How College Students’ Home Region Influences Their Risk Perceptions and Behaviors During Tornados

Amy Hyman
Arkansas State University

Joseph Richmond
Arkansas State University

Publication Date: 2024

Abstract

College students can be vulnerable to disasters because many have recently settled in the region where they are attending college and, as a result, lack knowledge about local natural hazards and how to protect themselves in the event of an emergency. In addition, college students are an understudied population in disaster research. Where college students are from may influence their perception of tornado risk due to lack of experience and knowledge of tornados. Prior experience with and knowledge of hazards has been found to influence risk perception and protective action behavior. On March 31, 2023, two EF3 tornadoes occurred in Arkansas. For this quantitative study, we surveyed 498 college students at Arkansas State University and the University of Arkansas at Little Rock, which are located in the two regions affected by the tornados. The purpose of the survey was to examine how students’ place of origin shaped their risk perceptions, protective action behaviors, and previous tornado experience. Our results show that students’ home region shaped their tornado experience and knowledge. However, there were no significant differences in the protective actions that students took to stay safe during the emergencies. These findings have practical implications for campus emergency planners and risk communicators, which we discuss in the conclusion. We published our survey instrument and data on DesignSafe—as Project 4724— so that these materials can be accessed by other researchers interested in using our data or doing a similar study.

Introduction

The United States experiences more tornados than any other country in the world, with over 1,000 occurring each year on average (Edwards, 20231; Burgueño Salas, 20232). On March 31, 2023, two separate EF3 tornados touched down in Arkansas, causing devastating damage in the cities of Little Rock and Wynne. Little Rock is a metropolitan area and home to the University of Arkansas at Little Rock. Wynne is a small town in rural Northeast Arkansas. Many students from Wynne commute to Arkansas State University in the nearby city of Jonesboro). Jonesboro was under the same tornado watch and warning advisories as Wynne prior to the tornado touching down. While neither campus was physically impacted by these tornados, both closed early on March 31 due to the forecasted severe weather. Both universities used text and email notifications to alert students, faculty, and staff of the school closure and the potential for a tornado emergency. Arkansas State continued to post social media posts throughout the day including information on where shelters were available to students and the public on campus. Keeping students informed and up to date with emergency information can help them make informed decisions related to protective actions.

Literature Review

The states of Texas, Oklahoma, Kansas, Nebraska, South Dakota, Iowa, and Nebraska have traditionally been considered part of “Tornado Alley” because this region has recorded the most tornados and tornadic storms in the U.S. history. However, there is a recent effort to focus on tornados in the southeastern states of Arkansas, Louisiana, Mississippi, Alabama, Tennessee, Kentucky, and Illinois (Agee et al., 20163; Cao et al., 20214). Research has shown that the frequency and power of tornados in the southeastern United States has increased in recent years (Elsner et al., 20195; Elsner et al., 20156). Some argue that this shift can be explained by increased population density in specific areas and better detection systems (Agee & Taylor, 20197; Nouri et al., 20218) while others argue that these shifts are due to climate change (Agee et al., 2016). Regardless of the cause, populations in the area are increasingly exposed to tornados. This shift in tornadic activity creates challenges for the emergency managers and risk communicators responsible for disseminating alerts, warnings, and risk messages to the public.

College Student Vulnerability to Disasters

Research shows that college students are vulnerable to disasters for multiple reasons, including their tendency to underestimate risk, their residence in geographic areas that they may be unfamiliar with, their lack of knowledge about local hazards, and their lack of resources such as limited financial security, unreliable transportation, or subpar housing that more permanent groups possess (Jauernic & Van Den Broeke, 20179; Lovekamp & Tate, 200810; Rohli et al., 201811; Simms et al. 201312; & Wu et al., 201713). Jauernic and Van. Den Broeke (2017) found that students who originated from the Great Plains had a higher risk perception of tornados than international students. While this study is informative, it focused on students from the Great Plains and did not assess the risk perceptions and tornado experience of respondents from the Southeastern United States. The current study seeks to fill this gap. In addition, this study will also focus on how risk perception influences hazard adjustments and actions students take to protect themselves, they did not focus on if home region played a role (Greer et al., 201814; Wu et al., 2017)., topics that have yet to be explored sufficiently in the literature.

Risk Perception and Experiences

How individuals perceive risk influences the protective actions and other decisions that they make during disasters (Savoia et al., 201715). Slovic (1987: 280)16 defined risk perception as the “intuitive risk judgments” that individuals use to evaluate hazards. While intuition plays a role in the risk perception process, other factors also influence one’s risk perception; these factors include gender (Stokoe, 201617), age (Balog-Way et al., 202018), and past experiences (Demuth et al., 201619). Past experiences can be a significant and complex factor influencing risk perception (Demuth, 201820). Knuth et al. (2014)21 found that survivors of emergencies had a heightened risk perception after experiencing an event. Other studies have found that individuals who have experienced tornados were more likely to accurately estimate, or even overestimate, their tornado risk (Ellis et al., 201822; Senkbeil et al., 201923).

Protective Actions

During disasters, individuals take different actions to protect themselves based on various factors, including their risk perception and past experiences. Greer et al. (2018) found that college students who experienced an earthquake previously were more likely to adopt hazard adjustments afterward due to increased risk perception. Research shows that individuals who experience a disaster are more likely to take protective actions than those who only hear about a disaster (Stock et al., 202124). However, other research suggests there are inconsistencies between disaster experience and protective actions (Liddell et al., 202025; Lindell & Hwang, 200826).

Research Questions

While there are existing studies on college students’ behaviors and risk perceptions concerning tornados (Huntsman et al., 202127; Lovekamp & Tate, 2008) and other hazards (Greer et al., 2018; Rohli et al., 2018; Simms et al., 2013), there is little knowledge on the differences between students from the region where the campus is located and those who are from outside the campus region. Two previous studies by Jauernic and Van Den Broeke (2017) and Senkbeil et al. (2019) compared college students’ risk perceptions of tornados based on home region, but these are the only known studies that do so. This study seeks to fill that gap by examining the different behaviors and perceptions related to tornadic events between college students based on their home region in order to better inform campus emergency planners and risk communicators. This study asked two research questions:

  1. What are the differences in risk perception between students from different home regions?
  2. What are the differences in protective action behavior between students from different home regions?

Research Design

We used a survey to examine how tornado risk perceptions and behaviors of college students from different regions varied. We have published our survey instrument and data on DesignSafe so that these materials can be accessed by other researchers interested in using our data or doing a similar study (Hyman & Richmond, 202328).

Study Site and Access

The areas of interest were the Little Rock Metro Area and Northeast Arkansas. These two regions were selected due to the tornados that occurred on March 31, 2023. In addition, both regions are home to higher education institutions. The Little Rock Metro Area has a population of 757,945 (Census Reporter, n.d.29). This region in Arkansas is an urban area and is home to the University of Arkansas at Little Rock. The Northeast Arkansas area has a population of approximately 135, 512 (City Population, n.d.30). Jonesboro is the regional hub for Northeast Arkansas and houses Arkansas State University. The northeastern part of Arkansas is nestled in the Delta and is mostly rural, with agriculture as the primary industry.

Survey

An online questionnaire was created and disseminated via email at Arkansas State University and the University of Arkansas at Little Rock through student listservs and newsletters to reach all students. The survey link was also posted on institution-related social media accounts. The survey was comprised of 26 questions, and participants were provided with an informed consent statement on the questionnaire landing page that they had to agree to before proceeding with the survey. In order to provide an incentive for respondents to take the survey, participants had the option of entering their email address to win one of six $100 gift cards.

Sampling Strategy

A convenience sample was selected for this study in order to collect time-sensitive data, also known as perishable data. After a disaster occurs, it is vital to collect perishable data to gather people’s thoughts, behaviors, and attitudes during a disaster before they are forgotten. The collection of real-time data after a disaster occurs can provide more accurate data, which results in better recommendations for future practical applications and research (Institute of Medicine, 201531). The sampling strategy for this study is considered a convenience sample since the sample was not randomly selected. An online survey was utilized to collect responses. Anyone who met the respondent requirements could take the survey. Participants had to be 18 years of age or older and current students enrolled at a higher education institution in the regions of Northeast Arkansas or the Little Rock Metro area. Participants had to confirm their participation consent before proceeding to the survey.

Survey Measures

The measurement scales used in the questionnaire are described below. In addition to the measurement scales discussed below, basic demographic questions were included.

Home region. Participants were also asked to identify their home state or country in order to categorize them into in-region or out-region for tornado risk regions. Respondents were grouped into in-region (living in tornado prone states), out-region (living in states less prone to tornados and students originating from a country outside of the United States). In-region states include states traditionally belonging to Tornado Alley (Texas, Oklahoma, Kansas, Nebraska, South Dakota, Iowa, and Nebraska) and states in the Southeast region including Arkansas, Louisiana, Mississippi, Alabama, Tennessee, Kentucky, and Illinois.

Risk Assessment. We adapted the measure for participants’ risk assessment during the event from a survey by Huang et al. (2012)32 on evacuation decisions and risk perception in response to Hurricane Ike. Our measure was adapted to fit a tornado event rather than a hurricane event. Five-point Likert scales (1-5) were used to measure risk assessment. Scale items for risk assessment ranged from not at all likely (1) to almost certain (5). This was used to measure participants’ risk perception of tornadic activity during the event based on environmental cues and how respondents perceived risk related to those cues.

Risk Perception of Home Region. Participants were asked to rate the level of tornado risk in their home state or country, using a five-point Likert scale ranging from marginal risk (1) to high risk (5).

Tornado Hazard Knowledge. Participants were asked to rate their tornado knowledge prior to the March 31, 2023, events on a scale from 1-5, with 1 being not at all knowledgeable and 5 being very knowledgeable.

Tornado Experience. Participants were asked if they had ever experienced a tornado before the March 31, 2023, event. This was a binary response with yes or no for an answer. If participants answered yes, they were then asked how many tornados they had experienced before March 31, 2023, and how many of these events were strong or violent tornados (EF3/F3 and greater).

Warning Sources. This measurement is from the NOAA Tornado Post-Event Survey (Natural Hazards Center, n.d.33). Participants identified the types of warning sources that they utilized to learn information about the event. This was a multiple-choice question where respondents could select all the warning sources they utilized during the event.

Protective Actions. We used measures of protective actions from the NOAA Tornado Post-Event Survey (Natural Hazards Center, n.d.). This portion of the survey asked respondents what actions they took once they received the tornado warning. The measure was a multiple-choice question where respondents could select all options that apply.

Survey Distribution

The online questionnaire was administered through the survey platform Qualtrics. The anonymous questionnaire did not collect any identifying information from the participants. Participants were able to choose to submit their email to a drawing to win a gift card, but their email addresses were not linked to their survey responses. The survey was disseminated via email at two institutions in Arkansas: Arkansas State University and the University of Arkansas at Little Rock. The survey was also disseminated via social media on Facebook and Instagram. Strategic marketing was utilized on social media to target college students in the Little Rock Metro and Northeast Arkansas areas.

Demographic Characteristics of Respondents

A total of 689 respondents completed the online questionnaire. More females responded than males and there was a wider age range than anticipated among respondents. The breakdown of demographics can be found in Table 1.

Table 1. Respondent Demographic Characteristics

Variable Description n %
Gender Male 139 28.8
Female 321 66.6
Prefer not to answer 22 4.6
Total 482 100.0
Race Hispanic/Latino 19 3.9
Native American/Alaskan 3 0.6
Asian 21 4.3
Black/African American 60 12.4
White 337 69.6
Two or more races 30 6.2
Prefer not to answer 14 2.9
Total 484 100.0
Age 18-22 Years 173 37.1
23-27 Years 127 27.3
28-32 Years 53 11.4
33-37 Years 42 9.0
38 Years & Older 71 15.2
Total 466 100.0

Table 2 shows the grouping of respondents by home region. Some respondents did not share their home region which forced us to eliminate their responses during data analysis comparing home regions. A total of 470 responses were analyzed. A surprising number of respondents were over the age of 27. The Education Data Initiative recently reported that 66.6% of college students in the United States are 24 years of age or younger (Hanson, n.d.34). The age demographic could be considered in future studies examining college students and an older age group of students could be a limitation to this study which will be discussed later on in this report.

Table 2. Respondent Home Region

Variable Description n %
Region In Region 356 75.7
Out Region 114 24.3
Total 470 100.0
Note. In-region respondents originated from states that are tornado prone such as Tornado Alley and the Southeast region. Out-region students originated from states that are less tornado-prone and countries outside of the United States.

Ethical Considerations

IRB approval was obtained by two institutions in the state of Arkansas. Arkansas State University provided IRB approval on April 10, 2023, and the University of Arkansas Little Rock provided IRB approval on June 14, 2023. IRB approval provided access to primary higher education institutions in the Northeast Arkansas region and the Little Rock Metro Area. This allowed recruitment emails to be disseminated via student listservs and placements in institution newsletters along with social media postings from institution affiliated accounts.

There are unique challenges when conducting research after a disaster has occurred, including ethical considerations. Those who have experienced a disaster must be considered when conducting research after an event, as populations who experienced a disaster may be traumatized and vulnerable (Hunt et al., 201635). Researchers must be considerate and thoughtful of the trauma and challenges that populations experience when conducting disaster research. However, collecting data and gathering information after a disaster can produce knowledge and insight that lead to planning improvements and other practical applications. Conducting research quickly after a disaster can help capture the experiences and behaviors of those who experienced the disaster while memories are still fresh (Browne & Peek, 201436). This can help improve the understanding of hazard effects on individuals (Edwards, 202137). This study disseminated an online questionnaire that participants could voluntarily take without feeling pressured. Participants were informed that their participation was voluntary and that they could withdraw from the questionnaire at any time.

Data Analysis Procedures

Respondents were grouped by their home region. This included grouping respondents into in-region students and out-region students. In-region students were identified as respondents living in the Southeast region or Tornado Alley, while out-region students were identified as living in any region outside of the Southeast region or Tornado Alley, including other countries. The original intent of this study was to compare respondents living in Tornado Alley versus the Southeastern region, and to compare international students to domestic students. However, many respondents’ home region was in Southeast region, so respondents from the Southeast region and Tornado Alley were grouped together in the in-region group. International students and those living outside of tornado-prone states were grouped into the out-region group. Descriptive statistics were used to examine demographics and other questions asked in the survey. A comparison of means was used to examine the differences between college groups based on their home region.

Results

Risk Assessment

We examined students’ assessment of risk for the March 31, 2023, events in order to gauge if environmental cues in the moments leading up to and during the event influenced risk assessment. Respondents were asked to assess their feelings about the likelihood of being affected by the event based on environmental cues. Out-region respondents had very similar risk assessment index scores compared to in-region respondents. There were no statistically significant differences between the groups (U=19,819.0, p=0.920).

Risk Perception of Home Region

We used mean comparisons to compare how students perceived tornado risk in their home region or state. Respondents were asked to rate what level of tornado risk that they perceived existed in their home region. Out-region students perceived tornado risk in their home states or country as “moderate,” on average. In-region students perceived higher tornado risk, rating their states as at “enhanced risk” for tornados. The mean ranked difference between these groups was statistically significant (U=12,831.5, p<0.001) with a moderately weak effect size (r=0.278).

Tornado Experience and Knowledge

Respondents’ past experience with tornadic events and their knowledge regarding tornado hazards were examined using mean comparisons. Out-region respondents tended to have less experience with tornadic events (N=57 or 18.5%) when compared to in-region respondents (N=251 or 81.5%). The difference between these groups on the hazard experience variable was statistically significant (X2(1,469)=16.408, p<0.001), but the effect size was small (ϕ=0.187).

Out-region respondents had slightly less prior knowledge about tornadic events (N=114, Some knowledge) when compared to in-region respondents (N=305, Knowledgeable). The mean rank difference between these groups was statistically significant (U=16,126.5, p<0.001), but the effect size was small (r=0.157).

Warning Sources and Protective Actions

During the event, out-region respondents used roughly the same number of warning sources as in-region respondents, and out-region respondents took roughly the same number of protective actions as in-region respondents. No statistically significant differences were found between these groups on the warning sources variable (U=17,950.5, p=0.101) or the protective actions variable (U=19,526.5, p=0.465).

Discussion

Tornado Risk Perceptions Among Students From Different Regions

As outlined above, the first research question in this study aimed to examine how risk perceptions varied among students from different home regions. This section interprets results from survey questions measuring students’ risk assessment, risk perceptions for home region, and prior tornado experience and knowledge.

Risk Assessment

The risk assessment portion of the questionnaire asked participants to examine their perception of risk in the moments leading up to and during the March 31, 2023, events. Out-region respondents had similar index scores as in-region students which could be due to several factors. First, while there are high-risk areas in the United States, a large portion of the country does experience tornadic activity from time to time, so while out-region students may live in a low-risk tornado area, they might have experienced tornados prior to the March 31, 2023, events. Second, out-region students might have an increased self-efficacy due to knowledge and familiarity with tornado risks. Studies have found that knowledge can increase one’s self-efficacy (Bandura, 200138; Kilic & Simsek, 201939). While out-region students were found to have less knowledge compared to in-region students in this study, it does not mean they had no knowledge about tornados at all. Another explanation for these results is that students had just experienced a tornado and were now familiar with the environmental cues that occur during tornadic storms. Regardless of home region, their risk perceptions during the event were heightened. Future studies could examine college students’ risk perceptions before an event occurs and then re-survey after the event in order to better gauge how experiencing an event can affect one’s risk perception.

Risk Perception for Home Region

Out-region students were found to have a lower risk perception for their home region than in-region students. While this is not surprising, this suggests that students who move from a lower-risk tornado regions might have lower overall tornado risk perceptions even after moving to a high-tornado-risk region.

Tornado Experience and Knowledge

Out-region students were found to have less experience with tornados compared to in-region students. This is unsurprising as out-region students originate from areas that are less tornado-prone.. As discussed above, research shows that experience is often an antecedent to risk perception (Demuth, 2018). Since our findings show that out-region students had lower tornado experience with tornados, it could be argued that they also had lower tornado risk perceptions prior to the March 31 event. As discussed above, while risk perception may be lower before an event for those less familiar with tornados, experiencing an actual tornado may increase their risk perception afterwards. This interpretation supports past findings that experience with hazards can increase one’s risk perception (Knuth et al., 2014; Wallace et al., 201540).

Out-region respondents also had less knowledge about tornadic events compared to in-region respondents. Again, this is unsurprising but is useful as knowledge about tornados has been found to influence protective actions in past studies. Drost (2013)41 found that college students with higher levels of knowledge about tornados were more likely to take protective actions compared to those that had less tornado knowledge. Drost (2013) further examined how experience and knowledge played a role in college students’ behaviors towards tornados and found that those who have experienced a tornadic event but had less knowledge were the least likely to take a tornado warning seriously. These findings are supported with the findings of other studies indicate the complexity of these factors and the importance of examining them.

Warning Sources and Protective Actions Among Students From Different Home Regions

The second research question examined how student behavior during the tornado event varied among students from different home regions. Below we discuss whether home region affected the warning sources they used and the action they took to protect themselves.

The findings show that during the March 31, 2023, event, out-region respondents used a similar number of warning sources and also took the same number of protective actions as in-region respondents. The lack of significant differences between the groups is significant in itself as out-region students took the same number of protective actions despite having less experience with tornados in the past. The findings from this study regarding protective actions do not align with other studies (Drost, 2013; Lindell & Perry).

One explanation for the similar behavior of in- and out-region could be that they both were paying attention to the warning sources and therefore were informed by media or other information sources on what actions to take. Several warning messages and social media posts put out during the event provided students with information on students what actions to take. Most college campuses have an emergency alert system in place to notify students during disasters in order to comply with the Clery Act requirements (Han et al., 201542). In addition, FEMA conducted a national test in 2023 using IPAWS to reach millions who use mobile phones (FEMA, n.d.). These warning systems can play a very important role in educating individuals on what protective actions to take even if they have no prior knowledge. Future studies could examine this topic further to explore why protective actions did not differ between these groups and the relationship between warning sources and protective actions.

Conclusions

This study examined if college students’ home region influenced their risk perceptions and protective action behaviors during the March 31, 2023, tornados in Arkansas. With tornadic activity increasing in the southeastern United States, it is vital that campus planners in the region adequately prepare for their population and identify various risk perceptions and behaviors between group populations. In-region students were found to have more experience and knowledge regarding tornados compared to out-region students. The results also found that there are differences in home region risk perception prior to events which could influence one’s perception of risk. The relationship between risk perception, experiences, and home region should be further examined in order to better understand how college students’ backgrounds affect their time on campus. However, there were little differences when it came to college groups taking protective actions during the events which could be explained by the emergency alert systems informing students of what protective actions to take. Future studies could study the connection between emergency alert systems and protective actions among college students to see if prior experience and knowledge play a role in the decision-making process.

Implications for Practice or Policy

This study has several practical implications for college campus emergency managers, risk communicators, and any emergency managers that have higher education institutions within their jurisdiction. Students on college campuses come from a wide array of geographical areas, backgrounds, cultures, and life experiences which can create unique challenges for emergency planners and communicators regarding hazards and their risks. In addition to the diverse nature of college campuses, college students are considered a vulnerable population due to their limited financial resources and unfamiliarity with local hazards (Jauernic and Van Den Broeke, 2017; Lovekamp & Tate, 2008; Rohli et al., 2018; Simms et al. 2013; & Wu et al., 2017). College student groups are understudied in relation to risk perception and protective action behaviors in regard to natural hazards, particularly tornados. Understanding the different groups across campuses and how they interpret risk can help campus planners tailor plans and engage in educational campaigns to manage perceptions of risk and behaviors during disasters and help ensure that all students take actions to protect themselves. This study can further influence higher education and emergency management legislation at the local and state levels and help ensure that policy effectively protects vulnerable college populations from natural hazard events. Higher education institutions could consider implementing a section about local hazards into First Year Experience (FYE) courses or orientation sessions to educate incoming students unfamiliar with their new residence. Promoting local warning system enrollment is another practice higher education institutions could consider if enrollment is not already automatic or mandated. Universities and colleges could also gather more information on their student demographics related to hazard knowledge and experience to tailor messages and plans for their campus. Understanding students and their backgrounds can help campus planners update plans that are inclusive of the entire student population. This can help ensure that limited financial resources, including those for structural mitigation projects, are directed to areas with the greatest return on investment and with the most positive impact on campus life safety.

Limitations

The geographical area limits this study as the results reported may not be generalizable to other populations who experience tornado hazards differently. Future studies should focus on higher education institutions outside of the Southeast region of the United States as well as international student populations. This will help better gauge how home region plays a role in college groups’ risk perception and protective action decisions during tornadic events. Another limitation identified is the sampling method itself as a convenience sample can also create challenges with generalizability. The higher age range of students could also be a limitation as out-region students who are older may have lived in a tornado-prone area longer and be more knowledgeable about the risks of tornados. Future studies could set an age limit on respondents, but with the evolving college student demographic, response rates may decrease with the age limit. The low international student respondent rate is also a limitation. Future studies could make an effort to reach more international respondents to get a better representation of the population.

Lastly, to measure respondents’ tornado hazard knowledge we asked respondents were asked to rate their knowledge on a scale from 1-5. Future studies could ask respondents specific tornado hazard facts to truly gauge respondent knowledge.

Future Research Directions

Future research could examine higher institution campuses from multiple regions to gain a deeper understanding of how hazard experience plays a role in college student groups’ risk perception and protective action behavior. This would allow for groups with a wider range of hazard experiences to be examined and potentially produce more significant results in college group comparisons. Gathering a larger number of international respondents could also provide more insight into how international students view tornadic risks and their knowledge of protective action behaviors.

References

  1. Edwards, Roger. (2023, February 19). The Online Tornado FAQ. National Oceanic and Atmospheric Administration, Storm Prediction Center. https://www.spc.noaa.gov/faq/tornado/ ↩

  2. Burgueño Salas, Erick. (November 9, 2023). Number of tornadoes in the United States from 1995 to 2021 [Graph]. In Statista. https://www.statista.com/statistics/203682/number-of-tornadoes-in-the-us-since-1995/ ↩

  3. Agee, E., Larson, J., Childs, S., & Marmo, A. (2016). Spatial redistribution of U.S. tornado activity between 1954 and 2013. Journal of Applied Meteorology and Climatology, 55(8), 1681-1697. https://doi.org/10.1175/JAMC-D-15-0342.1 ↩

  4. Cao, Z., Cai, H., & Zhang, G. J. (2021). Geographic shift and environment change of US tornado activities in a warming climate. Atmosphere, 12(5), 567. https://doi.org/10.3390/atmos12050567 ↩

  5. Elsner, J. B., Fricker, T., & Schroder, Z. (2019). Increasingly powerful tornadoes in the United States. Geophysical Research Letters, 46(1), 392-398. https://doi.org/10.1029/2018GL080819 ↩

  6. Elsner, J. B., Elsner, S. C., & Jagger, T. H. (2015). The increasing efficiency of tornado days in the united states. Climate Dynamics, 45(3-4), 651-659. https://doi.org/10.1007/s00382-014-2277-3 ↩

  7. Agee, E., & Taylor, L. (2019). Historical analysis of U.S. tornado fatalities (1808–2017): Population, science, and technology. Weather, Climate, and Society, 11(2), 355-368. https://doi.org/10.1175/WCAS-D-18-0078.1 ↩

  8. Nouri, N., Devineni, N., Were, V., & Khanbilvardi, R. (2021). Explaining the trends and variability in the united states tornado records using climate teleconnections and shifts in observational practices. Scientific Reports, 11(1), 1741-1741. https://doi.org/10.1038/s41598-021-81143-5 ↩

  9. Jauernic, S. T., & Van Den Broeke., M. S. (2017). Tornado Warning Response and Perceptions among Undergraduates in Nebraska. Weather, Climate, and Society, 9(2), 125-139. https://doi.org/10.1175/WCAS-D-16-0031.1 ↩

  10. Lovekamp, W. E., & Tate, M. L. (2008). College Student Disaster Risk, Fear and Preparedness. International Journal of Mass Emergencies & Disasters, 26(2), 70–90. https://doi.org/10.1177/028072700802600201 ↩

  11. Rohli, R. V., Collins, J. M., Ersing, R. L., Lunsford, G. D., & Ludwig, A. M. (2018). Hurricane preparedness among university residential housing assistants and staff. Weather, Climate, and Society, 10(2), 341-359. https://doi.org/10.1175/WCAS-D-17-0012.1 ↩

  12. Simms, J. L., Kusenbach, M., & Tobin, G. A. (2013). Equally unprepared: Assessing the hurricane vulnerability of undergraduate students. Weather, Climate, and Society, 5(3), 233-243. https://doi.org/10.1175/WCAS-D-12-00056.1 ↩

  13. Wu, H-C., Greer, A., Murphy, H. C., & Chang, R. (2017). Preparing for the new normal: Students and earthquake hazard adjustments in Oklahoma. International Journal of Disaster Risk Reduction, 25, 312-323. https://doi.org/10.1016/j.ijdrr.2017.09.033 ↩

  14. Greer, A., Wu, H., & Murphy, H. (2018). A serendipitous, quasi-natural experiment: Earthquake risk perceptions and hazard adjustments among college students. Natural Hazards (Dordrecht), 93(2), 987-1011. https://doi.org/10.1007/s11069-018-3337-5 ↩

  15. Savoia, E., Lin, L., & Gamhewage, G. M. (2017). A conceptual framework for the evaluation of emergency risk communications. American Journal of Public Health, 107(S2), S208-S214. https://doi.org/10.2105/AJPH.2017.304040 ↩

  16. Slovic, P. (1987). Perception of risk. Science, 236(4799), 280-285. https://doi.org/10.1126/science.3563507 ↩

  17. Stokoe, R. M. (2016). Putting people at the centre of tornado warnings: How perception analysis can cut fatalities. International Journal of Disaster Risk Reduction, 17, 137-153. https://doi.org/10.1016/j.ijdrr.2016.04.004 ↩

  18. Balog-Way, D. H. P., Evensen, D., & Löfstedt, R. E. (2020). Pharmaceutical Benefit–Risk perception and age differences in the USA and Germany. Drug Safety, 43(11), 1141-1156. https://doi.org/10.1007/s40264-020-00977-6 ↩

  19. Demuth, J. L., Morss, R. E., Lazo, J. K., & Trumbo, C. (2016). The effects of past hurricane experiences on evacuation intentions through risk perception and efficacy beliefs: A mediation analysis. Weather, Climate, and Society, 8(4), 327-344. https://doi.org/10.1175/WCAS-D-15-0074.1 ↩

  20. Demuth, J. L. (2018). Explicating experience: Development of a valid scale of past hazard experience for tornadoes. Risk Analysis, 38(9), 1921-1943. https://doi.org/10.1111/risa.12983 ↩

  21. Knuth, D., Kehl, D., Hulse, L., & Schmidt, S. (2014). Risk perception, experience, and objective risk: A cross-national study with European emergency survivors. Risk Analysis, 34(7), 1286-1298. https://doi.org/10.1111/risa.12157 ↩

  22. Ellis, K. N., Lisa, R. M., Gassert, K. N., Elsner, J. B., & Fricker, T. (2018). Public perception of climatological tornado risk in Tennessee, USA. International Journal of Biometeorology, 62(9), 1557-1566. https://doi.org/10.1007/s00484-018-1547-x ↩

  23. Senkbeil, J. C., Ellis, K. N., & Reed, J. R. (2019). The influence of tornado activity, impact, memory, and sentiment on tornado perception accuracy among college students. Atmosphere, 10(12), 732. https://doi.org/10.3390/atmos10120732 ↩

  24. Stock, A., Davidson, R. A., Trainor, J. E., Slotter, R., Nozick, L. K., & Kruse, J. B. (2021). What makes homeowners consider protective actions to reduce disaster risk? an application of the precaution adoption process model and life course theory. International Journal of Disaster Risk Science, 12(3), 312-325. https://doi.org/10.1007/s13753-021-00350-w ↩

  25. Liddell, J. L., Saltzman, L. Y., Ferreira, R. J., & Lesen, A. E. (2020). Cumulative disaster exposure, gender and the protective action decision model. Progress in Disaster Science, 5, 100042. https://doi.org/10.1016/j.pdisas.2019.100042 ↩

  26. Lindell, M. K., & Hwang, S. N. (2008). Households' perceived personal risk and responses in a multihazard environment. Risk Analysis, 28(2), 539-556. https://doi.org/10.1111/j.1539-6924.2008.01032.x ↩

  27. Huntsman, D., Wu, H., & Greer, A. (2021). What matters? Exploring drivers of basic and complex adjustments to tornadoes among college students. Weather, Climate, and Society, 13(3), 665-676. https://doi.org/10.1175/WCAS-D-21-0008.1 ↩

  28. Hyman, A., J. Richmond. (2023) "College Students’ Risk Perceptions and Behaviors During Tornadic and Severe Weather Events", in College Students’ Risk Perceptions and Behaviors During Tornadic and Severe Weather Events. DesignSafe-CI. https://doi.org/10.17603/ds2-y3za-8092 ↩

  29. Census Reporter. (n.d.). Little Rock-North Little Rock-Conway, AR Metro Area. Retrieved January 3, 2024, from http://censusreporter.org/profiles/31000US30780-little-rock-north-little-rock-conway-ar-metro-area/ ↩

  30. City Population (n.d.). Jonesboro Metropolitan Statistical Area in USA. Retrieved January 3, 2024, from: https://www.citypopulation.de/en/usa/metro/27860__jonesboro/ ↩

  31. Institute of Medicine. (2015). Enabling rapid and sustainable public health research during disasters: Summary of a joint workshop by the institute of medicine and the U.S. Department of Health and Human Services. National Academies Press. https://doi.org/10.17226/18967 ↩

  32. Huang, S., Lindell, M. K., Prater, C. S., Wu, H., & Siebeneck, L. K. (2012). Household evacuation decision making in response to Hurricane Ike. Natural Hazards Review, 13(4), 283-296. https://doi.org/10.1061/(ASCE)NH.1527-6996.0000074 ↩

  33. Natural Hazards Center. (n.d.). NOAA Tornado Post-Event Survey. Special Call: Tornado Ready Quick Response Research. Retrieved January 4, 2024, from https://hazards.colorado.edu/uploads/basicpage/NOAA%20Tornado%20Post-Event%20Survey_NHC.pdf ↩

  34. Hanson, M. (n.d.). College enrollment & student demographic statistics. Education Data Initiative. Retrieved January 22, 2024, from https://educationdata.org/college-enrollment-statistics ↩

  35. Hunt, M., Tansey, C. M., Anderson, J., Boulanger, R. F., Eckenwiler, L., Pringle, J., & Schwartz, L. (2016). The challenge of timely, responsive and rigorous ethics review of disaster research: Views of research ethics committee members. PloS One, 11(6), e0157142-e0157142. https://doi.org/10.1371/journal.pone.0157142 ↩

  36. Brown, K. E., & Peek, L. (2014). Beyond the IRB: An ethical toolkit for long-term disaster research. International Journal of Mass Emergencies and Disasters, 32(1), 82–120. ↩

  37. Edwards, F. L. (2021). Practical Considerations for Ethical Research in Post-disaster Communities. In J. D. Rivera (Ed.), Disaster and emergency management methods: Social science approaches in application (pp. 11–30). Routledge. ↩

  38. Bandura, A. (2001). Social cognitive theory of mass communication. Media Psychology, 3(3), 265–299. https://doi.org/10.1207/S1532785XMEP0303_03 ↩

  39. Kilic, N., & Simsek, N. (2019). The effects of psychological first aid training on disaster preparedness perception and self-efficacy. Nurse Education Today, 83, 104203-104203. https://doi.org/10.1016/j.nedt.2019.104203 ↩

  40. Wallace, Z. C., Keys-Mathews, L., & Hill, A. A. (2015). The role of experience in defining tornado risk perceptions: A case from the 27 April 2011 outbreak in rural Alabama. Southeastern Geographer, 55(4), 400-416. https://doi.org/10.1353/sgo.2015.0035 ↩

  41. Drost, R. (2013). Memory and decision making: Determining action when the sirens sound. Weather, Climate, and Society, 5(1), 43-54. https://doi.org/10.1175/WCAS-D-11-00042.1 ↩

  42. Han, W., Ada, S., Sharman, R., Rao, H. R., University at Buffalo, University of Illinois at Urbana-Champaign, & Selçuk University. (2015). Campus emergency notification systems: An examination of factors affecting compliance with alerts. MIS Quarterly, 39(4), 909-930. https://doi.org/10.25300/misq/2015/39.4.8 ↩

Suggested Citation:

Hyman, A., & Richmond, J. (2024). How College Students’ Home Region Influences Their Risk Perceptions and Behaviors During Tornados (Natural Hazards Center Weather Ready Research Report Series, Report 11). Natural Hazards Center, University of Colorado Boulder. https://hazards.colorado.edu/weather-ready-research/how-college-students-home-region-influences-their-risk-perceptions-and-behaviors-during-tornados

Hyman, A., & Richmond, J. (2024). How College Students’ Home Region Influences Their Risk Perceptions and Behaviors During Tornados (Natural Hazards Center Weather Ready Research Report Series, Report 11). Natural Hazards Center, University of Colorado Boulder. https://hazards.colorado.edu/weather-ready-research/how-college-students-home-region-influences-their-risk-perceptions-and-behaviors-during-tornados