Natural Hazards

, Volume 66, Issue 2, pp 1029–1036

Hurricane preparedness and planning in coastal public school districts in the United States


  • Jessica Van Meter
    • Department of GeographyKent State University
    • Department of GeosciencesMississippi State University
    • Department of GeographyKent State University
Original Paper

DOI: 10.1007/s11069-012-0534-5

Cite this article as:
Van Meter, J. & Schmidlin, T.W. Nat Hazards (2013) 66: 1029. doi:10.1007/s11069-012-0534-5


One hundred school districts were surveyed along the Atlantic and Gulf of Mexico coasts from North Carolina to Texas. Nearly all had recent experience with a tropical storm or hurricane and had hurricane plans in place. About half teach hurricane preparedness to students and 85 % train staff members in hurricane preparedness. Sources of information about cyclone threats were the National Hurricane Center (91 %), local television news (74 %), The Weather Channel (67 %), and the internet (67 %). Only 36 % would cancel classes for a hurricane warning but 89 % would cancel classes for a mandatory evacuation. Most districts (75 %) would use schools as storm shelters, and 92 % would use school busses to assist in community evacuations. Districts with a higher percentage of Hispanic population provided hurricane information in Spanish. Larger school districts were less likely to cancel classes in the middle of the day for a storm threat. Districts with higher home values were less likely to use school busses for evacuations, and smaller school districts were less likely to provide schools as storm shelters. There were no other significant associations between hurricane preparedness of the districts and district demographic variables of poverty, percent black, percent Hispanic, population, district size, or median home values.


HurricaneTropical cycloneSchoolsPreparednessUnited States

1 Introduction

This research examines the status of hurricane preparedness and planning in coastal school districts along the Atlantic and Gulf of Mexico coasts of the United States in the eight states from North Carolina to Texas and the relationships between school preparedness and the demographics of the school districts. Hurricanes and tropical storms bring strong winds, coastal storm surge, heavy rain, and river flooding as they approach a coast or make landfall and move inland. Hurricanes are a common threat along these coasts. The return period for a category 1 or stronger hurricane (1-min sustained wind speed of 33 ms−1 (74 mph) or greater) passing within 93 km (58 mi) of any given location ranges from 5 years to 15 years along this portion of the coast (National Hurricane Center 2012). Tropical storms (>17.5 ms−1, 39 mph) have even shorter return periods.

Schools along these coasts are in session from August to November during the Atlantic hurricane season so it is reasonable to expect that school administrations have experience with hurricanes and plans in place to minimize threats to human life and school property during a hurricane. In addition, schools are integral parts of the social community. They may serve as focal points for information and preparedness before the storm, emergency shelter during a storm, and recovery after a storm (Kelman 2011). Following Hurricane Andrew in 1992, schools in the affected area of Florida provided continuity and stability in the lives of the students and a return to normalcy after the storm (Provenzo and Fradd 1995). In addition to the education of students, school serve the community with day care, meals, transportation, sporting and cultural events, and after-school programs that many families rely upon. Students learn in schools about safety procedures for fires and natural hazards, and this information is taken home for dissemination in their families (Kurita et al. 2006; Mitchell et al. 2008; Palm and Hodgson 1993); thus, hazards education in the schools can form the core of community knowledge.

Hazard preparedness, planning, and response at all levels in society may be affected by demographics of the families or the communities. (Dow and Cutter 2000; Pielke and Pielke 1997; Tobin and Montz 1997; Whitehead et al. 2000). Factors such as lower income, nonwhite race and ethnicity, and smaller community size, for example, may increase hazard vulnerability through decreased ability to absorb and recover from losses, language and cultural barriers that affect access to resources and access to funds to prepare for and recover from a disaster (Cutter et al. 2003). Associations between hurricane preparedness and demographic factors in school districts could provide guidance on where efforts should be focused to improve hurricane preparedness in coastal regions.

The goals of this research were to (a) assess the status of hurricane planning and preparedness of coastal school districts from North Carolina to Texas, (b) assess the districts’ actions during recent hurricane threats, and (c) determine whether associations exists between planning/preparedness and the demographics of the school district.

2 Methods

A 41-question survey (available from T. Schmidlin) was designed to determine the size of the district, administrators’ contacts with outside agencies regarding hurricanes, presence of a hurricane plan, experience with hurricane watches and warnings in the recent 5 years, impacts from hurricanes, frequency and audience for hurricane preparedness training, decisions about closing schools due to an approaching storm, use of schools as public shelters, and use of busses for evacuation. The survey was mailed on September 8, 2010, to every school district in every county along the U.S. Atlantic and Gulf of Mexico coasts in the eight states from North Carolina to Texas. In addition, surveys were mailed to every district in about one-half of counties about 80 km inland and about one-third of counties 160 km inland. A total of 272 surveys were mailed. One hundred surveys were returned by November 1, 2010, a 37 % return rate (Fig. 1). These 100 surveys were the study sample to be analyzed.
Fig. 1

Locations of counties with at least one school district responding to the survey

The status of hurricane preparedness and planning was assessed through several survey questions. These included questions about recent experiences with hurricanes, vulnerability to flooding, whether a written hurricane plan exists, the presence of hurricane safety programs for students, staff, and community, sources of forecast and warning information, school cancelation policies, use of school busses for community evacuations, use of school buildings as storm shelters, and policies for resuming normal activities after the storm.

Demographic data for the school districts were obtained from the school district websites and from the U.S. Census Bureau. These included the number of students in the district, total population of the district, median home value, land area of the district, percent population in poverty, percent population black, percent population Hispanic, and percent with difficulty in English. If the district included an entire county, then the county demographics were used. If the district covered a portion of a county, then the demographics for the city of the district’s mailing address were used. Data were analyzed with Statistical Package for the Social Sciences (SPSS 2009).

3 Results

3.1 Demographics

Responses were received from 100 school districts (Fig. 1) with a wide diversity of demographics (Table 1) of student population, home values, and percent black, Hispanic, in poverty, and with difficulty in English. Median home values in the sample are lower than the eight-state average, perhaps because much of the high home values are in suburbs of large cities (Atlanta, Charlotte, and Austin for example) and many of these are inland and outside of the sampled region. Sample values of percent population in poverty, percent with difficulty in English, and percent black are near the eight-state averages, but percent Hispanic is lower in the sample than the eight-state average (Table 1).
Table 1

Summary of demographics of the 100 school districts with completed surveys

Demographic variable


Range of values

Eight-state average

Students in the district




Total population




Median home value ($)




Land area (mi2)




% Poverty




% Difficulty in English




% Black




% Hispanic




Eight-state average values were taken from 2008 U.S. Census data (US Census Bureau 2012)

Table 2

Status of hurricane preparedness and planning among school districts


Yes (%)

No (%)

Does your district have a written hurricane plan?



Are policies influenced by government agencies?



Has your district had a tropical storm watch or warning in the last 5 years?



Has your district been impacted by a tropical storm in the last 5 years?



Do you have school buildings that could be affected by storm surge?



Have any schools been impacted by any flood?



Is hurricane preparedness taught to students?



Is hurricane preparedness taught to school staff?



Do student or teacher handbooks contain hurricane information?



Would hurricane information be provided in school in a language other than English?



Would classes be canceled before the end of the day if tropical storm conditions intensified?



If evacuations were ordered in your district, would school busses and drivers assist in evacuation?



Would any of your district buildings be used as hurricane shelters?



3.2 Status of preparedness and planning

Most (94 %) of districts had a written hurricane plan, and these were typically influenced by government agencies, such as FEMA and state and county emergency management agencies (Table 2). Nearly all (99 %) districts reported being under a tropical storm or hurricane watch or warning in the last 5 years. Nearly half (46 %) reported being under such a watch or warning five or more times in the last 5 years. This is not surprising given the number of hurricanes during the 2005–2010 period (Fig. 2). Most districts (85 %) reported at least one instance of being warned of a tropical storm or hurricane but receiving no impact, a “false alarm” in their view. Most districts (79 %) had been impacted by a tropical storm or hurricane in the last 5 years. Nearly all put their hurricane plan into place during that storm and 75 % of those rated the effectiveness of the plan as 8 or better on a scale from 1 to 10 (best). Of the districts that put their plan in place during a storm, 79 % reported making changes to the plan after the storm. This indicates that plans are flexible, and efforts are made to improve the plans.
Fig. 2

Hurricane and tropical storm paths in the study area 2005–2010. Track information was provided by the NOAA Coastal Services Center

About half of districts (52 %) teach hurricane preparedness to students. Most of these teach at all grade levels, once or twice annually, and 82 % of those use a class or student assembly to convey information to students. A greater percentage of districts (85 %) train staff members in hurricane preparedness. Most of those (64 %) train all staff, but in districts where only select staff receives training, administrators and principals are the most common recipients. Staff training is typically given once or twice annually at a seminar or staff meeting. About one-third (32 %) of districts provide some hurricane information or programs to their community. The most common types of information were shelter location and evacuation procedures.

School districts receive information about tropical cyclone threats from a variety of sources. The most frequently mentioned sources were the National Hurricane Center (91 %), local news stations (74 %), The Weather Channel (67 %), and the internet (67 %). Although they may not consult the National Hurricane Center (NHC) directly, district staff seem to know that the watches and warnings originate with the NHC.

Most districts would not cancel classes for a tropical storm watch (11 %) or hurricane watch (3 %). This is expected since the watch is issued 48 h before impacts are expected. More districts would cancel classes for a tropical storm warning (29 %) or hurricane warning (36 %). Local evacuation was a stronger trigger for closing schools with 41 % closing for a voluntary evacuation and 89 % closing with a mandatory evacuation. Districts would announce school closures by television (99 %), radio (91 %), on a website (85 %), and by phone calls to parents (78 %).

Most districts (97 %) reported that their buildings may be used as a community shelter and most of those (77 %) would be staffed with emergency workers other than school employees. Districts would reopen schools as soon as possible after a storm, with most (71 %) expecting to reopen within 1 day and 94 % within 3 days of the storm.

3.3 Relationships between district preparedness and district demographics

Demographic data were binned into 3–4 nearly equal groups for each variable. Analysis was completed for all combinations by cross-tabulations and chi-square tests for associations between hurricane preparedness and demographics. The hypothesis of no association between variables was rejected if p < 0.05. Not surprisingly, providing hurricane policy information in a language other than English was associated (p = 0.02) with percent Hispanic population in the district. All of the districts in the highest sixth of Hispanic population provide information in a language other than English, while 61 % of districts with the lowest half of Hispanic population provide that information. Whether a district would close and send students home in the middle of the day in response to a hurricane threat was associated (p = 0.01) with student population. Larger districts were less likely to close in the middle of the day (55 %) than smaller districts (79 %), perhaps due to the complexity of transportation. Using school busses and drivers to help with hurricane evacuations in a community was associated with median home value. In general, school busses would be used in evacuations (92 % of all districts), but districts with higher home values were less likely to use their busses in evacuations (80 %) than districts with lower home values (100 %). Perhaps private transportation was sufficiently available in affluent districts to reduce the need for school busses. Most (75 %) school districts made their school buildings available as hurricane storm shelters. Use of schools as shelters was associated (p = 0.01) with student population (p = 0.01) and median home value (p = 0.00). Smaller districts were less likely to provide schools as shelters (62 %) than larger districts (93 %). Districts with lower home values were also less likely to provide schools as shelters than districts with higher home values. The latter association is surprising as one may expect poorer districts to need more community shelters.

Most of the other possible associations between preparedness and demographic variables were not statistically significant. Hurricane preparedness in these coastal school districts has little or no association with characteristics of the districts, such as poverty, percent black or Hispanic, population, district size, median home values, or distance from the coast.

4 Conclusions

These coastal or near-coastal school districts reported substantial experience with tropical storm or hurricanes in their districts in the recent 5 years. Hurricane plans were nearly universal, the plans were influenced by FEMA or state emergency management agencies, those plans were put into place during a recent tropical cyclone, districts were pleased with the effectiveness of their plans, and most made adjustments to the plan after a recent storm. These results show a positive status of hurricane planning in coastal school districts in the United States. Less positive were the rates of teaching hurricane preparedness to students (52 %) and the community (32 %). Past research shows that schools play an important role in moving emergency knowledge and planning from the classroom into the homes and community so it seems more hazard education should be conveyed to students in coastal districts. Evacuation orders (typically issued by local authorities) were more important (89 % would cancel) than hurricane watches or warnings (3 % would cancel with watch, 35 % with warning) issued by the National Hurricane Center in the decision to cancel classes. Whether this is an enhanced response to local authorities or an indicator of the perceived severity of the situation is not known. Schools were commonly used as community hurricane shelters. Busses and school staff would be used to assist in evacuations, further embedding the school district into community disaster response.

The lack of a general association between school district demographics and school hurricane preparedness was unexpected because wealth, race, and other demographics have been shown to affect hazard vulnerability (Tobin and Montz 1997). Perhaps an association exists but the measures of preparedness used here were not efficient assessors of actual preparedness or perhaps the correct demographic variables were not assessed. On the other hand, there is precedent in the literature for lack of association between preparedness and demographics. Kano and Bourque (2008) found no relationship between school preparedness for disasters and school characteristics such as size, urbanicity, resource base, and prior experience among 157 California schools. Horney et al. (2008) found no significant relationships between demographic factors and hurricane preparedness at the household level in a coastal North Carolina county.

Based on these results, hurricane preparedness is nearly ubiquitous in these coastal school districts where the threat from tropical cyclones is so frequent and guidance from federal and state emergency management agencies is effective. Differences in preparedness based on income, wealth, race, or ethnicity, if they ever existed, have become minimal.

While television and radio will likely remain important sources of information flow from the hurricane forecasters through school districts and into the community, it will be important for administrators in school districts to maintain awareness of tropical cyclone threats and communicate these threats through the internet (Sherman-Morris et al. 2011, for example) and emerging social media.

Copyright information

© Springer Science+Business Media Dordrecht 2013