You do not have to be a
spreadsheet-wielding manager to earn the bucks. Redefining "career advancement"
is on the agenda of many systems managers.
Working at a university, I normally see a few of the better students
apply to graduate school. This year, none of my students is applying to graduate
school. Rather, a sizable number of students (undergraduate and graduate) are
delaying their schooling to enter the work force, to take advantage of
opportunities before the bubble bursts. Systems managers are being pulled along
with this tide of money. Competition for the smart, bright, motivated, and
skilled person is intense. Toys are tossed in our paths to entice us to another
"opportunity." It's industry, trolling for minds. Startups are not the sole cause
of this vacuum. Long-standing industries are reengineering their computing
infrastructure; all commerce is being channeled to the Web. It is not surprising
that some of the highest-paid individuals run Web farms.
The complete report of the 1999 SAGE System Administrator Salary
Profile is now available to anyone for the asking at
<http://www.usenix.org/sage/jobs/
salary_survey/>. I urge everyone to download a copy. It has much more
information than could be presented here, including salary information by zip
code, industry, and SAGE job categories. Read it, and let us know what you think.
I look forward to active discussion on the <sage-members> mailing
list!
The salary profile evolved from a spontaneous paper-based "sysadmin
profile" at the LISA conferences. This year's results are based on our new
Web-based questionnaire, which will continue to evolve each year. The 2000 Salary
Profile will be refined and expanded to consider the international aspects of our
membership and professional responsibilities. I feel the sample size of 2,314 is
too small and would like to grow that number tremendously with the next
iteration.
I'd like to take a moment to thank all the respondents. The SAGE
Executive Committee hopes this information will improve your daily working life
and give more form to the fine art of systems management!
Report on the SAGE System Administrator Profile
Survey
As part of its ongoing effort to gain recognition and advancement
for system administrators, SAGE annually conducts a System Administrator Profile
Survey. This report is based on results of the 1999 profile that was administered
at the LISA Conference and on the USENIX/
SAGE Web site during November and December 1999.
The 2,314 respondents to the 1999 survey worked in 48 different
countries. The majority (81.6%) worked in the U.S. The other most common
countries were Canada (5.0%), Australia (3.8%), and the United Kingdom (1.9%).
Most analyses of salary, bonuses, and total cash (total cash is salary, wages,
bonuses, and other nondeferred cash payments) were based on only the U.S.
respondents, because of the small sample sizes for other countries.
Most respondents were salaried workers (90.1%), working for a single
employer (90.9%), male (87.3%), and had a bachelor's degree (46.0%) or some
college (21.5%) as their highest level of education. Most worked with Solaris
(77.7%), Windows NT (63.7%), Linux (55.9%), and/or Windows 95/98 (47.8%). On
average, respondents worked with 4.7 operating systems. The majority did not
supervise any subordinates (72.0%) and were not certified on any operating system
(65.4%). They averaged 11.3 days of travel per year, 47.0 hours of work per week,
7.9 years of experience in the field, 2.8 different employers while in the field,
and 34 years of age. Half had worked for their current employer two years or
less. The majority (89.1%) indicated that system administration was their primary
line of work. Over one-fourth of respondents were in the computer/software/
Internet industry, over 15% worked in university or college education, and over
10% were in consulting/business services.
For U.S. system administrators, mean salary was $64,271 and the
median was $62,500; the mean total cash was $70,565 and the median was $65,200;
the mean bonus was $3,464 and the median was zero. Mean and median salaries were
lower in other countries and areas of the world than in the U.S., particularly in
Eastern Europe/Western Asia. Mean salaries were over $75,000 for two New York
City zip codes (10 and 11), one North Carolina zip (28), and the Sacramento and
San Francisco area (94 and 95) zip codes.
For those in the U.S., the average 1999 pay increase from the same
employer for the same job was 7.9%; from the same employer for a promotion,
14.9%; and from changing employers, 23.3%. Over 61% had received an increase from
the same employer for the same job. More than one in six had received an increase
from changing employers, and more than one in ten had received a promotional
increase.
Figure 1: Income ranges, 1998 and 1999
Figure 1 shows the percentage of respondents for each of the two
years who had income in the ranges shown. The percentage with salary less than
$41,000 was the same in 1999 (15%) as in 1998 (15%), because the sample included
a higher proportion of respondents from outside the US (29.0% in 1999 compared
with 14.4% in 1998), where salaries are lower, on average. Percentages in the
middle salary ranges decreased, while the percentage making $76,000 or more
increased in 1999 (31%) compared to 1998 (22%), which demonstrates the trend of
increasing salaries for system administrators.
Over half (52.1%) of the system administrators in the total sample
did not receive a bonus. For those who did, Figure 2 shows the percentage of the
sample who received each type of bonus. Over 22% received a bonus for individual
performance; 17.1% received a bonus for organizational performance, 10.1% for
group or unit performance, and 6.7% for a special project. The median bonus for
individual and organizational performance was $3,000, for group/unit performance
it was $2,000, and for a special project it was $1,500. Of the 21 respondents who
reported a bonus related to stock, the median was $27,000.
Figure 2: Types of bonuses
Eight percent of the system administrators in the sample said they
were contractors, and 1.9% said they were independent, self-employed consultants.
Mean salary was lower for salaried ($62,152) than contractor ($77,307) or
consultant ($96,121) system administrators in the U.S. The same pattern was found
for mean total cash, as shown in Table 1.
Table 1. Total Cash in 1999 by Type of Primary Job
(U.S. System Administrators)
| Job Type | # of
Respondents | Mean Total Cash in U.S.
$ |
|
Salaried | |
1,470 | | $68,742 |
|
Contractor | |
13 | | $78,346 |
| Independent, |
| self-employed |
|
consultant | |
24 | | $106,746 |
Mean salary for those who "assist in the administration of a systems
facility" and "perform routine tasks under the direct supervision of a more
experienced system administrator" (Level 1 of the four SAGE Job Description
Levels) was $47,847. For those at Level 2, who "work under the general
supervision of a computer system manager, carrying out more complex tasks with
some independence," the mean was $54,765. For those at Level 3, who "initiate
some new responsibilities and help to plan for the future of the facility,"
"manage the work of novice system administrators," and "evaluate and/or recommend
purchases," the mean was $61,703. And for those at Level 4, who "design and
manage the computing infrastructure," "work under general direction from senior
management," and "establish or recommend policies on system use and services,"
the mean was $71,725. The same pattern of relationship between increasing level
and increasing pay is seen for total cash in the Table 2.
Table 2. Total Cash in 1999 by SAGE Job Description Level
(U.S. System Administrators)
| Job Level | #
of Respondents | Mean Total Cash in U.S.
$ |
| Level
1 | | 43 | | $50,766 |
| Level
2 | | 181 | | $57,806 |
| Level
3 | | 851 | | $66,578 |
| Level
4 | | 554 | | $81,021 |
| Mean for All
Levels | | 1,629 | | $70,098 |
In the U.S., supervision of subordinates was related to the
following average salaries: no subordinates, $62,448; 1 to 2, $63,297; 3 to 5,
$67,779; 6 to 10, $70,044; and 11 or more, $78,168.
For U.S. system administrators, salary also tended to increase with
hours worked (e.g., $41,848 average for 019 hours; $54,449 for 2034
hours; $62,900 for 4145 hours; $66,156 for 5160 hours).
For U.S. system administrator respondents, 15.4% reported receiving
overtime pay, 0.1% shift pay, and 15.9% on-call/pager pay. Overtime pay for U.S.
system administrators was most prevalent in aerospace, government, and the
military. Shift pay was most prevalent in the aerospace industry and federal
government (nonmilitary). Over one-quarter of U.S. respondents in health care,
state/local government, and transportation organizations said they get paid for
being on call or wearing a pager.
The number of operating systems supported was not related to the
amount of pay, but those who worked with Solaris, BSDI, and HP-UX tended to have
significantly higher salaries, while those who worked with FreeBSD, MacOS
(non-UNIX), OpenBSD, or DOS/Win 3.1 tended to have lower salaries, on average.
About a third of the sample (34.5%) had a certification for an operating system.
Those with certifications for HP-UX, Solaris, and SunOS tended to earn higher
salaries or more total cash. Other operating systems and certifications,
including Linux and Windows NT, did not have a statistically significant
relationship with the amount of salary or total cash earned.
Table 3. 1999 Salary by Level of Education (U.S. system
administrators)
| Level of Education | # of Respondents | Mean Salary in U.S.
$ |
|
Certificate | |
6 | | $46,050 |
| High
School | | 44 | |
$58,055 |
| Some Technical
School | | 77 | | $60,533 |
| Some
College | | 383 | | $61,076 |
| Associate's
Degree | | 123 | | $61,585 |
| Bachelor's
Degree | | 770 | | $64,641 |
| Master's
Degree | | 237 | | $67,998 |
| Ph.D. | | 20 | | $72,160 |
As shown in Table 3, U.S. system administrators with high school as
their highest level of education averaged $58,055 in salary, while those with
increasingly higher levels of education made increasingly higher average
salaries. The number of certifications one had earned was related to salary in
the U.S.: those with no certifications averaged $62,992 in salary; those with one
certification averaged $64,116; and those with six or more averaged $71,322.
Members of SAGE tended to receive higher salaries ($65,818, on average) and more
total cash ($73,990, on average) than those who were not members of a
professional organization (average $61,637 in salary and $65,226 in total cash).
As shown in Table 4, male system administrators reported higher average salary
and total cash than the female system administrator respondents from the U.S.
Regression equations including variables such as education, years of experience,
and job level along with sex indicated that females make lower salaries but
higher bonuses, on average, and do not have significantly lower total cash when
the other variables are taken into account.
Table 4. 1999 Salary and Total Cash by Sex (U.S. System
Administrators)
Sex of Respondent # of Respondents Mean Salary in U.S. $
Mean Total Cash in U.S. $
Male 1,435 $64,883 $71,214
Female 225 $57,777 $62,863
The number of years of experience as a system administrator (or in
similar work) was positively related to higher salaries and total cash (see Table
5), but the number of years with one's current employer was not. Those with two
years or less of experience averaged less than $50,000 salary; those with three
to ten years averaged between $50,000 and $60,000; those with 11 to 14 years
averaged between $70,000 and $75,000; and those with over 15 years averaged more
than $76,000. Age was also positively related to compensation, although the
relationship was not as strong as that between years of experience and
compensation.
Table 5. Salary by Years of Experience
(U.S. System Administrators)
| Years of Experience | # of Respondents | Mean Salary in U.S.
$ |
| 1 or
less | | 58 | | $38,163 |
| 2 | | 90 | | $46,178 |
| 3 | | 138 | | $54,288 |
| 4 | | 159 | | $55,508 |
| 5 | | 204 | | $60,246 |
| 6 | | 145 | | $63,643 |
| 7-8 | | 224 | | $67,814 |
| 9-10 | | 196 | | $67,510 |
| 11-12 | | 142 | | $72,167 |
| 13-14 | | 83 | | $74,749 |
| 15-16 | | 109 | | $76,958 |
| 17-20 | | 73 | | $76,934 |
| 21-25 | | 29 | | $77,950 |
| 26 or
more | | 8 | | $77,780 |
Table 6. 1999 Salary and Total Cash by Number of Employers (U.S.
system administrators)
| # of Employers | # of Respondents | Mean Salary in U.S.
$ | Mean Total Cash in U.S.
$ |
|
1 | | 291 | |
$55,338 | | $58,460 |
|
2 | | 445 | |
$58,925 | | $62,537 |
|
3 | | 419 | |
$65,002 | | $68,901 |
|
4 | | 243 | |
$70,397 | | $77,853 |
|
5 | | 123 | |
$74,876 | | $83,240 |
|
6 | | 55 | |
$71,836 | | $92,523 |
|
7 | | 26 | |
$78,595 | | $137,815 |
| 8 or
more | | 26 | |
$87,667 | | $108,536 |
Table 6 illustrates how U.S. system administrators' compensation was
positively related to the number of employers they had had during their career.
This was not because those who had had more employers also had more experience;
nor was it due to the fact that those with more employers were more likely to be
contractors or consultants. In addition, those who worked for more than one
employer at the time of the survey (i.e., more than one at the same time)
averaged about $5,000 more salary ($68,573 vs. $63,463) and about $6,000 more
total cash ($75,622 vs. $69,589) than those with one employer.
Compensation varied considerably by industry for U.S. system
administrators. The median salary was lowest in state or local government
($49,000) and colleges or universities ($50,000) and highest in finance,
insurance, and real estate ($71,500) and entertainment ($80,000). These were also
the industries with the lowest and highest median total cash, respectively.
State/local government, entertainment, and utilities were among the most generous
in providing fully paid insurance, although this varied somewhat by type of
insurance.
The numbers of computers supported, users supported, and system
administrators employed where one worked were not related to pay levels. The
number of computers or users per administrator in one's facility was also not
related to compensation.
Table 7. 1999 Salary by Organization Size (U.S. system
administrators)
| Number of Employees |
Percent of Responses | Mean Salary in
U.S. $ |
|
1 | | 0.5% | |
$58,788 |
|
210 | |
1.8% | | $48,032 |
|
1150 | |
7.1% | | $61,356 |
|
51500 | |
25.9% | | $64,174 |
|
50110,000 | |
37.2% | | $62,478 |
| 10,001 or
more | | 27.5% | |
$67,941 |
The number of employees in one's organization, worldwide, was
significantly related to salary (see Table 7). Salary tended to go up with the
number of employees. The mean number of paid holidays, vacation days, and
training days, and the percentage with a retirement plan, childcare assistance,
or tuition assistance also tended to increase with the number of employees in an
organization. The percentage who could telecommute or use flextime was lower for
those in organizations with 51 to 2,500 employees than in either smaller or
larger organizations.
U.S. system administrators averaged 15.4 days of paid vacation, 11.1
days of paid sick leave, 8.8 paid holidays, and 8.3 days of paid training per
year. Paid time off tended to be highest in government, college/university,
military, and not-for-profit organizations. Paid training days were highest in
the military. The only industry for which U.S. respondents reported a higher
percentage of defined benefit (i.e., pension) than defined contribution (e.g.,
401k, 403b) retirement plans was state/local government.
The majority (71.9%) of U.S. respondents said "yes," their
organization does have difficulty filling all of the system administrator
positions it would like to fill; 17.9% said "no" and 10.2% said "not sure" to
this question. The industries with percentages above 75% were retail and
wholesale trade, consulting and business services, federal
governmentnonmilitary, aerospace, college/university education, and
advertising/public relations/
communication/marketing. Geographi-cal areas with the highest percentages were
San Diego, the Research Triangle of North Carolina, and the San Francisco
area.
Over half of the respondents said the factors that would be most
important in making them think seriously about switching jobs were pay (83.1%),
location (76.4%), and benefits (60.0%). Over one-third marked organizational
stability (38.9%), hours (38.6%), and organizational reputation (38.3%).
(Respondents could mark more than one category.) The majority of respondents
(80.1%) said they expect still to be a system administrator in five years.
Respondents were asked to note any special benefits or working
conditions they particularly liked and the most problematic or bothersome aspects
of their jobs. The following are ranked according to the number of responses
(e.g., #1 had the most respondents).
Regression analysis was used to determine which of the various job,
organizational, and personal background characteristics on the survey were most
highly related to compensation. Over half (52.0%) of the variance in salary,
23.4% of the variance in total cash, and 8% of the variance in bonuses were
accounted for by the equations. Thus, there were more systematic relationships
between survey topics and salary than between survey topics and total cash or
bonuses. The most significant factors associated with salary were being a
consultant or contractor, location, industry, operating systems used, job level,
education, number of employers, and experience.
More detailed versions of these data and analyses are available from
USENIX/SAGE upon request.
Benefits/Working Conditions That Were Liked
1. Flexible work schedules
2. Jobs that provide challenge, autonomy, variety,
learning
3. Organizational cultures that are university-like,
noncorporate, friendly, or nonbureaucratic
4. Good managers, co-workers, and users
5. Equipment for home offices provided by employers
6. Casual dress or lack of a dress code
7. Free beverages and/or food
8. Telecommuting
9. Working with new technology and good resources
10. Good pay, special awards, stock/options,
or paid overtime
Most Problematic/Bothersome Aspects of Jobs
1. Poor management
2. Long hours, heavy workloads, or being on call
3. Office politics and bureaucracy
4. Low pay or lack of pay for overtime or on-call time
5. Poor resources, low budgets, or lack of help desk
support
6. Routine, unchallenging, menial, or administrative
tasks
7. Users not using resources to solve problems
themselves
8. Lack of training, career development, and career paths
9. Understaffing, recruiting, and retention problems
10. Long commutes, extensive travel, or travel without
notice
|
news
Need help? Use our Contacts page.
Issue index