Should I Stay or Should I Go? Determining Parental Preferences for School Severe Weather Decisions

While plenty of research has been devoted to preparedness and response activities of both schools and private residents, to my knowledge, no studies shed light on parents’ feedback on severe weather actions of schools, notably sheltering-in-place and early dismissal decisions. While schools have authority in making their severe weather protocol, limiting or omitting parent input may place students at greater risk, especially in a tornado-prone region with a greater density of manufactured homes and a more socially vulnerable population. This study attempts to obtain parent preferences for their children on school days under a tornadic threat.

A Qualtrics survey was distributed to 500 K-12 parents. This survey included questions related to parent demographic variables, child and school characteristics, perceptions of infrastructure and tornadic intensity, resource and actionable items, tornado knowledge and experience of the parent, and the parent’s perspective on school actions and communications. Recruitment requirements were that individuals were parents/guardians of K-12 age children, currently living in the U.S., and were a demographic representative of the nation. An additional screener was added to limit participation to 13 tornado-prone states (i.e., AR, AL, GA, KY, LA, MS, OK, SC, TN, and TX) in the Southern U.S. to ensure that parents were accustomed to severe weather based early dismissals and provide a greater pool of potential participants to fulfill sample size needs.

Multinomial logistic regression was selected to analyze the data with a stepwise selection process. The results of a general early dismissal tornadic scenario (i.e., no mention of the timing of the dismissal) show a slim majority of parents (51%) prefer their child return home before a threat emerges in comparison to nearly 47% wanting their child to remain in school until the event passes. In contrast, when asking parents where they wish their child to shelter based on the timing of the dismissal, both morning and afternoon dismissals saw nearly 75% of parents want their child to return home before a severe threat occurs.

For a general dismissal scenario, it was observed that greater confidence in school personnel keeping their child safe during a crisis, not having an alternate sheltering plan at home and greater confidence in a child’s school withstanding tornado-force winds made a parent more likely to want their child to shelter in place at school. In contrast, guardians were more unlikely to want their child to remain in school if they were non-white, if they perceived communications between themselves and the school during normal operations as good, more so than excellent, and if their confidence in detached single-family brick or cinderblock homes was slightly lower.

With the morning dismissal scenario, parents were more likely to want their child to remain in school if they had greater confidence in school personnel keeping their child safe and had higher assurance that their child(ren)’s school building could withstand tornado-force winds. Parents were also more likely to want their children to remain in school if they were not in the home regularly during school hours. Guardians were more likely to desire an early dismissal if they had better confidence in brick or cinderblock homes and a worse perception of the school’s severe weather communications.

For the afternoon dismissal scenario, like each of the other models before it, parents were more likely to want their child to stay in school if they had greater confidence in the school personnel keeping the students safe and in the resistance of the school building itself. Similar to the general scenario, guardians would have a higher chance of wanting their child to be in school if the family lacked an alternate sheltering location. Caretakers were less likely to want their child to shelter in place at school if they felt that school severe weather communications were average to good in comparison to good or excellent. Unlike other models, having more confidence in one’s housing enduring a strong tornado (EF-2+) made a guardian less likely to want their child to be in school when the event occurs.

This is the first study to conduct an assessment of parents on tornado actions in schools. These results indicate that parents are influenced heavily by the schools regarding their trust in personnel, perceptions of infrastructure, and dissemination of information. Additionally, perceptions of housing type, tornadic intensity, and actionable items like another place to shelter can sway parents to want their child to return home. These insights provide us with a better understanding of school-parent relationships, the importance of infrastructure, and give us future pathways to explore decision-making and protocols for severe weather events. For example, school districts, especially those in high mobile home density areas or lower-income neighborhoods, may need to contemplate how to raise the trust of K-12 guardians to shift preferences and prevent parents from desiring their children to return to unsafe structures. Likewise, caretakers should evaluate their home’s durability to hazards. A desire for reunification before an event is understandable, but they may be putting their child in harm’s way because of flawed perceptions.

This study provides a baseline on preferences and perceptions related to severe weather decisions in schools. Future projects will be devoted to uncovering the preferences of children during these situations as well as what training, plans, and procedures bus drivers have when transporting students during early dismissals, particularly for those transferring students with disabilities. What methods schools can use to boost the parents’ confidence during crises is also warranted. Additional research should also consider comparing the building codes of both schools and housing stock within a district to determine if sheltering in place at school or sending children back to their homes is the more appropriate, less harmful protocol. Early dismissals are still a novel response. We should continue investigating these decisions and their effect on relevant stakeholders and determine the best response for each district in tornadic environments.