.
The research question "What is the relationship between user classification and level of utilization of networking technology for educational and research purposes at Lehigh University?" was answered through the testing of a number of hypotheses. It was hypothesized that there were no significant differences among user classification groups regarding the perceived users of networking technology. This was tested through an analysis of responses to items 5 through 8 (percentages of undergraduate students, graduate students, faculty members, and staff members which respondents felt utilized networking technology) with respect to item 2 (campus status). Individuals may refrain from using electronic mail to communicate with certain groups (e.g., graduate students) with the impression that few members of that group utilize networking technology. This hypothesis was meant to determine whether this perception existed and whether it was valid.
It was hypothesized that there was no significant relationship between user classification and the frequency with which members of the groups utilized campus networking technology. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 37, 38, 39, and 40 (frequency of use of the Network Server, Compute Servers, remote use of AFS workstations, and direct use of AFS workstations, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of electronic mail for communications by members of the groups. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 9 and 10 (frequency of utilizing on-campus and off-campus electronic mail, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of electronic mail for student-faculty communications. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 11, 12, and 13 (frequency of utilizing electronic mail with undergraduate students, graduate students, and faculty members, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of electronic mail by faculty members for non-student-faculty communication. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 13 and 15 (frequency of utilizing electronic mail with faculty members and with off-campus individuals, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of electronic mail by students for non-student-faculty communication. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 11, 12, and 15 (frequency of utilizing electronic mail with undergraduate students, graduate students, and off-campus individuals, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of sharing files electronically. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 18 (frequency of uploading or downloading a file for sharing purposes) and 19 (frequency of utilizing a LAN to exchange files).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of off-campus computing (excluding library and database systems which are analyzed elsewhere). This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 20, 21, 22, and 36 (frequency of utilizing archive servers, anonymous ftp, Gopher and Mosaic, and connecting to off-campus computers for purposes other than library or database usage, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of Usenet News and BITNET LISTSERVs. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 16 (frequency of reading a news group) and 17 (frequency of posting to a news group).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of local library resources. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 23, 30, 31, 32, and 33 (frequency of remote connection to ASA, and frequency of utilization of the library's CD-ROM LAN, on-line forms, REFERENCE aids, and electronic texts, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of locally mounted databases. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 26 and 27 (frequency of utilization of ERIC and EngIndex, respectively).
It was hypothesized that there was no significant relationship between user classification and the frequency of use of off-campus libraries and databases. This was tested through analyses of crosstabulations of responses to item 2 (campus status) with responses to items 24, 25, 28, 29, 34, and 35 (frequency of utilization of CARL/Uncover, WorldCat, Dow Jones News Retrieval, Lexis/Nexis, off-campus library systems, and off-campus database systems, respectively).
The questionnaire items 56 through 60 related to the intellectual skills being utilized through networking technology. These items were derived from the Taylor Residential College Networking Questionnaire which was created to obtain feedback from participants of a pilot project to network the Taylor Residential College residence hall at Lehigh University. Item 56, "I constantly find new information available to me through the use of network services," pertains to the knowledge acquisition intellectual skill. Item 57, "By accessing information through the Network Server and the Internet, my abilities to distinguish the value of information I encounter has greatly increased," pertains to the comprehension intellectual skill. Item 58, "Utilizing different search techniques to find information through the Network Server and the Internet has increased the scope of information available to me," pertains to the application intellectual skill. Item 59, "My skills in assembling information into a more usable form have increased due to the use of networking technology," pertains to analysis, synthesis, and evaluation intellectual skills. And, item 60, "Using networking technology has had a significant positive impact on my ability to get my work done," pertains to all of the intellectual skills found within the cognitive domain of learning. It was hypothesized that there were no significant relationships among user classification groups for any of these items.
The second research question "What are the attitudes of users toward the utilization of networking technology for education and research at Lehigh University?" was answered through the testing of a number of hypotheses. All hypotheses in this section were analyzed through a Likert scale by obtaining the mean for responses for the corresponding group of questionnaire items for each user and then performing an analysis of variance (ANOVA) to determine whether any significant differences existed among the user classification groups or within each user classification group. As stated by Oppenheim (1966), "Likert scales tend to perform very well when it comes to a reliable, rough ordering of people with regard to a particular attitude" (p.141).
It was hypothesized that there were no significant differences among user classification groups regarding the degree to which members of the groups enjoyed the utilization of networking technology for retrieving information. This was tested through an analysis of combined means of responses to item 41, "I like working with networked information resources," item 44, "The challenge of obtaining information through computer networks does not appeal to me," item 47, "I think that working with networked information resources is enjoyable and stimulating," item 50, "I do as little work with networked information resources as possible," and item 53, "I enjoy talking to others about networked information resources and networks." For this specific question, a low score (1 on a scale of 1 to 5) signified a high level of enjoyment. Since items 44 and 50 were negatively worded, responses were coded in reverse for analysis.
It was hypothesized that there were no significant differences among user classification groups regarding the degree to which members of the groups had a high anxiety level toward the utilization of networking technology. This was tested through an analysis of combined means of responses to item 43, "Whenever I use something that is computerized I am afraid I will break it," item 46, "I know that I will not understand how to access networked information resources," item 49, "Using networked computing resources is too time consuming," item 52, "I have had bad experiences accessing networked computing resources," and item 55, "I do not like using networked information resources because I do not know from where the information is coming." For this specific question, a low score (1 on a scale of 1 to 5) signified a high level of anxiety.
It was hypothesized that there were no significant differences among user classification groups regarding the degree to which members of the groups were confident in their abilities to utilize networking technology. This was tested through an analysis of combined means of responses to item 42, "I would prefer finding information through a computerized database instead of a book," item 45, "Having a networked computer available to me helps with my job, research, or studies," item 48, "I prefer not to learn more about computer networking and available resources," item 51, "I would like to have a networked computer connection in my home," and item 54 "I like to browse through networked information resources to find what is available." For this specific question, a low score (1 on a scale of 1 to 5) signified a high confidence level. Since item 48 was negatively worded, responses were coded in reverse for analysis.