ArticlePDF Available

Ecological Footprint of Student Population and its use in policy analysis at Jackson State University

Authors:

Abstract and Figures

Scientists continuously look for suitable indicators to measure the impact of humans on the ecosystem. Ideally, such indicators should also be suited as benchmarks for ascertaining the effects of administrative policies. One such indicator is the Ecological Footprint, which although it has a short history, has gained a widespread popularity. Since the first introduction of the Ecological Footprint, studies analyzing its impact on the ecosystem of nations, regions, or individuals have become widespread. The level of analysis has tended to shift toward institutions and, due to their role as opinion leaders, universities were the first to calculate the impact they have on nature. This study provides a complementary view by analyzing the Ecological Footprint of the student population at Jackson State University and the role that a possible campus policy may have in reducing it. It finds that the average Ecological Footprint of a student is lower than that of the average U.S. citizen by five acres. Our data suggests a policy that discourages freshmen and sophomore students from bringing their cars on campus would reduce the Ecological Footprint of the student population by about 4,000 acres or sixteen times the campus area. The study provides an example of how the Ecological Footprint can be used to plan for future campus development and create a campus that is aesthetically and ecologically balanced.
Content may be subject to copyright.
Journal of the Mississippi Academy of Sciences
Volume 52 October 2007 Number 4
Editor
Michelle Tucci
University of Mississippi Medical Center
Associate Editor
Edwin Swiatlo
University of Mississippi Medical Center
Editorial Board
Maria Begonia
Jackson State University
Gregorio Begonia
Jackson State University
Ibrahim Farah
Jackson State University
Robin Rockhold
University of Mississippi Medical Center
Abstract Editor
John Boyle
Mississippi State University
Program Editor
Ann Marie Kinnell
University of Southern Mississippi
The Journal of the Mississippi Acad-
emy of Sciences (ISSN 0076-9436) is
published in January (annual meeting
abstracts), April, July, and October, by the
Mississippi Academy of Sciences. Mem-
bers of the Academy receive the journal as
part of their regular (non-student)
membership. Inquiries regarding
subscriptions, availability of back issues,
and address changes should be addressed
to The Mississippi Academy of Sciences,
Post Office Box 55907, Jackson, MS
39296-5709, telephone 601-977-0627, or
email-msacad@bellsouth.net.
Table of Contents
Katrina: Special Report
223 Challenges of Collecting Survey Data on the Mississippi
Gulf Coast After Hurricane Katrina: An In-Depth
Interview Study of Survey Team Members-Ann Marie
Kinnell and Kristen Dellinger.
228 Assessing Katrina’s Demographic and Social Impacts on
The Mississippi Gulf Coast- David Swanson, Rich Forgette,
Mark Van Boening, Cliff Holley, and Ann Marie Kinnell.
243 The Role of Information and Communication Technology
(ICT) in the Resilience of Educational Institutions in the
Wake of Hurricane Katrina. Bertta Sokura and Arthur
Cosby.
262 College Students’ Experiences with Hurricane Katrina: A
Comparison Between Students From Mississippi State
University and Three New Orleans Universities-Duane Gill,
Anthony E. Ladd, and John Marszalek.
281 Hurricane Katrina As A Natural Experiment of ‘Creative
Destruction’-Ronald Cossman.
289 Water Quality Studies On Freshwater Bodies in New
Orleans Louisiana One Year After Hurricane Katrina
Alex D. Acholonu and Tiffari Jenkins.
Note
295 A Note on Additional Plants Found at the Sixteenth Section
(Osborn) Prairie- JoVonn G. Hill and Jennifer L. Seltzer.
Articles
298 Ecological Footprint of Student Population and its Use in
Policy Analysis at Jackson State University- Cristina Nica,
Elegenaid I. Hamadain, Ibrahim Farah, and P.C. Yuan.
310 Intraosseous Dental Implants: The Relationship Between
Morphological Configuration and Biological Response;
Thoughts for Consideration-Robert A. Deville
Departments
318 Executive Director Column
319 MAS 2008 Meeting Information
222 October 2007, Vol. 52 No 4
OFFICERS OF THE
MISSISSIPPI ACADEMY OF SCIENCES
President………………………………………………………………………Joseph A. Cameron
President-Elect………………………………………………………………..Rodney Baker
Immediate Past-President……………………………………………………..Juan Silva
Executive Officer……………………………………………………………..Hamed Benghuzzi
Junior Academy Co-Director…………………………………………………Maxine Woolsley
Junior Academy Co-Director………………………………………………… Ken Sleeper
Directors……………………………………………………………………………John Boyle
………………………………………………………………………………….Ann Marie Kinnell
…………………………………………………………………………………..…Michelle Tucci
Administrative Assistant…………………………………………………………..Cynthia Huff
The Mississippi Academy of Sciences recognizes the following
Gold Booth Exhibitor, 2006 Annual Meeting:
Base Pair
Dr. Robin Rockhold
University of Mississippi Medical Center
2500 North State St.
Jackson, MS 39216-4505
601-984-1634 (phone)
rrockhold@pharmacology.umsmed.edu
The Mississippi Center for Supercomputing Research (MCSR) provides free, high performance
computing cycles and consulting in support of research and instruction, for all interested
students, faculty, or researchers associated with any of Mississippi’s eight publicly funded
institutions of higher learning. The MCSR actively supports the Mississippi Academy of
Sciences with regular participation in the Mathematics, Computer Science, and Statistics
Division. Please visit http://www.mcsr.olemiss.edu, email assist@mcsr.olemiss.edu, or call 662-
915-3922 to inquire about how we might support your HPC research or instructional computing
projects at your university. Or, simply apply for an account today at
http://www.mcsr.olemiss.edu/accounts.
Journal of the Mississippi Academy of Sciences 223
CHALLENGES OF COLLECTING SURVEY DATA ON THE MISSISSIPPI
GULF COAST AFTER
HURRICANE KATRINA: AN IN-DEPTH INTERVIEW STUDY OF
SURVEY TEAM MEMBERS
Ann Marie Kinnell and Kirsten Dellinger
University of Southern Mississippi, Hattiesburg, MS, 39406, and
University of Mississippi, University, MS, 38677
Ann Marie Kinnell corresponding author: ann.kinnell@usm.edu
ABSTRACT
The purpose of this paper is to offer a brief description of the unique context faced by a team of
researchers collecting survey data on the Mississippi Gulf Coast four months after Hurricane
Katrina (see Swanson et. al. this issue for more information on the larger study). Based on in-
depth interviews with survey team members, we discuss several challenges faced during data
collection: locating subjects, soliciting subjects' participation, and collecting completed surveys.
We conclude by discussing the methodological implications of these challenges.
INTRODUCTION
Although studies of disasters and
their aftermath usually include a
discussion of methods and data, few
studies have examined the actual process
of gathering data after a disaster (for an
exception see Killian 1956/2005). In the
area of disaster research, there is a
growing focus on methodology.
Researchers recognize that the context of
a disaster and its aftermath can pose
unique methodological challenges to a
study. For instance, the difficulty of
recruiting subjects and guaranteeing the
safety of data collectors may be
heightened (Knack et. al., 2006). As a
recently edited volume on the methods
of disaster research notes, although the
actual methods used by disaster
researchers are not unique, their
application in the field needs to be better
studied and understood (Stallings, 2002).
The purpose of this paper is to
offer a brief description of the unique
context faced by a team of researchers
collecting survey data on the Mississippi
Gulf Coast four months after Hurricane
Katrina. The NSF-funded study had two
objectives:
(1) to gather pre- and post-
Katrina information on
housing and population; and
(2) to distribute to coastal
residents a self-administered,
115-item questionnaire to
collect retrospective
information on the roles that
social and kinship networks
played in sustaining a
respondent’s well-being after
Hurricane Katrina (see
Swanson et. al. this volume
for specific information on
the NSF-funded study and
preliminary analysis of
results).
Over a one-week period in early
January 2006, the team of researchers
went door-to-door handing out
questionnaires and arranging with
respondents a time to return for the
completed questionnaire. In doing so,
224 October 2007, Vol. 52 No 4
they had to 1) locate subjects on
designated blocks and 2) gain subject’s
consent to participate. Both of these
activities posed unique challenges to
team members as described below.
METHODS
The data discussed in this paper
were collected via in-depth interviews
conducted by the two authors with
eighteen members of the NSF-funded
research team after the primary data
collection period for the NSF-funded
study was over. This interview study
was designed to better understand the
subjective experiences and challenges
faced by the survey team members who
gathered data in a disaster context. The
NSF-study research team consisted of
twenty-one researchers, two of whom
were the authors of this paper. After
receiving IRB approval, the two authors
approached the other nineteen team
members by phone or in-person
requesting permission to interview them
about their experiences in the field
during data collection period of the NSF-
funded study. Of these nineteen team
members, only one could not be
scheduled for an interview. All
interviews with NSF-study research
team members who agreed to participate
took place between late January and
March 2006. Depending on the team
member's location, the interviews were
either done in person or by phone. This
paper also draws on the ethnographic
observations of the authors, both of
whom were members of the research
team.
CHALLENGES OF DATA
COLLECTION
Although the following list of
challenges is not exhaustive, it does
encompass some of the major issues
with which the research team had to
deal.
Locating Subjects. One unique
challenge of doing a survey in a disaster
zone is that the disaster often
dramatically changes the landscape such
that normal navigation and standardized
procedures become problematic.
For example, team members
were given a list of census blocks and
instructed to systematically locate and
give questionnaires to residents still
residing in those blocks. First, arriving
at the designated area was problematic in
that the hurricane had destroyed one of
the main arteries into the study area—
parts of HWY 90 and the Bay of St.
Louis bridge. As the project’s “home
base” was by necessity located in Biloxi,
at one of the few hotels still operating,
the drive to the Waveland/Bay St. Louis
area took much longer than usual,
leaving less time to canvass the blocks.
Teams only went out during daylight
hours due to safety concerns. Once in
the study area, missing street signs and
other landmarks made locating specific
blocks in the study area difficult. Several
team members reported relying on local
residents to orient them. Team members
might also finally locate a designated
block only to find that no structures
remained standing. Although some pre-
canvassing of census blocks was done in
November prior to the start of data
collection in January, the short time
frame to implementation of the NSF-
study and the large number of census
blocks in the sample made pre-
canvassing every block unrealistic.
Second, if structures were found,
canvassing a block and finding subjects
could be problematic. Team members
had to be constantly aware of hazardous
field conditions—debris, ruptured gas
lines, dogs, and insects, among others.
Journal of the Mississippi Academy of Sciences 225
Often team members had to scout around
houses or other properties to locate
possible temporary housing such as
FEMA trailers which were not always
visible. The unusual experience of
walking on foundations, peeking into
windows for clues about whether a
house was occupied, and the general
need for a more vigorous “search” of the
property to find trailers or tents
introduced a heightened concern among
the volunteers that they might be
invading people’s privacy. Some team
members suggested that this led them to
experience emotional exhaustion and
some distress (1).
Team members were also
instructed to call back to each potentially
habitable house two additional times if
subjects could not be initially located.
Although standard procedure, team
members acknowledged that a lot of
time was spent going back to houses of
unclear habitation status.
Gaining Subjects’ Participation
and Consent. Once subjects were
located, team members had to solicit
their participation in the survey. As
Lindsay (2005: 120) argues, while this
process is usually presented as objective
and predetermined, in reality it is
"shaped through interactions with
participants in the field." By January
2006, the relationship between coast
residents and local, state, and national
governments and organizations was
becoming strained. Residents had filled
out multiple forms for FEMA and their
insurance companies for, in some cases,
very little return. Several team members
reported that residents seemed “formed-
out” and were less likely to participate if
they thought the survey was connected
to the state or national government.
Although the written protocol instructed
the research team to simply introduce
themselves as representatives of the
University of Mississippi, many team
members found that additional
clarification was needed to convince
potential subjects that the team was not
associated with any other state or federal
agency. While most survey research
precedes under the assumption that
official sponsorship by the government
increases survey participation, a context
in which “officials” have been
discredited may require more careful
analysis of this assumption (Quarantelli
2002). On the other hand, team
members also indicated that some
individuals expressed gratitude that
someone was listening to them. They
may have seen the survey as an avenue
to voice concerns they believed no one
else was heeding.
One event that may have created
more support for the survey was a
WLOX TV news spot done on the
project toward the middle of the week.
Several team members reported that
residents indicated having seen the spot
and wanted to participate. At least one
team member reported that one subject
who had refused to participate the day
before, changed his mind due to the TV
spot. In a context where there is a high
level of suspicion of people asking them
to fill out forms, TV spots could help by
clarifying researchers’ goals and
affiliations even before they knock at the
front door (for an alternative view see
Quarantelli 2002).
Another barrier to gaining
participation according to team members
was that many residents indicated a lack
of time to fill it out due to rebuilding.
Most team members found that the
original plan of locating subjects in the
morning, handing out the questionnaire,
and picking it up several hours later was
not going to be feasible due to the length
226 October 2007, Vol. 52 No 4
of the survey itself (approximately one
hour to complete) and the fact that
residents had more pressing issues to
address. However, they also found that
if they arranged with a respondent to
pick up the questionnaire the next day,
giving them more time to fill it out, the
respondent may or may not be there and
may or may not have left the
questionnaire in the designated spot. One
team member gave a survey to a man
whose home had been reduced to a slab.
He told the team member that he would
leave the completed survey on a chair
sitting on the middle of the concrete
foundation, but did not end up doing so.
The physical destruction of the hurricane
removed some of the “normal” places
people might feel comfortable leaving a
survey and made arrangements for
retrieval complicated. One positive
aspect of residents’ rebuilding, however,
might be that more residents were home
during the day light hours instead of at
work.
Although team members were
very cognizant of the need to get people
to participate and to follow standardized
procedures, they were also aware of the
emotional, psychological, and physical
issues with which coast residents were
coping. Several team members reported
not wanting to push subjects to
participate given the scope of what they
had already been through. Team
members stated that they wanted to be as
little of a burden as possible to residents,
to not interrupt their work and add to
their distress. Because of this they did
not try to “press” or “convince” residents
to participate after an initial refusal (2).
At the same time, some of the team
members reported that building rapport
required a careful negotiation of the
assumption that surveyors should be
completely objective, and value-neutral
observers. They explained that some of
the coastal residents wanted to discuss
controversial and pervasive political
issues, especially the response of FEMA
and insurance companies to the plight of
residents affected by Katrina. Team
members who obliged felt that it
increased the likelihood that people
would participate.
CONCLUSIONS
Overall, team members reported
that they felt they were fairly successful
in locating subjects and gaining their
participation. At the same time, the
experience could be very frustrating and
emotionally draining. They reported
being careful to follow the standardized
rules of the research project, while
modifying the rules as needed (3).
This study contributes to the
small but growing literature on the
process and context of collecting data
after a disaster by suggesting several
methodological areas that warrant
consideration. It is important to
understand in what ways the physical
destruction of a landscape may require
adjustments to sampling techniques. For
example, if time does not allow more
thorough scouting trips to the area to
confirm where residents and houses still
remain, research teams may need more
training on how to find neighborhoods
and streets no longer marked by street
signs or other landmarks. It is also
important to consider and investigate
how this physical destruction impacts
the physical and emotional experiences
of research team members in such
situations. Last, although it is often
argued that potential subjects are willing
to participate in research studies after a
disaster (Quarantelli 2002; Bourque,
Shoaf, and Nguyen 2002), it is important
to further explore how disaster
Journal of the Mississippi Academy of Sciences 227
conditions affect individuals’ likelihood
to participate in a survey and what
techniques researchers actually use to
gain informed consent and participation
in these situations.
REFERENCES
Bourque, L.B., K.I. Shoaf, and L.H.
Nguyen. Survey research. Pages
157-193 in R.A. Stallings, ed.
Methods of Disaster Research.
Xlibris Corporation, PA.
Glass, J. and H. Frankiel. 1968. The
influence of subjects on the
researcher: a problem in observing
social interaction. The Pacific
Sociological Review, 11, 75-80.
Killian, L.M. 1956/2002. An introduction
to methodological problems of field
studies in disasters. Pages 49-93 in
R.A. Stallings, ed. Methods of
Disaster Research. Xlibris
Corporation, PA.
Knack, J.M., Z. Chen, K.D. Williams, and
L.A. Jensen-Campbell. 2006.
Opportunities and challenges
for studying disaster survivors.
Analyses of Social Issues and Public
Policy, 6: 175-189.
Lindsay, J. 2005. Getting the numbers: the
unacknowledged work in recruiting
for survey research. Field
Methods, 17: 119-128.
Quarantelli, E.L. 2002. The disaster research
center (DRC) field studies of
organized behavior in the crisis
time period of disasters. Pages 94-
126 in R.A. Stallings, ed. Methods
of Disaster Research. Xlibris
Corporation, PA.
Shumsky, A. 1962. The subjective
experience of the researcher.
Journal of Educational Sociology,
36, 134-138.
Stallings, R.A. 2003. Methods of disaster
research: Unique or not? Pages 21-
46 in R.A. Stallings, ed.
Methods of Disaster Research.
Xlibris Corporation, PA.
Footnotes
1. The subjective or emotional experience
of the researcher is an issue usually
glossed over or treated in discussions of
"bias" in reports of survey and other
"objective" research studies. For some
exceptions, see Shumsky (1962), Glass
and Frankiel (1968) and Lindsay
(2005). As Glass and Frankiel (1968)
note, "the notion that as researchers we
can turn off our emotions...is a heritage
which the social sciences have carried
along from the physical sciences...it is a
view which does not well agree with
what we know of human behavior." (P.
78)
2. Quarantelli (2002), in his history of the
Disaster Research Center (now at the
University of Delaware), notes that
decisions such as whether to press for a
particular interview or to seek
information about some sensitive topic
were made in the field and based on the
team's judgment of how pressing the
subject would affect the team's
reputation and ability to make future
contacts (p. 109).
3. For example, although there was a
written protocol that the principle
investigators gave the team members to
read to potential subjects, introducing
the survey and requesting consent to
participate, team members quickly
learned that reading the protocol word
by word was problematic. For example,
one potential subject told a team
member that she did not need to keep
repeating the word "Katrina" as
everyone knew what the name of the
hurricane was.
228 October 2007, Vol. 52 No 4
ASSESSING KATRINA’S DEMOGRAPHIC AND SOCIAL IMPACTS
ON THE MISSISSIPPI GULF COAST.1
1David A. Swanson, 2Rich Forgette, 3Mark Van Boening,
4Cliff Holley, and 5Ann Marie Kinnell(2)
1Department of Sociology and Anthropology , University of Mississippi; 2Department of Political
Science, University of Mississippi; 3Department of Economics, University of Mississippi; 4Center
for Population Studies, University of Mississippi; and 5Department of Anthropology and
Sociology, University of Southern Mississippi
Corresponding Author: David Swanson
ABSTRACT
This paper provides results from a study funded by the National Science Foundation to
examine the effects of Hurricane Katrina on an area of the Mississippi Gulf Coast immediately to
the west and east of St. Louis Bay. This Study Area includes portions of three towns in
Mississippi, Bay St. Louis, Waveland, and Pass Christian. Specifically, the paper describes
selected housing, demographic, and social impacts of Katrina on the Study Area. In regard to
housing and demographic effects, we find that 27% of the housing was destroyed in the Study
Area and 47% significantly damaged. Related to the effects on housing, Katrina caused a 40%
decline in the Study Area’s household population. In regard to social effects, the results of one of
our research hypotheses about the effect of social networks on the well-being of people show that
social isolation significantly increases perceptions of disaster disturbance and decreases perceived
rates of disaster relief. Recommendations (and potential implications for other areas affected by
large-scale disasters) based on our results are provided, as well as descriptions of the Study Area,
study design, and data collection procedures.
INTRODUCTION
According to most measures, the
landfall of Hurricane Katrina on the Gulf
Coast on August 29th, 2005 represented
the greatest natural disaster in American
history. The geographic spread of the
disaster stretched 90,000 square miles,
roughly the size of Great Britain
(Johnson, 2006). In human terms, at
least 1,836 people lost their lives from
Katrina (only 65 did so due to Hurricane
Andrew in August of 1992 and 265 from
Hurricane Camille in August of 1969).
In economic terms, hundreds of
thousands of Gulf Coast residents lost
their homes and jobs. One authoritative
source estimates economic losses at
$81.2 billion in 2005 dollars (Johnson,
2006), nearly double the costs associated
with the next most costly disaster,
Hurricane Andrew ($45 billion in 2005
dollars) and nine times more than
Hurricane Camille ($9 billion in 2005
dollars).
While the preceding numbers are
staggering and likely in the general
ballpark, they are only estimates. As
such, they fall within the tradition of
post disaster assessment in that estimates
of damage and destruction are the norm
(Johnson, 2006). Because of the
ephemeral nature of the data and the
high costs, it is not surprising that
estimates rather than complete counts
Journal of the Mississippi Academy of Sciences 229
are made in regard to the damage from
hurricanes and other large scale
disasters. However, as observed by
Chang (1983), Dynes et al. (1987) and
Smith and McCarty (1996), the lack of
reliable data poses a major problem in
measuring and evaluating the
demographic and economic effects of
major disasters such as hurricanes,
floods, tornadoes, earthquakes, and
volcanic eruptions. It is this problem
that formed one of the two major
objectives of the study we report on here
The other major objective of our
study was to examine the effects of
social networks in regard to mitigating
the effects of a disaster for people. We
followed this line of inquiry because
social science research clearly shows the
importance of social networks in many
activities, including obtaining
employment (Montgomery 1992),
maintaining one’s health (Haines and
Hurlbert, 1992), building safer and
healthier communities (Coleman 1990;
Portes 1998; Putnam 2000) and in
mitigating the effects of unexpected
events such as hurricanes (Haines,
Beggs, and Hurlbert, 2002; Haines,
Hurlbert, and Beggs, 1999; Hurlbert,
Haines, and Beggs, 2000; Hurlbert,
Beggs, and Haines, 2005;
Kirschenbaum, 2004). In designing our
study, we built on this and other
previous disaster-related research by
measuring social networks before,
during, and after Hurricane Katrina.
Thus, we developed temporal measures
of social networks and disaster effects
using primary data of Hurricane Katrina
survivors in our Study Area. Because of
space limitations, we do not provide an
exhaustive report on the results of this
part of our study. However, we do give
an idea of the effects of social networks
by examining one of the social network
hypotheses we examined in our study:
A person embedded in a larger
personal network group will perceive
lower levels of disturbance in his or
her economic, health, and social well-
being than a person in a smaller
personal group network, where
“disturbance” refers to the difference
between responses for “now” (four
months after the hurricane) and those
for “before” (before the hurricane).
DATA AND METHODS
As one of nine “social network”
post-Katrina research projects funded by
the National Science Foundation under
the provisions of the SGER program,3
the recipients of SGER Grant #0555136
(Swanson, Van Boening, and Forgette)
received $96,212 to conduct a study that:
(1) gathered pre- and post-Katrina
information on housing and
population from 573 targeted census
blocks at the epicenter of Katrina’s
impact on the Mississippi gulf coast
that the 2000 census showed as
containing people (the “Short
Form”); and
(2) employed a random start, systematic
selection, cluster sample targeting
126 of these 573 blocks for
administration of a 115-item
questionnaire (the “Long Form”),
such that at least 350 completed
questionnaires would be obtained.
The Long Form was designed for
several purposes, one of which was
to collect retrospective information
on the roles that social and kinship
networks played in determining
respondents’ success (i.e., the
capacity for respondents to sustain
their physical and emotional well-
being after Hurricane Katrina).
230 October 2007, Vol. 52 No 4
The geographic context in which our Study Area is found is provided in Exhibit 1 and the
specific blocks are shown in Exhibit 2.
EXHIBIT 1. THE STUDY AREA AND ITS GEOGRAPHIC CONTEXT
EXHIBIT 2. THE STUDY AREA AND ITS TARGET BLOCKS
Journal of the Mississippi Academy of Sciences 231
The primary data collection team
included faculty and graduate students
from the University of Mississippi,
Mississippi State University, the
University of Southern Mississippi, and
the University of Tennessee Medical
Center (Memphis), as well as several
residents from the MS Gulf Coast. A
secondary team was comprised of
members of the geography division of
the U.S. Census Bureau. This team
geocoded selected sites and assisted with
Short Form data collection. Collectively,
the primary and secondary team
members canvassed the Study Area to
count and assess housing using the Short
Form and to administer the Long Form
Questionnaire.4
The collection of data entailed a
number of operational challenges.5
However, the team was successful in
collecting Short Form data comprised of
10,547 completed surveys from 346 of
the targeted 573 blocks and Long Form
data comprised of 400 completed
surveys from 71 blocks, 68 of which
were from the 126 blocks targeted for
Long Form data collection and three of
which were from Short Form blocks
erroneously canvassed.
The data collection process also
captured information needed to provide
a general assessment of survey data
quality (American Association for Public
Opinion Research, 2000; Dillman,
2000). Using these criteria, our
assessment suggests that the data are of
good quality.
The Short Form
The Short Form contained
identifying information (housing unit
sequence number, block, tract, and as
much information on a street address as
possible) and captured four pieces of
information: structure type (single or
multiple unit dwelling, trailer, mobile
home), whether it was permanent or
temporary, its condition (habitable,
heavily damaged, destroyed), and its
occupancy status (occupied or vacant).
The Short Form was approved for use by
the Institutional Review Board of the
University of Mississippi in the late fall
of 2005. Short Form data were collected
during two periods, January 8th to 15th,
2006 and March 10th to 19th, 2006, with
the bulk of data being collected during
the March 10th to 19th period.
The Short Form data (N=10,547)
represent a complete enumeration of all
housing in the 346 blocks, both
permanent and temporary, as well as a
determination of their condition
(habitable, damaged, or destroyed) and
occupancy status. These 346 blocks
represent portions of two census tracts in
Hancock County, MS (03010 and
03020) and four in Harrison County, MS
(02700, 02800, 02900, & 0300), areas
that were at the epicenter of Katrina’s
Landfall in Mississippi.
Because we used census
definitions and conventions, the Short
Form (and the corresponding control
sheets for the Long Forms in a given
block) allow for a direct comparison of
our housing unit counts with Census
2000 housing unit counts on a block-by-
block basis. From this we also can
account for virtually all housing stock
change between census 2000 (officially,
the date is April 1st) and August 29th,
2005. This allows not only for a
comparison of pre- and post-Katrina
housing, but also pre-and post-Katrina
household populations.
The Long Form
The Long Form was a self-
administered questionnaire containing
115 items regarding sources, constraints,
and assessments of Hurricane Katrina
relief and recovery as well as basic
232 October 2007, Vol. 52 No 4
demographic information, the latter of
which used census definitions and
conventions in the same manner as the
Short Form described earlier. It was
approved by the Institutional Review
Board (IRB) at the University of
Mississippi in the late fall of 2005. Each
block in the Long Form sample, had a
Control Sheet corresponding to the items
found in the Short Form. The Long Form
was informally tested and revised nine
time before a formal pre-test was done in
the field. This field pre-test also allowed
the study team to assess and refine
protocols and procedures associated with
the data collection effort.
The primary dependent variables
that can be derived from questions in the
Long Form are the differences in a
survivor’s responses of his or her
economic, health and social well-being.
To measure differences, respondents
were asked to retrospectively and
prospectively assess their satisfaction-
levels in addition to stating their current
perceptions.
The Long Form also contained a
Post-Traumatic Stress Scale as well as
items that allow us to control for other
variables affecting post-disaster
assessments. These variables include
the level of property loss, access to
insurance, the amount and sources of
governmental relief (FEMA, National
Guard, state and local emergency
management), as well as ascribed (e.g.,
age, sex) and achieved characteristics
(e.g., income, education).
Controlling for these alternative
explanations, we could test many
specific hypotheses, one of which we
present later in this paper.
As stated earlier, it was
administered to a representative sample
comprised of 126 targeted blocks of the
total of 573 in the Study Area. Seventy
one of these blocks were found to
contain habitable housing. Team
members went door-to-door handing out
questionnaires and arranging with
respondents a time to return for the
completed questionnaire. A minimum of
two callback attempts was made at each
housing unit canvassed that potentially
was occupied, including damaged
permanent units and all temporary units.
The Long Form data were
collected January 8th to 15th, 2006, with
mail–out/mail-back callbacks collected
from January 8th to February 15th. Four
hundred completed Long Forms were
obtained from canvassing and callbacks.
KATRINA’S HOUSING AND
DEMOGRAPHIC IMPACTS
As shown in Table 1.a, our data
indicate that just before Katrina stuck,
there were 8,535 (permanent) housing
units in the 346 blocks we canvassed, an
increase of nearly 10% over the Census
2000 count of 7,793 (Table 1.b). Table
1.a also shows that of the 8,535 housing
units, 2,227 (27%) were destroyed and
3,997 substantially damaged (47%),
leaving 2,261 habitable (26%).6 Table
1.c shows that 2,012 temporary units
were in the Study Area after Katrina
struck, of which 94% were occupied.7
Journal of the Mississippi Academy of Sciences 233
TABLE 1a. 2006 SPECIAL CENSUS DATA
County/Tract
Total Housing
Units
(Permanent)
Habitable
Permanent
HUs
Occupied
Habitable
Permanent
HUs
Damaged
Permanent
HUs
Occupied
Damaged
Permanent
HUs
Destroyed
Permanent
HUs
Harrison/0027 2,035 1,331 1,198 651 103 53
Harrison/0028 698 53 46 232 20 413
Harrison,/0029 741 35 24 166 8 540
Harrison/0030 1,479 32 19 763 0 684
Hancock/0301 2,721 519 446 1,777 43 425
Hancock/0302 861 291 174 408 2 162
GRAND TOTAL 8,535 2,261 1,907 3,997 176 2,277
For definitions see endnote # 3.
TABLE 1b.2000 DECENNIAL CENSUS DATA
County/Tract
Total Housing
Units
(Permanent)
Occupied
Housing Units
(Permanent) Vacant Housing
Units (Permanent)
Harrison/0027 1,788 1,666 122
Harrison/0028 660 548 112
Harrison,/0029 670 353 317
Harrison/0030 1,361 1,086 275
Hancock/0301 2,528 2,177 351
Hancock/0302 786 656 130
GRAND TOTAL 7,793 6,486 1,307
For definitions, see endnote # 3.
TABLE 1c. 2006 SPECIAL CENSUS DATA
County/Tract Temporary
HUs
Occupied
Temporary
HUs Total Occupied HUs
(Perm & Temp.)
Harrison/0027 245 237 1,538
Harrison/0028 38 34 100
Harrison,/0029 29 26 58
Harrison/0030 425 422 441
Hancock/0301 805 745 1,234
Hancock/0302 470 436 612
GRAND TOTAL 2,012 1,900 3,983
For definitions, see endnote # 4.
234 October 2007, Vol. 52 No 4
There were approximately 16,540
people residing in 6,486 (occupied)
permanent housing units in the 346
blocks as of Census 2000. Just prior to
the impact of Katrina on August 29th,
2005, there were approximately 7,100
occupied permanent housing units (83%
of the total number of permanent
housing units) containing 18,105 people
in these same 346 blocks. By January of
2006, we found approximately 10, 950
people residing in 3,938 permanent and
temporary housing units in these same
346 blocks. At the time of Census 2000
and just prior to when Katrina struck, the
average number of persons per
household (PPH) in the Study Area was
2.55. Subsequent to Katrina the PPH was
2.78. Thus, for the 346 blocks
comprising our Study Area, Hurricane
Katrina resulted in:
(1) a decline of 7,155 for the
household population – a 40%
drop from the pre-Katrina
household population of 18,105;8
and
(2) an increase of 0.23 persons per
household– a 9% increase from
the pre-Katrina PPH of 2.55.
Our estimates of the effects of
Katrina on the household population in
the Study Area are consistent with the
special estimates of Hancock and
Harrison counties that the Census
Bureau released for January of 2006.
These estimates were designed to show
the impact of Katrina in the 117 counties
designated by the Federal Emergency
Management Agency (FEMA) as being
eligible for individual and public
assistance (U.S. Census Bureau, 2006).
For Hancock County, the Census Bureau
estimated that Katrina caused a 24%
decline in the household population (As
of July 1st, 2005 the Census Bureau
estimated the population of Hancock
County to be 46, 240 and for January of
2006, it estimated that the estimated the
household population was 35, 129) and
for Harrison County, it estimated a
16.5% decline (As of July 1st, 2005 the
Census Bureau estimated the population
of Harrison County to be 186,530 and
for January of 2006, it estimated that the
household population was 155,817). We
say that our numbers are consistent with
the Census Bureau’s because our Study
Area is in the portions of Hancock and
Harrison counties that received the brunt
of Katrina’s impact. As such, our
estimate of a 40% decline in the
household population for the Study Area
is consistent with the Bureau’s estimates
of 24 and 16% declines for all of
Hancock and Harrison counties,
respectively.
THE EFFECT OF SOCIAL
NETWORKS ON KATRINA’S
IMPACT
Beyond counts of housing and
population, there is also the question of
how an individual’s economic, health,
and social well-being is affected by a
disaster such as Katrina. Why are some
individuals better able than others to
respond and recover in the immediate
aftermath of major disasters? Does the
economic, educational, racial or
experiential (previous catastrophic event
experiences) status of a particular
refugee largely determine an individual’s
perceptions of relief and recovery?
To assist in answering these
questions, we turned to Social Network
Theory, which offers an interdisciplinary
framework for understanding the health
and resilience of communities (Coleman
1990; Portes 1998; Putnam 2000). Thus,
in our research, we applied social
Journal of the Mississippi Academy of Sciences 235
network theory to generate hypotheses
addressing these questions. Our
hypotheses were designed to clarify the
conditions under which some individuals
are better able to solve immediate
problems presented by a natural disaster
and to create a strong perception of
eventual community recovery. We first
examined the size and scope of the
respondents’ social networks and then
analyzed whether or not being embedded
in a social network affected the
respondents’ post-Katrina well-being.
Exhibit 3 shows that gender,
race, and church attendance are
correlated with the number of formal
group network memberships. Females,
whites and frequent church attendees
average about twice as many formal
group network memberships as males,
minorities, and infrequent church
attendees, respectively. These respective
differences were statistically significant
(α <.05).
Exhibit 4 shows a second
dimension of relative social isolation
among Hurricane Katrina survivors: the
average number of people in a personal
group network by selected personal
characteristics. As Exhibit 4 indicates,
gender and employment are correlated
with the size of a personal group
network. Females and employed
persons are in larger personal group
networks than males and unemployed
persons, respectively. Again, these
respective differences were found to be
statistically significant (α <.05). The
difference between the white
respondents and minority respondents is
not statistically significant (α =.083).
EXHIBIT 3. SIZE OF GROUP NETWORKS BY GENDER, RACE, AND CHURCH
ATTENDANCE
Size of Group Networks
0 0.5 1 1.5 2 2.5
Male
Female
White
Minority
Attend Church: Never or Yearly
Attend Church: 1-2 Times a Month
Attend Church: Weekly
Attend Church: More than once a week
Total Number of Group Membership
236 October 2007, Vol. 52 No 4
EXHIBIT 4. SIZE OF PERSONAL NETWORKS BY GENDER, RACE, AND
EMPLOYMENT STATUS
Size of Personal Networks
01234567
Male
Female
White
Minor ity
Employed
Unemployed
Number of Networks
Exhibit 5 shows the general
composition of these survivors’ personal
group networks. As it shows, about 90%
of survivors’ self-reported networks are
“friends,” “immediate family” or
“extended family members.”
As stated earlier, the specific
hypothesis we report on here that a
person embedded in a larger personal
network group will perceive lower levels
of disturbance in his or her economic,
health, and social well-being than a
person in a smaller personal group
network. We examined this hypothesis
using ordinary least-squares (OLS)
multiple regression (Tabachnick and
Fidell, 1996: 127-193). The results of
our analysis are shown in Table 2. The
dependent variables measure six types of
disturbances perceived by respondents.
These disturbances are in terms of their
financial, economic, psychological, and
physical well-being, as well as their
personal and professional relationships.
For each of these six types of
disturbances, we used our measure of the
respondent’s reported well- being or
situation prior to Katrina and our
measure of the respondent’s reported
well-being or situation after Katrina. Our
measure of disturbance is the difference
between the six sets of pre-Katrina and
post-Katrina measures, respectively.
Looking at Table 2, what is the
effect of personal group network size on
survivors’ perceptions of Hurricane
Katrina’s impact while controlling for
the effects of other (independent)
variables? Our results indicate strong
support for the hypothesis that a person
embedded in a larger personal network
group (shown as “Personal Networks” in
Table 2) will perceive lower levels of
disturbance in his or her economic,
Journal of the Mississippi Academy of Sciences 237
health, and social well-being than a
person in a smaller personal group
network The Personal Networks
coefficients are negative and statistically
significant for financial, economic,
psychological well-being, as well as that
of professional relationships; there is no
statistically significant effect of Personal
Networks on physical well-being and on
personal relationships.
Overall, Damage and Personal
Networks are the most consistently
statistically significant of the
independent variables across the six
multiple regression models shown in
Table 2. These two variables have the
expected signs: Damage is positively
associated with disturbance (more
damage increases the disturbance) and
Personal Networks is negatively
associated with disturbance (larger
personal group networks lessen the
disturbance).
EXHIBIT 5. NETWORK COMPOSITION
Network Composition
0
5
10
15
20
25
30
35
40
45
Friends
Imme di ate F amily
Extended Family
Spouse
Co-Worker
Neighbors
Other
Percentage of Total Network
238 October 2007, Vol. 52 No 4
TABLE 2. RESULTS OF TEST OF THE HYPOTHESIS
Independent Variable
Model Characteristics
Disturbance
(Dependent
Variable)
Damage
Insurance Govt.
Aid Private
Aid Personal
Networks
Injuries
Constant
R2
N
Financial .484***
(.115)
.129
(.108)
.153
(.100)
-.065
(.076)
-.053**
(.021)
----
-1.90***
(.393)
.07
335
Economic .176*
(.104)
.129
(.097)
.160*
(.090)
-.055
(.070)
-.040**
(.019)
----
-2.38***
(.352)
.03
332
Psychological .208**
(.099)
.088
(.091)
-.010
(.085)
-.068
(.065)
-.030*
(.018)
----
-1.43***
(.339)
.02
335
Physical .321***
(.092)
-.025
(.086)
.039
(.080)
-.058
(.061)
-.028
(.017)
.006
(.206)
-1.23***
(.397)
.05
335
Personal
Relationships .131
(.085)
-.078
(.080)
-.089
(.074)
-.045
(.056)
-.009
(.015)
----
-.431
(.289)
.02
338
Professional
Relationships .210**
(.104)
-.036
(.097)
.020
(.089)
.003
(.069)
-.045**
(.019)
----
-.819**
(.358)
.03
287
The number in the upper part of each cell under the Independent Variables represents the OLS Multiple
Regression coefficient for the Independent variable in question relative to the Disturbance in the same row.
The number in parentheses in each of the same cells is the standard error associated with the coefficient.
Statistical significance of each coefficient (or lack thereof) is shown by the number of asterisks, where:
*** is statistically significant at the .01 level; ** is statistically significant at the .05 level; and * is
statistically significant at the .10 level. Lack of statistical significance is indicated by a lack of asterisks.
DISCUSSION
What are the housing and
demographic effects of a major disaster?
Our research shows that they are
substantial, with 27% of the pre-Katrina
housing stock destroyed and 47%
damaged, it is not surprising that we find
a 40% decline in the Pre-Katrina
household population. Neither should it
be surprising that 48% of the occupied
post-Katrina housing stock consisted of
temporary units (Table 1.a and Table
1.b) and that there was a slight increase
in the average number of persons per
household after Katrina. Given the
magnitude of these measurements, we
concur with the recommendation by
Henderson et al. (forthcoming) that
methods need to be developed to quickly
and accurately assess the TOTAL
housing and demographic effects of
large scale disasters like Hurricane
Katrina, particularly at their epicenters.
What about social networks? Our
results point toward a revised
understanding of who is at-risk and who
recovers from disasters. Vulnerable or
at-risk populations are typically defined
by personal or physical attributes that
include an individual’s socio-economic
status, employment, disabilities, age,
housing quality, availability of personal
transportation (National Research
Council, 2007). We find that social
networks are an important part of a
person’s attributes that should be
considered in better understanding who
is at risk. Consequently, we suggest that
the methods for identifying “at-risk and
vulnerable” populations found in the
National Research Council’s 2007 report
be extended to include community-based
assessments of social networks.
Understanding the spatial or geographic
correlates of socially isolated disaster
survivors may allow governmental and
non-governmental emergency manage-
Journal of the Mississippi Academy of Sciences 239
ment teams to better target relief and
recovery efforts. Finally, we suggest
that governmental emergency
management needs to be sensitized to
the importance of informal networks in
ameliorating various types of
disturbances associated with natural
disasters, which suggests that they also
play an important role in relief and
recovery.
REFERENCES
American Association for Public
Opinion Research. (2000).
Standard Definitions: Final
Dispositions of Case Codes and
Outcome Rates for Surveys.
Lenexa, KS: AAPOR.
Chang, S. (1983). Disasters and Fiscal
Policy: Hurricane Impact on Municipal
Revenue. Urban Affairs Quarterly 18:
511-523.
Coleman, J. (1990). Foundations of
Social Theory. Cambridge, MA:
Harvard University Press.
Dillman, D. (2000). Mail and Internet
surveys: The Tailored Design
Method, 2nd Edition. New York,
NY: Wiley.
Dynes, R., B. deMarchi, and C. Pelanda.
(1987) Sociology of Disasters:
Contribution of Sociology to
Disaster Research. Milan, Italy:
Franco Angeli.
Haines, V., Beggs, J., & Hurlbert, J.
(2002). Exploring the structural
contexts of the support process:
Social networks, social statuses,
social support, and psychological
distress. Advances in Medical
Sociology, 8, 269-292.
Haines, V., & Hurlbert, J. (1992).
Network range and health.
Journal of Health and Social
Behavior, 33, 254-266.
Haines, V., Hurlbert, J., & Beggs, J.
(1999). The disaster framing of
the stress process: A test of an
expanded model. International
Journal of Mass Emergencies
and Disasters, 17, 57-78.
Henderson, T. L., D. Banks, A. Chia-
Chen Chen, C. Airriess, D.
Swanson, and M. Sirois (
Forthcoming) After a Disaster:
Lessons in Survey Methodology
from Hurricane Katrina.
Population Research and Policy
Review.
Hurlbert, J. S., Haines, V. A., & Beggs,
J. J. (2000). Core networks and tie
activation:
What kinds of routine networks
allocate resources in nonroutine
situations? American Sociological
Review, 65, 598-618.
Hurlbert, J., Beggs, J., & Haines, V.
(2005). Bridges over troubled
waters: What are the optimal
networks for Katrina’s victims?
Unpublished manuscript. Available
online at
http://understandingkatrina.ssrc.org
/Hurlbert_Beggs_Haines/, last
accessed, June 2007.
Johnson, D. L. (2006). Service
Assessment, Hurricane Katrina
August 23-31, 2005. Silver Spring,
MD: U.S. Department of
240 October 2007, Vol. 52 No 4
Commerce, National Oceanic and
Atmospheric Administration,
National Weather Service.
www.weather.gov/s/assessments/p
dfs/katrina.pdf, last accessed April
12, 2007)
Kirschenbaum, A. (2004). Generic
sources of disaster communities: A
social network
approach. International Journal of
Sociology and Social Policy,
24(10/11), 94-129.
Montgomery, James D. (1992). “Job
Search and Network Composition:
Implications of the Strength-of-
Weak-Ties Hypothesis.” American
Sociological Review 57: 586-96.
National Research Council. (2007).
Tools and Methods for Estimating
Populations at Risk from Natural
Disasters and Complex
Humanitarian Crises. Washington,
D.C.” The National Academies
Press.
Portes, A. (1998). Social capital: Its
origins and applications in modern
sociology. American Review of
Sociology, 24(1), 25.
Putnam, R. (2000). Bowling Alone: The
Collapse and Revival of
American Community. New
York, NY: Simon and Schuster.
Smith, S. and C. McCarty. (1996).
“Demographic Effects of Natural
Disasters: A Case Study of
Hurricane Andrew.”
Demography 33: 265-275.
Swanson, David. A. and G. E. Stephan.
(2004). The Methods and
Materials of Demography, 2nd
Edition. New York, NY:
Elsevier Academic Press.
Tabachnick, B. and L. Fidell (1996).
Using Multivariate Statistics, 3rd
Edition. New York, NY: Harper
Collins College Publishers.
U.S. Census Bureau (2006). Special
Population Estimates for
Impacted Counties in the Gulf
Coast Area
(http://www.census.gov/PressRelease/w
w/2005/katrina.htm, last accessed
June 2007).
ENDNOTES
1. This material is based upon work supported by the National Science Foundation under Grant No.
0555136, awarded to the University of Mississippi (D. Swanson (Sociology & Anthropology), PI; Mark
Van Boening (Economics) and Rich Forgette (Political Science), Co-PIs). Any opinions, findings, and
conclusions or recommendations expressed in this material are those of the author(s) and do not
necessarily reflect the views of the National Science Foundation.
2. David A. Swanson, Department of Sociology and Anthropology , University of Mississippi; Rich
Forgette, Department of Political Science, University of Mississippi; Mark Van Boening, Department of
Economics, University of Mississippi; Cliff Holley, Center for Population Studies, University of
Mississippi; and Ann Marie Kinnell, Department of Anthropology and Sociology, University of Southern
Mississippi
Journal of the Mississippi Academy of Sciences 241
3. The Acronym “SGER” stands for “Small Grants for Exploratory Research.” Very soon after Katrina
struck the Mississippi Gulf Coast, The National Science Foundation issued a call for “SGER” grants to
assess its impact. There were nine SGER grants that used a social network perspective. The grantees
using a social network perspective were assembled at an NSF-funded conference organized by RAND
that took place in Arlington, Virginia August 17-18, 2006. Papers from this conference are forthcoming in
a special issue of Population Research and Policy Review (volume 27, no. 6, December, 2008).
4. In addition to the authors, the study team included:
University of Mississippi
Stefan Schulenberg, (Psychology Department, faculty)
Jerry McKibben, (Center for Population Studies, Adjunct Faculty)
Kirsten Dellinger(Sociology and Anthropology Department, faculty)
Jeff Jackson (Sociology and Anthropology Department, faculty)
Ed Sharpe (Sociology and Anthropology Department, adjunct faculty)
Johnny Ducking (Economics Department, graduate student)
Bryan Dettrey (Political Science Department, graduate student)
Chrissy Glider (Economics Department, graduate student)
Mike Hirschel (Psychology Department, graduate student)
Gwen Wages (Sociology and Anthropology Department, graduate student)
Jennifer Sukanek (Sociology and Anthropology Department, undergraduate student)
Mississippi State University
Amie Brushaber (Sociology, Anthropology and Social Work Department, graduate student)
Ian Monaghan (Social Science Research Center, research assistant)
University of Southern Mississippi
Barbara Hester (Anthropology and Sociology Department, Graduate student; resident of Pass Christian,
MS)
Brooke Roberts (Anthropology and Sociology Department, Graduate student, resident of Jackson County,
MS)
University of Tennessee Health Science Center, Memphis, Tennessee
Rick Thomas (Department of Preventive Medicine, faculty)
Community
Mary Ellen Calvert (interviewer, Long Beach, MS; alumna of the University of Mississippi)
Rita Swanson (Volunteer, first aid support for Field Work Team, Oxford, MS; RN, BSN, MSN)
Census Bureau Personnel (GPS recording of housing units and sites of housing units)
Greg Hanks (Team Leader)
Steve Bainter
Sharon Cochran
Ross Davis
Kevin Donnalley
Jennifer Harrop
Jennifer Holland
John Kennedy
All primary study team members completed training models on gathering data from human
subjects and were certified through their respective Institutional Review Boards. The secondary team
242 October 2007, Vol. 52 No 4
members completed similar training under the provisions of Title 13 as part of their normal work with the
US Census Bureau. Training specific to the data collection (both the Short Form and the Long Form) was
done at the University of Mississippi and Mississippi State University before the Pilot Study was done on
January 7th, 2006. The Pilot Study also served as On-Job-Training and a final test of procedures and
protocols. Data collection training included modules on dealing with a distressed population.
5. Operational challenges included, among others: (1) an extremely limited number of locations to use as
a “home base,” which resulted in long driving times to the field and, in turn, limited the time available for
team members to canvass blocks; (2) a lack of street signs and other landmarks to use to locate specific
blocks in the study area; (3) hazardous field conditions – debris, ruptured gas lines, and insects; and (4) a
lack of facilities for food, gas, and sanitation.
6. The definition of a housing unit follows that of the Census Bureau’s definition as used in the 2000
Decennial Census. However, the Census Bureau has no definition for a “damaged” or “destroyed”
housing unit. Given the intent of our study, we needed such a definition. Therefore, we defined a
“damaged housing unit as one that had received observable damage, but was still standing and appeared
to be structurally sound. For example, a house with a blue tarp for a roof and all of the doors, windows,
and interior walls missing was defined as damaged. A Destroyed house was one that was either
completely gone (e.g., only a slab remained) or sustained structural damage (e.g., supporting beams for
the roof had collapsed, a wall was caved in). In cases where it was difficult to distinguish whether a house
was damaged or destroyed, we classified it as damaged.
7. The Census Bureau does not distinguish between a temporary and permanent housing unit.
Specifically, the Census Bureau defines a housing unit as a shelter intended for “separate use’ by its
occupants such that there is independent access to the outside and the shelter is not a group quarters
(Swanson and Stephan, 2004: 762). Given the intent of our study we needed to identify temporary
housing units. Therefore, we defined temporary housing units using the following protocol. First, we
defined as temporary housing units, any non-permanent structure in which people were residing. This
included tents, lean-to, campsites, motor vehicles, recreational vehicles, travel trailers, house trailers and
mobile homes with their axles and wheels in place. The recreational vehicles, travel trailers, house
trailers, and mobile homes classified as temporary housing units generally were on lots next to destroyed
or damaged permanent housing units or in parks and usually were connected to power and other utilities.
In such cases, even if they were not occupied, we counted them as temporary housing units. If we
encountered tents, cars, and trucks that were not occupied, we did not count them as housing units.
Similarly, if we encountered un-occupied recreational vehicles, travel trailers, house trailers, and mobile
homes on sales lots we did not count them (these were usually either heavily damaged or destroyed
anyway)
8. The household population is comprised of those who live in housing units (as opposed to those who are
homeless or living in group quarters – prisons, long-term care hospitals, military barracks, and school and
college dormitories (Swanson and Stephan, 2004: 762).
Journal of the Mississippi Academy of Sciences 243
THE ROLE OF INFORMATION AND COMMUNICATION
TECHNOLOGY (ICT) IN THE RESILIENCE OF EDUCATIONAL
INSTITUTIONS IN THE WAKE OF HURRICANE KATRINA
1Bertta Sokura and 2Arthur G. Cosby
1Department of Business Technology, Information Systems Science
Helsinki School of Economics and 2 Social Science Research Center, Mississippi StateUniversity
Corresponding Author: Bertta.Sokura@hse.fi
ABSTRACT
In this paper we examine the role of information and communication technology in the
resilience of two educational institutions in the wake of hurricane Katrina. More specifically, we
primarily focus on trying to understand the roles played by ICT from the perspectives of the IT
department, the registrar’s office and instruction. Also some recommendations how universities
can use ICT in disaster recovery are provided. We use case study method that is both qualitative
and exploratory in nature. The theoretical framework is drawn from the concepts of business
process re-design, IT capabilities, and intellectual capital. This paper draws on qualitative results
from the data generated by interviews with the following: (1) a department chair or teaching
professional; (2) a representatives of IT Administration; and (3) a representative of the registrar’s
office. The results document the significance of ICT in the survivability and resilience of these
institutions after Katrina.
INTRODUCTION
The fall semester began for many
southern universities only one week before
Hurricane Katrina struck the Gulf Coast on
August 29, 2005. Hurricane Katrina caused
massive damage and destruction along the
coastlines of Louisiana, Mississippi, and
Alabama. It is estimated that Katrina was
responsible for $81 billion in damages
(Johnson, 2006). The storm caused not only
property damage, power outages, and fuel
shortages but killed at least 1,836 people
(Johnson, 2006). So far, the storm has left
many people homeless and families
separated. Eighty percent of the city of New
Orleans was flooded and over one million
people from the Gulf Coast region were
displaced. Among those displaced were
over 50,000 students forced to evacuate
from the city’s campuses. In this situation,
every university in New Orleans closed its
on-campus operations for the 2005 fall term.
The leadership of the universities in
New Orleans was profoundly stressed in
keeping their universities viable.
Immediately after Katrina, the public was
barred from returning to New Orleans by the
National Guard and local officials. During
this period, University of New Orleans’s
Chancellor Timothy Ryan made a special
request to the Coast Guard to escort him
onto the university’s campus to retrieve
certain very important items. The Coast
Guard complied with the request and
transported him to the campus via Coast
Guard vessel and armed escorts. What was
so important to the Chancellor that in the
midst of one of the nation’s greatest
disasters that he felt compelled to make this
244 October 2007, Vol. 52 No 4
extraordinary visit to a university campus?
The answer was that he needed to retrieve
the university’s computer servers and, of
course, the extensive information about the
university, its students, its faculty and its
employees. Think of the personal and
institutional damage that could occur if
information about student records, faculty
employment, and diplomas awarded were
forever lost. The university, of course, did
have IT back-up but it was located in
another area of New Orleans that was not
accessible at this time during the flooding.
The servers were retrieved and relocated to
Baton Rouge. The information was saved
and the university began to reestablish itself
by its web presence. Significantly, the
university was able to make its mid-
September payroll within two days of its
standard payroll procedures. Clearly,
modern information technology was playing
a critical role in the survivability and
resilience of the University of New Orleans.
Many disaster related studies have
focused on reporting “bad news” such as
financial losses, physical damages or
negative impacts on mental or physical
health (Freudenburg, 1997; Gill, 1998;
Picou, Marshall, and Gill, 1997;Chappell et
al, forthcoming). The study that we report
here, however, might be interpreted as a
success story for Information and
Communication Technology (ICT) in that it
is a case study that investigates the role that
ICT played in mitigating the effects of
Hurricane Katrina on two of the universities
in New Orleans. Is it possible that
information technology is so important in
responding to disasters that it can be
considered a new form of “intellectual
capital” that transcends, or at least extends,
the traditional capabilities of human,
financial, and infrastructure capital?
In our case study, we primarily focus
on trying to understand the roles -
facilitating, enabling or IT-driven - played
by ICT from the perspectives of three
important university activities: (1) technical
support; (2) the registrar’s office; and (3)
instruction. As a secondary goal, we will
elaborate on how universities can use ICT in
disaster recovery - very little has been done
on the effects of disasters on organizations
and even less on institutions of learning.
This paper is divided into four parts.
First, we offer a short review of Intellectual
Capital and ICT-related models and theories.
This serves as the analytic framework for
our research. Second, we present our
methodology and analysis. Third, the key
results are .provided. And finally, we
discuss and summarize the results of our
research and give some recommendations on
how ICT can be used by organizations in
disaster recovery.
INTELLECTUAL CAPITAL AND IT
CAPABILITIES
When we were planning this study,
we hypothesized that intangible
organizational assets would play a role in
disaster recovery even though tangible
organizational assets such as buildings and
other infrastructure were damaged or
destroyed. In fact, we anticipated that ICT
would become even more important when
physical infrastructure became unavailable
or unusable. This line of reasoning stems, in
part, from research reported by, among
others, Beggs, Haines, and Hurlbert (1995),
Lin, Cook, and Burt (2001). One way to
think about intangible assets is in terms of
Intellectual Capital (MERITUM, 2001;
Sullivan, 1998). The concept of Intellectual
Capital can be approached from the
perspective of value creation or value
extraction (Sullivan, 2000, 184). The value
creation aspect points out the human effort
in creating innovations, formulating and
sharing knowledge. The typical activities in
value creation are education, knowledge,
innovations, building organizational
Journal of the Mississippi Academy of Sciences 245
structures, the development of the
interaction among customers, organizations,
and individuals, and the management of
values and organizational culture (Sullivan,
1998, 20). In this study, we use the value
creation approach, where the primary
objective is to improve the knowledge and
skills of an organization’s personnel. This is
primarily accomplished through professional
development. Typical activities include
knowledge creation and sharing, learning,
and organizational dynamics, and the
development of information systems
(Sullivan, 2000, 184).
Intellectual capital can be divided
into three main dimensions (MERITUM,
2001): (1) human capital; (2) structural
capital, and (3) relational capital. Human
capital includes the competencies and
capabilities of the personnel, their
motivation, commitment and interaction.
Structural capital consists mainly of
organizational structures, processes,
management, culture, values and
information systems and other items related
to these. Conceivable features of relational
capital are all the potential network related
issues, like customers, partners, and other
stakeholders. In our study, we view
relational capital as consisting mainly of
faculty and staff contacts with counterparts
in other institutions via professional and
other associations, alumni, and
governmental bodies. Students can be
viewed both as structural and relational
capital. In our view, we examine them more
from the perspective of relational capital.
It is useful to note that Saint-Onge
also argues that value can be created only
when all three forms of intellectual capital,
human, structural, and relational are
integrated (Edvinsson - Malone, 1997). In
terms of such integration, the role of
knowledge management can be thus seen as
an integrating mechanism that pulls the three
forms of intellectual capital into closer
interaction with each other. Figure 1
describes this “value creation” process.
Figure 1. MERITUM model of intangible assets and knowledge management (model was created
by Saint-Onge, Armstrong, Petrash and Edvinsson and originally presented as value creation
platform (Edvinsson and Malone, 1997)
The role of ICT in mitigating the
effects of a disaster can be approached in
several ways. We can take the perspective
of management, for example, or a more
technical approach, one based primarily on
information technology (See Zmud, 1988;
Schein, 1989; Sääksjärvi, 2000).
Significantly common information
Relational
capital
Human
capital Structural
capital
ValueValue
Knowledge
management
246 October 2007, Vol. 52 No 4
technology can be seen as an agent of
empowerment in an organization. Thus, it
has the capacity to be a vehicle for change in
an organization’s fundamental relationships
with its stakeholders, both internal and
external.
ICT can serve as an agent of
empowerment through the different ways. It
can be used to redesign organizational
functions and processes. Davenport (1990),
for example, classifies these ways into three
categories: (1) facilitating; (2) IT-enabled
redesign; and (3) and IT-driven process
redesign. Using this classification scheme,
Hannus (1994) goes on to describe three
different roles in business process re-design,
which we use in our case study. Table 1
provides a description of these three roles.
Table 1: Three roles of ICT modified from Hannus (1994)
ROLE OF ICT DESCRIPTION LEVEL OF AMBITION
Facilitating role ICT is utilized in
implementing of the
separately defined strategy
Improvement; the
activity itself has not
been questioned
Enabling role The core processes are
redesigned by utilizing
innovatively the possibilities
of ICT
Re-engineering
Driving role Major changes in processes
and mission on the basis of
ICT
Rethinking / reinventing
Traditionally, information technology
has been seen to have a facilitating role. The
basic idea is that the functional needs of an
organization, such as keeping student
records and processing payrolls are first
defined, so that the activities required to
support these functional needs are then
developed and implemented. One
shortcoming of this approach is that it fails
to recognize the reciprocal interaction
among processes, which often serve as the
potential enabling agents of ICT (Hannus,
2000, 109).
Bharadwaj (2000) defines the
capability of information technology based
on the definition of a resource-based view
(RBV) by Barney (1991) as follows: “A
firm’s IT infrastructure, its human IT skills,
and its ability to leverage IT for intangible
benefits serve as firm-specific resources,
which in combination create a firm-wide IT
capability”. These ideas are also similar to
those proposed by Langdon (2006). The IT
infrastructure means the tangible resources
consisting of the physical infrastructure
components such as the computer and
communication technologies and shareable
technical platforms, and databases. Human
IT-skills refer to the human resources
comprising the technical and managerial
skills. This definition is analogous to that of
IC (Intellectual Capital) where St. Onge
(Edvinsson and Malone, 1997) argues that
value can be created only when all three
forms of IC are integrated. The RBV theory,
however, states that the resources and
performance of a firm are linked in a way
that the unique resources and skills are
organization specific, valuable, and rare
(Barney, 1991). Thus the combination of IT
infrastructure, human IT skills, and the
ability to leverage IT for intangible benefits
creates the school-wide IT capability and
Journal of the Mississippi Academy of Sciences 247
increases the performance as the dimension interact. (See Fig. 2).
METHODS AND DATA
In our study, we use the case study method
(Broder et al., 2003). This case study
method has a long and rich history among
institutions engaged in education, training
and professional development. We selected
the “historical narration form” of the case
study method (see e.g. Patton and Swanson,
2003), which is well suited for our
objectives. In our case study, we
strategically selected two universities, “A”
and “B,” which are of interest not only
because of their unique features, but also for
the potential of obtaining important
knowledge about the “lessons
learned”(Stake, 2000) in the aftermath of
Katrina.
Our data were generated by interviews with
the following: (1) a department chair or
teaching professional; (2) a representative of
IT Administration; and (3) a representative
of the registrar’s office. The interviews
were done by telephone and conducted
during June and July of 2006. The questions
used in the interviews were common across
each of the institutions but varied by type of
representative. That is, the interview with a
representative of IT administration was
different than the interview with a
representative of the registrar’s office, and
both were different than the interview with a
department chair or teaching professional.
All interviews were semi structured and
open-ended. Questionnaires were sent to
respondents prior to the telephone interview.
The interviews were recorded and
transcribed. Table 2 provides a description
of this process.
Ability
to leverage IT
for intangible benefits
IT-infastructure Human IT skills
PerformancePerformance
These three
dimensions of IT
capability in
combination
serve as school-
specific
recourses
Figure 2. Visualization of a school-wide IT capability
248 October 2007, Vol. 52 No 4
Table 2. Background of the interviewed experts.
Background Number of
interviews
University A
- Chair of department
- Assistant professor
- Professor
- IT director
- Director of RO
1
1
1
1
1
University B
- Chair of department
- IT professional
- Director of RO
3
1
1
Total 10
The data used in this study initially covered
three educational institutions located in New
Orleans. The first phase of data collection
was the selection of the case organizations.
The criteria for selection were that all these
universities were located in the destroyed
area and they both had different “survival”
strategies. In the second phase a conference
call with the representatives of each
university was conducted in order to identify
the most important things that they perceive
as being the foundations for recovery from
disaster. After that we designed and pre-
tested questions. Three different sets of
questions were created. Then we chose the
interviewees one from IT and RO and three
from instruction. The questionnaires were
sent to respondents and phone interviews
conducted. The interviews were recorded
and transcribed. The analysis involved the
classification of responses, first by themes
and finally by activities. The third
university was deleted from the study
because we were unable to arrange
interviews. The analysis is described in
more detail in Appendix A.
University A was established in the
mid 19th Century and is one of 28 Jesuit
institutions of higher learning in the United
States. Although a Jesuit university, it is
open to students of all faiths and in its
mission statement, welcomes students of
diverse backgrounds. The leading academic
areas of this institution are communications,
music and religion and in the academic year
prior to Katrina, it had a total enrollment of
5,900 students, of which 3,800 were
undergraduates. The students come from all
50 states, the District of Columbia, Puerto
Rico, and 46 foreign countries.
Approximately 36 percent of the students
were minority students.
University A was extremely
fortunate in that it sustained only minor
building damage. It also lost some video
cameras, computers, and printers. More
serious was the threat of losing lab contents
and electronically-stored information due to
the combined loss of power and air
conditioning for a sustained period. In the
biology department, for example, the loss of
power and air conditioning was a
catastrophe for lab contents such as frozen
Journal of the Mississippi Academy of Sciences 249
tissues. However, even in the biology
department very little electronically stored
data were lost, which was the case for
University A as a whole. E-mail and
telephone service, and the student records
system, for example, were operational about
a week after the storm. More of a problem
was the lack of electricity, which meant that
electronically-stored data could not be
accessed.
Although it sustained only minor
damage to its buildings, the loss of power,
other utilities, and the workforce needed to
operate the university forced University A to
close for the 2005 fall term. Many students
did not attend school during the fall. The
ones who did were spread over 500 different
universities. The institution was re-opened
in January and out of 5,644 students who
had registered for the fall semester only
4,993 returned. The IT and registrar’s
offices did not lose many staff, but their
workloads increased. This was particularly
true for the Registrar’s Office as it tried to
track its students scattered across many
universities. However, many departments
lost faculty and like the students, not all
returned for the spring term.
University B was established in 1956
as a metropolitan campus of the Louisiana
State University System. Since the early
1970s it has also administered a large
summer program in Europe. University B
positions itself as “the urban research
university of the state of Louisiana and
provides essential support for the
educational, economic, cultural, and social
well being of the culturally rich and diverse
New Orleans metropolitan area.” University
B had a total enrollment of over 17,000
students, most of whom were from the
greater New Orleans area.
Unlike University A, University B
suffered major damage to its buildings. For
example, it still lacked functioning toilets in
some buildings nearly a year after Katrina
struck. In addition, some buildings received
interior water damage due to roof leaks,
which ruined carpets and furniture,
computers, video cameras, and printers.
Many faculty members set up temporary
offices on campus while the remainder
worked largely from home. The following
quote from the Registrar’s Office describes
the general situation across offices and
departments even a year after Katrina:
we’re still living out of boxes because you
can’t just immediately unpack and put
everything back together when you don’t
have as many employees as you used to
have.”
University B lost some
electronically-stored data. Notably, e-mails
and data stored in personal computers at
home. However, as was the case for
University A, University B did not suffer
major losses in its electronically-stored data.
The Registrar’s Office was working within
four days of the storm, (although from a
satellite office in Baton Rouge). The
primary problem was the lack of electricity
needed to access electronically-stored data.
The power was brought back building by
building and it took until December of 2005
before campus-wide power was restored. It
is worth noting in this regard, that
University B’s IT Office did not lose any
employees, but the loss of personnel in
registrar’s office and some academic
departments was substantial. Re-designing
activities and arranging online-courses
resulted in a significant increase of
workload.
Like University A, University B also
had to close its on-campus operations for
fall term 2005, but the teaching activities did
not stop: 700 online courses and about 300
courses on satellite campuses in the area
were offered. The campus re-opened in
January. University B lost a significant
number of students because of Katrina, but
immediately began to recover (see Fig. 3)
250 October 2007, Vol. 52 No 4
Most of the general budget for the
institutions comes from student tuition and
state funding driven by enrollment. Due to
the major decrease in enrollment, both
University A and B were thrown into
financial crisis, and many faculty and staff
were terminated. As of the time of the
writing of this report, some programs have
been combined; while other departments and
even some colleges have been completely
eliminated.
RESULTS
In this section we report what kind of
roles IT departments had and what kind of
roles ICT played in registration and
instruction. We classified the roles
according to Hannus (1994); facilitating,
enabling, and IT-driven. We also tried to
assess the IT capabilities and the changes or
shifts in the intellectual capital of the
organizations.
FROM THE PERSPECTIVE OF
INFORMATION TECHNOLOGY
Both universities already had
information and information technology
strategies in place that included disaster
recovery (DR) plans and “business
continuation” (BC) plans. This phenomenon
tends to be typical nowadays, but even for
organizations with such plans an actual
disaster results in plan revisions (Carpenter,
1993; Burke, 2005). For those organization
lacking such a plan (e.g. Vijayaraman and
Ramakrishna, 1993), recovery plans are
often developed after their first experience
with a disaster. For example, after
9/11/2001, financial firms were required by
law to develop and implement a business
continuation plan in the event of business
disruptions (Burke, 2005). According to
Carpenter (1993) the IT recovery plan
encourages management to develop a total
disaster recovery solution for the whole
organization. The ICT acted hence in the
role of a forerunner. The development of
ICT recovery and continuation plans might
serve as a model for these institutions to
create a more strategic and comprehensive
plan for the entire university.
University B’s representative
describes the change as follows: “In reality,
I don’t think it’s changed that much, but
they say that we will be up soon after a big
storm.” This sounds like the typical
statement about recovery time in DR plans.
17 142
6 684
11 446
0
2 000
4 000
6 000
8 000
10 000
12 000
14 000
16 000
18 000
Enrollment fall 2005 pre Enrollment fall 2005 post Enrollment spring 2006
Total Hdct
Full-Time Hdct
Part-Time Hdct
Journal of the Mississippi Academy of Sciences 251
The recovery time in the business sector is
in general 48 hours (Carpenter, 1993) or two
business days. In educational settings this
time may be longer. In our case study, the
recovery time varied from 24 hours to some
weeks depending on the department. The IT-
person representing University A described
what happened in more detail: “we already
had a disaster recovery plan before. And we
tested and we used it during Katrina. But
we’ve basically updated it and we found out
some things that worked out and things that
didn’t work well. So we’ve made
modifications to the plan to hopefully, if we
ever have to use it again, it’ll go a little bit
smoother. The second thing that we’re
trying to do is we’re trying to get this
building to work by a generator.” This
statement implicates that they tested, revised
and upgraded their disaster recovery plan
and did some physical investments such as
generators in the infrastructure. Hence, IT
acted mainly in the facilitating role. No
radical changes have been made from the
perspective of IT offices. They were
primarily engaged in continuous
improvement.
Both University A and B have
information technology strategies in place.
They already had backup-systems,
secondary computer rooms, and disaster
recovery plans, so essentially they were
making minor changes in their existing
plans. University B has taken extra steps to
prepare for another disaster: “but we also
now have more than one way to get to the
outside world, so to speak, on the network.
So we always stay up, at least our major
system stays up.” And “If everything were
washed away, we would still have a
presence outside of New Orleans”. We’ve
actually changed the way we ship tapes now
out of town, I mean out of the office, once a
week and actually to another building every
day, I’m not that concerned. Even if, like,
we have a chemical spill or fire or
something, I think we’re pretty safe it still
takes a little bit of luck involved, I guess, but
we’re pretty safe. I think as safe as we can
be.” Both organizations made incremental
changes to their existing plans in such a
manner that they enabled (or enhanced) their
core processes rather than making radical
changes.
As Katrina approached, neither
institution implemented a plan for
contacting students. Please recall that
Katrina was not originally forecasted to
strike New Orleans. However, the
registration office in University B had
created an emergency list of employees.
This was originally intended for more
typical and often individual emergencies,
such as illness. She described the limitation
of her emergency list: “And it took me about
ten days to track everyone down and part of
that was because the area code 504, in
terms of cell phones, was not operational.”
But in the aftermath of the hurricane
many kinds of supporting roles became
evident and the roles also changed and
expanded The nature of supporting depends
on the department, as well on the
characteristics of the informants. The IT
offices were typically mediators; they were
involved in communications with the
various departments. A representative of IT
in University A described their role “And
once we got our web server -- a secondary
web server up our staff was updating the
web server via what the president told us.
So we were just trying to disseminate the
information that somebody was giving us.”
Traditionally, IT has a range of roles in
redesign, as well as its usual service role. It
provides the platform for information
systems and is responsible for the continuity
of the information system. But after Katrina
there was a shift from the traditional role
towards the role of end-user by
disseminating the information that others
gave them. In terms of operational
252 October 2007, Vol. 52 No 4
procedures, IT offices seemed to function
very similar to pre-storm patterns, i.e., “My
particular office I don’t think is functioning
much differently at all except for the fact
that boxes are sitting in there that nobody’s
undone. They don’t belong to us; they
belong to somebody else in the building.”
Summa samarium, both
organizations already had the information
and information technology strategies with
their existing disaster recovery plans. They
only tested, revised, and upgraded them. The
biggest changes are that University A
created a secondary email system (beyond
the office) for faculty, staff and
administrators - but not for students. The
other typical change was the timeline and
location of backup-systems. Third, they
have prepared for IT resilience by providing
generators for back-up power. In contrast,
there was little or no alteration in the DR
plan for computers or information systems.
The role of IT department can be
categorized as being a service provider,
mediator, facilitator, and end-user. In
addition it acted as a forerunner in a more
abstract way for other disaster plans.
FROM THE PERSPECTIVE OF THE
REGISTRAR
The registrar’s offices offer an
interesting perspective to this study. The
policies and procedures at registration
offices seem to have experienced major
changes. Very soon after the storm they
turned into virtual offices existing in
cyberspace and using web-based Internet
systems from different locations far away
from their own campus. Thus the role of
ICT was vital as it provided the means/tools
to establish the virtual offices and create a
presence outside the disaster area. ICT
worked on the other hand in the role of a
facilitator but it might be that it was also in
the role of an enabler. ICT became a
powerful and innovative source for the
registrar’s functions of the universities.
One phenomenon was the increased
workload; University B lost numerous
employees from its registrar’s office, while
University A did not lose as many
employees. Due to the lack of employees
and the demands for new services, the
registrars had to re-design their processes.
Both institutions rapidly innovated:
University B developed a new online system
for registration. The students can now
register themselves online and the program
appears to be functioning better than their
original system. They also rapidly
transitioned from printed materials to online
documents. A representative of registrar’s
office in University B describes the
improvement: “In terms of things like print
publications that related to registration like
the class bulletins, that is all online now and
there is no intention of going back to a
hardcopy printed bulletin of any kind.”
The changes at the registrar’s office
A are also innovative and related to their
information systems. They now do things
that they had never thought of doing before.
They have developed their policies and
procedures in such a manner that they can
either take or access their procedure manuals
in case of evacuations. A variety of
different types of reports are run and put on
a drive and are then always ready to be
relocated. They will have remote access to
information for their currently enrolled
students, their contact information, degrees
and other major information. The level of
ambition is judged to be high. For the
registrar function, ICT is playing an
important enabling role.
In addition to the improved
resilience of student information, the
universities have also improved access to
other organizational data. Both have created
emergency lists that include cell phone
numbers and spouse’s cell phone numbers.
Journal of the Mississippi Academy of Sciences 253
University A also requires everybody in its
office to have a secondary e-mail address.
In creating another e-mail address ICT has
again provided a facilitating role and by
improving communications in emergency
situations, it works in an enabling role.
The registration offices also acted as
a mediator. A representative of University
B describes “But that I did and, I mean, I
had staff (in) Houston and Dallas and
Oklahoma, Georgia, Tennessee. I mean,
they went everywhere. So that was probably
the biggest push.” IT played a major role in
providing the communication linkage and
facilitation of information exchange for a
university that, in many respects, was
operating in a virtual world.
The faculty, staff and students from
University A were also widely dispersed
after Katrina. However, there was
substantial and parallel effort to establish
face-to-face contact in addition to ICT
communication. The representative of
registrar’s office A tells “I think we handled
it as best we could under the circumstances.
We all had our phone numbers out there. I
was receiving phone calls in South Carolina
from students who were concerned or had
special issues. We had phones set up in
Houston where they could call in and speak
with our vice president or a provost or even
with some of their deans were there. So,
maybe it wasn’t face-to-face contact, we
were easily accessible after about a week.
As a matter of fact, our president, once
things were up and running fairly smoothly,
he started making visits to many of the cities
that were housing so many of our students.
He would do alumni visits and also those
alumni organizations would bring in our
students from those schools.” It looks like
they really wanted to support and inform
students very closely by hearing and
visiting. Here ICT has no direct role yet in
face-to-face contacts.
Both registrar offices made
substantial changes in response to Katrina.
They re-designed their service processes and
in some cases created “mobile offices.”
There were divergence views on the
persistence of innovation and change. When
asked, “What is your office doing different
now?”: The director of the registrar’s office
of University A believed there was little
change, “we are functioning as if this never
happened” while the representative of
University B maintained “our office will
newer be back to what it was because of
reductions in students, reductions in
faculty.”
From a theoretical perspective, the
roles played by ICT in registrar offices’
activities are primarily facilitating and
enabling. IT capabilities in both offices were
good because of the flexible and integrated
infrastructure so that both were able to add
new applications in response to changing
demands. In terms of intellectual capital, it
seems that there is a shift from human
capital to structural capital due to
developing new information systems.
FROM THE PERSPECTIVE OF
INSTRUCTION
We were interested in how Katrina
impacted teaching strategies, as well as
learning and teaching quality. But especially
we were interested what kind of role ICT
played in instruction. These two
organizations had markedly different
strategies for instruction after the storm.
University B turned very quickly to
cyberspace and by October 10th was
offering over 700 online courses. In
addition, they offered about 300 courses at
satellite campuses in the area. On the other
hand, University A, offered no online
courses, but instead was focusing on
relocating their students to other
universities. About 3,000 of their students
254 October 2007, Vol. 52 No 4
were attending about 500 different
universities.
So the teaching strategy changed
dramatically for University B. By spring
semester both organizations were mainly
back in classroom, but there was a dramatic
shift in online course offerings in University
B. Before Katrina the university had offered
about five percent online courses. In the
spring semester, after Katrina, online
courses soared to about 30-35% of course
offerings.
Many felt that there was diminished
teaching effectiveness and learning quality
because of the storm. The Chair of
Communications in University A stated: “I
probably cut students slack that I normally
wouldn’t have done.” A professor at the
same university noted: “There’s no doubt
that everybody was severely traumatized,
both the students as well as the faculty and
staff.” The third teaching representative of
University A emphasized psychological
issues: “Almost everybody found it to be a
stressful semester in terms of stress levels of
both the faculty, problems the universities
were having fiscally and otherwise, and of
course the problems with the students. So it
was an overload semester. I think there was
overall a lower level of performance. I
certainly saw some of that in my courses. I
heard my colleagues report some early in
theirs.” Our respondents felt that the
learning and teaching quality at University
A was lower in spring semester after Katrina
than that before.
According to a professor and chair at
University A, the quality from the students’
perspective is better due to smaller class
sizes. He continues: “This summer we’re
offering actually more courses than we
offered last summer but that’s just because I
was asked to offer some additional courses.
But I don’t think it’s significantly different.”
The other professor and department chair
reports: “We’re in a transition trying to
figure out exactly how large the school is
going to be and how many faculty we need
and stuff. But the quality of the teaching
and learning itself has remained consistent.”
A third respondent said: “Well, I mean if we
go to the spring semester, spring of ‘06, I
would say we were back to normal as far as
the types of courses we taught. We offered a
few more Internet classes than we had in the
past. But the majority of our classes were
still classroom lecture seminar types of
classes. Fall semester was an aberration. It
was about 90 percent Internet classes. So I
mean it was a big adjustment for everyone in
the fall. But by the spring we were just
about back to normal.” In interpreting these
quotes, it appears that the professors
interviewed in University A felt that the
quality of learning and teaching remained
constant or at least returned to the same
level shortly after the storm.
Both universities were required “to
do more with less.” Almost everyone felt
that their workload was increasing. The
increasing workload resulted from a
combination of many different factors.
Many professors were teaching extra
courses to “make up” for classes cancelled
during the fall semester. The number of
faculty and staff had decreased because of
displacement, cancellation of programs and
difficulty in retention of staff. Student
revenue was down because of reduced
enrollments, while course loads had
increased because of reductions in staff.
Pressures for downsizing were strong and
constituted a substantial cost in faculty
morale and cohesion.
The universities clearly relied on
their websites as a primary means of
conveying information to students, faculty,
staff and other interested publics. E-mail
and phone communications were also
widely used to communicate and coordinate
among the dispersed members of the
universities. ICT played a profoundly
Journal of the Mississippi Academy of Sciences 255
important facilitating role. It allowed
faculty members to locate and contact
student majors, minors and other advisees.
Professors made themselves available
through cell phones, e-mail and even home
visits. One professor said that through his
phone communication, he was able to reach
about 75% of his major and minor students.
The answers to the question “How is
your department functioning differently
now?” vary a lot. Some respondents
mention physical things such as the
department is smaller because there are not
as many faculty members and students.
Some point out the psychological aspects
like “I would say the level of cohesion has it
where it wasn’t that before. But I think
people often appreciate the contacts they
have among colleagues because they’re --
we were allowed for months so there were a
lot of joyful and tearful reunions and that
kind of stuff. It’s very emotional stuff. And I
think in a way that enhances cohesion
among people.” Some of them feel just
uncertainty because they don’t know how
many students are going to be back and how
many faculty members they are going to be
able to recruit.
The most important role that ICT had in
instruction was the IT-driven role evident in
the increased offering of online courses. IT
capacity was substantial for University B, in
that they were able to leverage IT resources
to provide a massive number of new online
courses. In the case of University A, IT
capacity was less of an issue because of a
much different strategy of resilience - the
placement of students in other universities
and colleges. Human capital decreased
dramatically in both universities, because
they lost many experienced faculty
members. Relational capital, in contrast,
increased due to faculty and staff contacts
with counterparts in other universities. But
because we examine the students more from
the perspective of relational capital, it can be
seen decreasing or staying equal.
DISCUSSION AND RECOMMENDATIONS
The aim of this study was to try to
investigate the roles played by ICT from the
perspectives of three important university
activities: (1) technical support; (2) the
registrar’s office; and (3) instruction. As a
secondary goal, we wanted to elaborate how
universities can use ICT in disaster
recovery. The empirical part of the study
comprised an analysis of data collected from
two universities by phone interviews. We
used the case study method and selected the
form of the “historical narration.” We
approached the roles by using business
process redesign, IT capabilities and
intellectual capital theories.
According to the staff and faculty
members interviewed, a major problem after
Katrina was communication between
students, faculty and staff. Both universities
have since devoted considerable effort
improving the resilience of their
communication infrastructure and processes.
Furthermore, the importance of functioning
infrastructure arose, especially the reliability
and availability of electrical power during
emergencies. A third IT-specific challenge
was to assure the continuation of IT
functions, especially the protection of
information through improved back-up
security. Both universities had DR plans for
their information and information
technology. Rather than radically revising
plans, both universities refined, modified
and improved their existing documents.
ICT was found to provide the three expected
roles of facilitating, enabling, and driving.
We also identified new or expanded roles
256 October 2007, Vol. 52 No 4
Table 3.
ROLE OF ICT Organization A Organization B
Department/Activity IT department Registrar's
office Instruction IT department Registrar's office Instruction
Facilitating + x x x x x x
Enabling+ x x x
Driving+ x
Forerunner* x x
Service provider* x x
End user* x x
Mediator* x x x x
IT departments expanded their roles
during Katrina to include more end-user and
mediator functions, e.g., putting content on
the web. Interestingly, some individuals
expanded their traditional ad