|
TCO Checklist:
Support
How Much Is Enough?
|
Stay Updated on taking TCO to the classroom
|
|
Sign up now for the periodic taking TCO to the classroom newsletter!
|
|
After computers are installed, a school district will need people to
help maintain the network and other hardware, and to help users solve
the problems they encounter with their computers and software packages.
The number of support staff required will depend on several variables,
including the number of workstations and the variety of operating systems
and software applications that must be supported.
When an educational PC fails, International Data Corporation (IDC) said, it can get taken out of service
for several days, while a business computer is usually repaired or replaced
within an hour or two. When the Milken
Exchange asked district technology directors how long it takes to
fix a problem when something breaks, and gave them the option of responding
in hours or days, the average number of hours reported was 5.6 and the
average number of days, 3.6.
It's easy to understand why this happens. When a business computer
breaks down, an office worker generally becomes totally unproductive.
When a classroom computer breaks down, a teacher is either expected
to go back to teaching "the old fashioned way" until it is fixed, or
students are expected to "double up" on the computers that are still
working.
School systems often fall back on technologically savvy teachers or
students to help with support. (The Milken Exchange study found that
39.6 percent of the districts surveyed said they "frequently" used teachers
to provide support and 11.5 percent frequently relied on students.)
Unfortunately this can mean that teachers are pulled away from their
primary duties. And when support is inadequate, the district will lose
some of the value of its investment in technology when hardware is not
repaired quickly. More than 29 percent of respondents in the Milken
survey said one reason computers were sitting idle in their schools
was because they needed to be repaired.
The frustrations that lack of support can create were described by
a 1998 congressional General Accounting Office study
of how five school districts covered their technology expenses. The
report noted: "Officials in all five districts reported having fewer
staff than needed. Some technology directors and trainers reported performing
maintenance or technical support at the expense of their other duties
due to a lack of sufficient support staff. Some district officials also
noted high stress levels among district technology trainers or maintenance
staff trying to serve many school sites. One result of a lack of staff
was lengthy equipment downtime when computers and other equipment were
not available for use. In several districts, repairs for some equipment
reportedly took as long as two weeks or more. Equipment downtime means
reduced access for teachers and students, and several officials observed
that this may frustrate teachers and discourage them from using the
equipment."
In what one published guide to school networking considers a "fully-staffed"
model, each full-time technician supports between 100 and 250 users.
Highly standardized networks can reduce the number of support staff required by a factor of 10, according
to some estimates-from one staff person for every 50 to 70 computers
to one for every 500 to 700. Some new, centralized network management
systems can also help control these costs by reducing travel time to
individual schools and permitting many installation, security and back-up
functions to be handled centrally. In addition, tight restrictions on
access to the network can also help contain support costs.
When the Denver
Public Schools set out to determine their Total Cost of Ownership,
for instance, they found that the ratio of technical support staff to
computers was 1:150 for instructional computers and 1:250 for administrative
computers. However, those numbers included persons who had limited technical
skills. When the district calculated the ratio of staff trained to support
networked computers, the ratio was one staff person for every 647 computers.
From 1983-1991, IBM Corp. and Digital Equipment Corp. worked with MIT
as part of the so-called Project Athena initiative to develop a formula
for calculating the number of staff needed to support a distributed
computing environment. It came up with this formula:
Staff members = (Number of workstations/500) +(Number of Users/1,000)
+(Number of clusters sharing servers, printers and other peripherals/15)
+(Number of applications supported/50) + (Number of distinct vendor
operating systems and applications/1) +(Number of Software Licenses
required/25)
Note, however, that the formula does not include factors for supporting
networks or phone systems, or for curricular support.
Smart Valley, a community networking initiative launched by Silicon
Valley companies, advised its schools in a networking guide
that a minimum staffing level for supporting a network was one network
manager at the county level, one network manager at the district level
and a half-time network technician at each school. It envisioned that
teachers, students and volunteers would also be used, but recommended
that they be used only to supplement paid, professional staff.
The benchmarks provided by the California Department of Education in
a four-year technology
plan envisioned that there would be one district or county-level
support person for every three schools, and that each school would have
a half-time technical person on site.
The Council of the Great City Schools found that among 29 of the nation's
largest school systems, there was a wide disparity-from four to 500-of
centralized technology staff reported. When the number of staff was
compared with the size of each district's technology budget, there was
one staff member, on average, for every $150,000 in the technology budget,
as defined by respondents. (Because the district's technology staff
reported its own numbers, it's possible that some school-based technicians
were not included in those numbers.)
A Formula for Tech Support
The Michigan Technology Training Resource, a statewide project funded by a
Technology Literacy Challenge Fund Grant, has been working in formulas and worksheets
for calculating an appropriate level of tech support for a school district.
In a study
comparing the Total Costs of Ownership of school districts and comparably-sized
businesses, International Data Corporation found that schools have "extremely
low" levels of support, usually one person for every 500 computer users,
compared to the 1:50 ratio it, too, found in the business environment.
The state of Maryland, for instance, recently completed a four-year
educational technology plan with a funding projection that assumed that
there would be one support person for every 500 PCs.
The Hidden Costs of Support: A Case Study
As part of a detailed cost analysis that was performed to help decide
whether to purchase PCs or lease them, the Fairfax County, VA, school
district tried to calculate the "hidden" costs it pays when it is
forced to rely on teachers to provide support for computers.
Fairfax County is a large district in the suburbs of Washington,
DC, with 155,000 students and 26,000 employees. It calculated that
if every teacher spends at least one hour a week trying to fix their
own computer problems, that equals 307 Full Time Equivalent positions,
at a cost to the district of $15.3 million in lost teaching time.
In addition, if 5 percent of teachers are regarded as "technical wizards"
by their peers, and are asked to provide 1.5 hours a week of informal
support, that equals 23 Full Time Equivalent positions, at a cost
of $1.2 million.
Thus, the district concluded that its "hidden" costs for technical
support could amount to an estimated $16.5 million.
Even in smaller school districts, the negative impact on budget and
instructional time is proportionately the same. For example, in a
school district of only 35 teachers in which each teacher spends one
hour a week solving their own tech support problems, the total number of hours
lost annually is 1,365=97almost a single full teacher position (calculated
as 1,447.5 hours per school year). Furthermore, if the equivalent of 5
percent of the teaching staff helps out their peers for 1.5 hours a week,
that translates into another 102.38 hours. Together they total the equivalent
of one teaching position a year worth of "hidden" tech support.
Why should this matter? It matters because many schools rely on teachers
for computer support-a responsibility that cuts into their available
time for teaching. A recent survey conducted by the Milken Exchange
on Education Technology found that 40 percent of the districts that
were surveyed "frequently" rely on teachers for support.
Surveys of who is really providing computer support and how much
can help school districts get a handle on the real costs of support.
Schools Are Taking Steps to Manage Costs of Tech Support, CoSN-NSBA Survey Shows
A survey of some of the nation’s most technologically advanced school districts
indicates that nine out of ten have adopted strategies to try to control the
costs of supporting their computer networks. However, only about a third of
them say they have actually adopted benchmarks or standards to measure how well
they are doing.
Those findings emerged from a recent online survey conducted by the Consortium
for School Networking and the National School Boards Association’s ITTE: Education
Technology Programs. The survey was developed as part of "Taking TCO to
the Classroom," a CoSN initiative designed to help school administrators
understand the concept of Total Cost of Ownership so they can budget adequately
for the long-term operation of their computers and networks. Intel Corp. provided
support for this phase of the project.
New Approaches to Networking
When, in the CoSN/NSBA survey, school district administrators were asked
which steps they had taken to try to control the amount budgeted
for school technology, nearly a third of the very largest districts
(over 20,000 students) and over a third of the smaller schools (2,500 to
9,999 students) indicated they had installed a "thin-client" solution in
at least some parts of their districts.
In thin-client architecture, all of the intelligence of the network
resides in a centralized server. All of the applications software
resides at a central location. And individual computers are used only
to communicate with the server.
In this approach, the reliability of the network becomes
critical. If the network is down, none of the terminals will work.
Some experts believe that applications that rely heavily on multimedia
are not as well suited to this kind of environment, and some
thin-client solutions require certain levels of bandwidth to work well.
The tradeoff is that with such tight centralization, the costs of
support can generally be controlled more easily.
This approach to centralization and cost control is taken one step
further with the use of Application Service Providers (ASPs), in which
the applications reside on the servers of an outside company responsible
for all the support operations. Relatively new to both business and
education, ASPs provide a means of limiting costs but with some loss of=20
flexibility and control.
The Gartner Group suggests that an ASP is appropriate for businesses
that are growing too fast, or are expanding geographically but do not
want to duplicate their current applications environment in many
locations, or that cannot afford to pay large up-front costs but can
afford predictable ongoing costs, or that have trouble acquiring and
retaining the technology staff needed in a tight labor market, or,
finally, that find the applications they need are being sold only in a
bundle with other service charges.
These descriptors may well also identify some school districts,
particularly those who have difficulty attracting and retaining support
staff and those who may need a more predictable, per-student or
per-workstation cost to manage their technology environment.
It is important for a school district to understand how it wants to use
new technology and whether the culture of the school district will be
comfortable with a particular approach before selecting a system path.
For example, some districts are starting to build networks that combine
thin clients with more traditional PCs, used in different settings.
Districts should make sure they understand the other
tradeoffs involved so that they will continue to be satisfied with their
decisions and their choices.
|
