[Printout | Contents | Search ]

Resnet: Student survey results

By Daniel A. Updegrove and George P. McKenna

In March the University's Board of Trustees approved plans for ResNet Phase 3, which will provide, by September 1995, PennNet Ethernet, cable television, and enhanced phone service to 1,260 students in five resi dences - DuBois House, Hill House, Mayer House, Stouffer College House, and Van Pelt College House. A total of 4,770 students will then have ResNet connections, leaving only the two Grad Towers and most of the Quad unwired. Engineering and cost studies will be undertaken during the next few months to determine if the remaining buildings can be completed in one additional summer or two.

Before approving the Phase 3 plan, Provost Stanley Chodorow requested an assessment of the use and impact of the first two phases of ResNet. To that end, a five-page questionnaire was distributed in all student residences (wired and unwired) in January. Of the 6,300 surveys distributed, 1,160 were returned, a 19 percent response rate. The survey results reveal much about ResNet as well as the state of student computing and network use on the campus.

Background

In March 1993 the Trustees approved a four-year plan to wire the residences, with funding from both residential and general student fees. In September 1993, 1,620 students moved into the first five wired buildings: Class of 1925/Modern Languages College House, English House, Kings Court, Ware College House (Quad), and High Rise North. To promote effective use of the Ethernet data connections, software was made available at no cost and installed, upon request, in student rooms. In addition, an automatic IP (Internet protocol) assignment system was developed, which simplifies start-up and permits students with laptops to roam from building to building. By year's end, about 310 (20 percent) of the students had used their Ethernet connection.

In 1994 wiring was extended to 1,890 students in Harrison and Harnwell Houses. The software distribution and installation processes were streamlined such that many students were able to obtain and install networking software without assistance; as of mid-March, over 880 ResNet students (26 percent) had connected their computers to PennNet. Television reception was improved by a change from the rooftop laser and microwave transceivers employed in Phase 1 to underground cable in Phase 2. In addition, the ResNet Video Network was merged with the Academic Video Network, which had delivered news, public affairs, and foreign language programming to 35 academic and administrative buildings.

The ResNet program also includes three additional components:

• Upgrading computer labs in the residence halls
• Expanding the modem pool
• Creating a central server, named "dolphin," for use by students in the eight "small" schools - Annenberg, Dental Medicine, Education, Fine Arts, Law, Nursing, Social Work, and Veterinary Medicine - to access electronic mail, NetNews, and World-Wide Web resources, including their own personal home pages

ResNet's goals

ResNet was conceived as a multi-faceted program with multi-dimensional goals:

• Contribute to University efforts to make residences "more academic" in character, via access to high-speed PennNet/Internet connections as well as public affairs and foreign-language television programs in rooms and lounges
• Promote integration of computing and information resources into curricula, via reduced barriers to access and increased student information technology literacy
• Enhance Penn's ability to prepare graduates for a networked, information-rich, international, multi-cultural world
• Remain competitive with peer institutions who already offer or are planning to offer similar services
• Increase residence hall occupancy
• Meet a growing demand for access to computing and information resources without concomitant investment in student computer labs by leveraging student computer ownership and the convenience of in-room network access
• Reduce growth of modem pool expenses
• Enhance residential telephone services and reduce maintenance costs
• Promote use of electronic communications in the small schools

Survey results

Of the 1,160 survey respondents, 628 live in ResNet buildings, 432 live in unwired residences, and 100 neglected to identify their residence. Response rates by building ranged from 37 to 5 percent, with wired buildings tending to elicit higher response rates. See the tables below for breakdowns by class, School, and number of roommates. A systematic assessment of non-response bias is in process.

Overall, 16 percent of students said they use computers intensively, 41 percent reported frequent use, 38 percent occasional use, and 5 percent never use them. These percentages did not differ substantially between ResNet and other residences.

A total 54 percent of ResNet students indicated that availability of ResNet services affected their room selection this year, as compared to 13 percent of those not in ResNet; 49 percent of the freshman, sophomores, and juniors in both ResNet and non-ResNet buildings agreed that ResNet will affect next year's housing selection.

In ResNet, 68 percent reported owning a computer (53 percent desktop and 15 percent laptop) compared to 62 percent non-ResNet (46 percent desktop, 16 percent laptop); 44 percent in ResNet reported sharing computing equipment with roommates versus 20 percent in other buildings. Reflecting this greater in-room access, ResNet students report less use of residence labs (12 vs. 22 percent) and School/Library labs (17 vs. 21 percent) for their primary computing.

An important finding is that 22 percent of ResNet students use a modem (vs. 45 percent outside ResNet). Further study is required to determine if ResNet modem use, which is discouraged, is due to the cost of Ethernet hardware, perceived complexity of Ethernet software installation, requirement for access to external bulletin boards not accessible via the Internet, or other factors.

In general, ResNet students make more use of network resources (although note that in Table II on page 9, "library use" was defined to mean network access to library resources from outside libraries).

Television

Although about 53 percent of all students reported owning televisions, 34 percent of ResNet students had access to roommates' sets, compared to 17 percent in non-ResNet buildings. Additionally, more ResNet students reported owning VCRs (36 vs. 27 percent). Reflecting both greater access to televisions and greater channel selection, ResNet students spend more time watching TV each week (15 percent watch over 14 hours; 47 percent watch from 5 to 14 hours; vs. 9 percent and 36 percent, respectively), and more frequently ("almost always" plus "frequently," as opposed to "sometimes" or "never") watch news (26 vs. 15 percent), sports (55 vs. 38 percent), the ResNet movie channel (51 vs. 7 percent), and foreign language programming (10 vs. 4 percent). The overall percentage of students reporting use of television for assignments was 12 percent.

Telephones

In ResNet buildings, 605 students reported that they opted to have a private telephone line installed, compared to 15 students in other residences.

"Bottom line" questions

Nine percent of ResNet students agreed that availability of ResNet services influenced their decision to attend Penn, compared to six percent of students in other buildings. When asked if ResNet had contributed to their academic experience, 76 percent in ResNet agreed vs. 26 percent in other buildings. Regarding ResNet's positive contribution to the quality of life at Penn, 87 percent in ResNet agreed, compared to 37 percent of students living elsewhere.

Looking forward, 57 percent in ResNet agreed that the services would help with future work goals (vs. 40 percent), 47 percent agreed that availability of ResNet-type services could influence their choice of graduate schools (vs. 30 percent), and, as noted above, 49 percent of freshman, sophomores, and juniors in both ResNet and non-ResNet buildings agreed that ResNet would influence their choice of residence next year.

Some observations

It is important to note that ResNet is an investment in "deep infrastructure" to support emerging changes in the way Penn carries out teaching and learning as well as changes in the role of on-campus residences in the academic and social life of the campus. It appears that the University launched ResNet at the ideal time to leverage the explosive growth of scholarly interest in the Internet. Nevertheless, since ResNet is only half finished - and, in particular, since very few freshman or graduate/professional rooms are wired - the full value of ResNet will not be apparent for several years.

Nettlesome issues remain, however. We need a better understanding of some students' preferences for modem use over Ethernet, and (potentially related) better understanding of how to accommodate the 15 percent of students with laptop computers. For example, in academic buildings there are few ports available to attach an Ethernet-ready laptop. We also have evidence that our network acceptable use policy will have to be adapted - and enforced - as new uses and abuses are discovered.

Finally, our biggest challenge is likely to be providing students who don't live in the residence halls with cost-effective access to data networking services equivalent to those available to ResNet students. This remote access is equally important to faculty and staff from their homes as networked computing becomes a key component in the way people live, learn, teach, and work at Penn.

                      Table I:  Survey respondents

                    Breakdown of respondents by class

              Freshman   Sophomore   Junior   Senior   Grad/Prof   Total

ResNet           88        260        175      90        14          627
No ResNet info   11         24         17      16        32          100
Non-ResNet      151         59         34      31       157          432
                ========================================================
Total           250        343        226     137       203         1159


                              Breakdown by School


              College    Wharton     Nursing   SEAS     Other      Total

ResNet          337       134          25     114        12          622
No ResNet info   36        23           8      15        18          100
Non-ResNet      179        84          10      55       103          431
                ========================================================
Total           552       241          43     184       133         1153


                       Breakdown by number of roommates 


                  Zero       One           Two         Three       Total

ResNet             94        226           134          174          628
No ResNet info     39         31            15           15          100
Non-ResNet        205        177            36           14          432
                  ======================================================
Total             338        434           185          203         1160

Note:  Because some students did not respond to every question, the 
totals vary slightly from the 1,160 total survey respondents. 



          Table II:  Students using specific resources (%)

              E-mail  News  Lynx  Mosaic  Gopher  PennInfo  Library  FTP

ResNet          97     76    37     50      72       55        71     41
No ResNet info  90     65    38     33      59       58        73     31
Non-ResNet      93     63    39     39      62       53        63     30
                ========================================================
Total           95     70    38     45      67       54        67     36