Armed personnel in educational settings: bridging policy and communication for safer schools Open Access

In 2023, there were 48 active shooter incidents, only three of which occurred in educational settings (Federal Bureau of Investigation, 2024), indicating that school shootings are rare. Nevertheless, several recent, high-profile school shootings across the United States (US) demonstrate that such acts of violence persist (Borum et al., 2010; Gorbett, 2017; Wang et al., 2020). In response to these tragedies and the loss of innocent life, legislation in many states allows for armed security in schools. The National Conference of State Legislatures identified 45 states having policies allowing the possession of firearms on K-12 school grounds by law enforcement (LE), with 21 states allowing firearms by non-LE school security (National Conference of State Legislatures, 2023). As of 2023, nine states have legislation specifically allowing school employees to carry firearms on K-12 school grounds (National Conference of State Legislatures, 2023, Florida Statute § 30.15 Public Safety, 2018, Kansas Statute § 75-7c10, Missouri Statute § 571.107, South Dakota Statute §13-64-1, Tennessee Code § 49-6-816, 2016, Wyoming Statute § 21-3-132, Idaho Statute § 18-3302C, Oklahoma House Bill 2527, Texas Education Code § 37.0814, 2023). In addition to these policies, many states have attempted to or are attempting to pass laws allowing armed school employees in public schools, charter schools and private schools (National Conference of State Legislatures, 2023). Other states do not specify guidelines, but current legislation leaves it open for school boards to allow for armed non-LE personnel on their campuses (Jarzombek-Torralva et al., 2025).

One result of these legislative actions is the increased number of armed, non-LE personnel who may be called upon to respond in the event of an active shooter. In Texas, recent legislation mandating an armed security presence on every school campus includes a provision for non-LE to fill the role of armed security. One of these types of non-LE security is guardians, and recent legislation affords districts the flexibility to define specific guardian roles and expectations (Texas Education Code § 37.0814, 2023). In the current study, researchers investigated the responsibilities and expectations of guardians in Texas to gain a better understanding of the level of communication occurring between districts and their responding LE agencies (LEAs). With consideration of key components of interagency threat response – coordination and collaboration (Federal Bureau of Investigation, 2025) – researchers examined the alignment between school districts and LEAs regarding their collaborative understanding of guardian responsibilities.

As one of the states allowing armed school employees (National Conference of State Legislatures, 2023), Texas has utilized guardians since 2008 (Walker and Sampson, 2018). The use of guardians has increased greatly in Texas over the years, with about 70 districts using them prior to 2014 (Domain, 2014), 180 districts using them in 2018 (Wood and Hampton, 2021) and 224 districts (22% of all districts in Texas) utilizing them in 2023 (Vaughn et al., 2024). Data from 2023 revealed that among districts with non-LE personnel, 80.4% of districts reported having a teacher, 71.4% a principal and 62.9% a superintendent serving in that capacity. These non-LE personnel primarily worked in rural or suburban schools (Texas School Safety Center, 2024; Vaughn et al., 2024).

The surge in guardian numbers can be attributed to several factors, including high-profile mass shootings such as Sandy Hook Elementary, Santa Fe High School and Robb Elementary (Isbell et al., 2019; Vaughn et al., 2024; Comer et al., 2023). These incidents heightened public awareness and concern over school safety, prompting many districts to take proactive steps to protect students. The Robb Elementary tragedy directly led to the passage of Texas House Bill 3 (HB3). Effective September 1, 2023, HB3 requires Texas districts to have at least one armed security officer at each campus during regular school hours (Texas Education Code § 37.0814, 2023). This bill provides three options to meet the mandate, including a: (a) school district peace officer, (b) school resource officer (SRO), or (c) commissioned peace officer to be employed as security personnel (Texas Association of School Boards, 2023). Alternatively, if a district lacks funding or does not have access to qualified candidates from the previous list, the school district can claim a good cause exception, which provides the option of employing a school marshal or guardian (i.e. a district employee or other person with whom the district contracts) (Texas Education Code § 37.0814, 2023). For the districts opting for guardians, their board of trustees are responsible for developing and maintaining written authorization (i.e. a Guardian Plan) outlining the armed security officer regulations (Texas Education Code § 37.0814, 2023; Penal Code 46.03(a)(1)(A)).

The good cause exception affords districts lacking funding and access to other LE personnel an option to meet the armed security mandate. This exception is important, as safety and security needs and perspectives across school districts often differ (DeVos et al., 2018; Isbell et al., 2019; Texas Association of School Boards, 2023). However, it is important to understand the factors involved when choosing to implement a good cause exception. School district peace officers and/or SROs have a great deal of training (including 728 hours of basic training required by the Texas Commission of Law Enforcement) and natural networks of communication with local LEAs (Texas Education Code § 37.0812). A commissioned peace officer employed as security personnel has extensive training and understanding of LE and how they operate (Texas Association of School Boards, 2023). However, the armed security options under the good cause exception (i.e. school marshals and guardians) have less required training and potentially fewer natural connections to LEAs (Texas Association of School Boards, 2023). Guardians require less training than school marshals (15–20 hours versus 80 hours, respectively) and are the most common option used under the good cause exception (224 school districts employed guardians compared to 41 districts employing school marshals in 2023) (Texas School Safety Center, 2024; Vaughn et al., 2024). Because of their widespread use and minimal training, guardians were the focus of this study.

Guardians are school district employees or contracted individuals required to have a license to carry (LTC) and complete a 15–20-hour school safety certification course (Texas Education Code § 37.0814, 2023; Texas Government Code 411.1901, 2013). The LTC requirements include completing an online application, undergoing a state and FBI criminal background check, and passing written exams and shooting proficiency tests (Texas Department of Public Safety, 2024). School safety certification courses must include training on the protection of students, interaction of license holders with first responders, tactics to deny an intruder classroom/school facility entry and methods to increase a license holder’s accuracy with a handgun while under duress (Texas Government Code 411.1901, 2013). Despite these standardized requirements, the content and duration of the training vary greatly due to instructor differences and the characteristics of training facilities available across the state (Vaughn et al., 2024).

Moreover, the flexibility allowing school boards to define the roles of guardians (Texas Education Code § 37.0814, 2023) raises curiosity about guardian responsibilities and how their duties are being communicated to others (e.g. district stakeholders and LEAs). The flexibility for schools to craft a Guardian Plan to fit their needs is important for Texas schools (DeVos et al., 2018; Isbell et al., 2019; Texas Association of School Boards, 2023), but it is equally important to consider the potential challenges associated with flexible plans.

Legislation around guardians does not list specific responsibilities (Texas Education Code § 37.0814, 2023), and school policies tend to concentrate on who will be authorized to carry firearms with only minimal guidance regarding responsibilities (Weiler et al., 2020). Ideally, guardians would be utilized to respond to an active shooter situation and thereby decrease response time, as this was the motive for their creation (Walker and Sampson, 2018). However, part of the allure of Guardian Plans is the flexibility they provide to districts to fit their specific needs as opposed to utilizing other armed security options. Once a district decides to use guardians, the specific responsibilities are flexible. Some districts may want their guardians to handle other threats beyond active shooters, such as bomb threats and natural disasters (Blair and Schweit, 2014). Other districts may want their guardian(s) involved in other school safety activities, such as patrolling school grounds or monitoring extracurricular events.

Although flexibility to define guardian responsibilities is favorable for many schools, one consequence is there are diverse combinations of armed security personnel and varieties in safety protocols on different campuses. Some districts rely heavily on guardians, while other districts use different combinations of security (DeVos et al., 2018; Isbell et al., 2019; Vaughn et al., 2024). Texas is particularly diverse in the kinds of armed security found in schools. In 2023, 412 Texas districts reported having a school district-operated police department, 420 reported contracting LE officers and 417 reported employing guardians and/or school marshals (Texas School Safety Center, 2024). Current literature provides little information about guardian responsibilities, instead focusing on policies about training and the decision-making behind implementing Guardian Plans (Domain, 2014; Jones, 2016; Ratliff, 2014; Hunter, 2018). This omission in literature leaves a significant gap in understanding the specific responsibilities assigned to guardians. As information about guardians is not easily accessible through public records (primarily due to concerns about protecting identities), it is critical that research be conducted to explore how districts are using guardians and what communication exists between districts and LEAs.

Prior research on communication between LEAs and districts has highlighted the need for these two entities to come together more often. For example, past research emphasized the need to clearly define SRO responsibilities (Ryan et al., 2018; Valerio et al., 2024) due to the wide variety of roles these personnel assume on campuses (McKenna and White, 2018; McKenna et al., 2016). In the case of SROs, a common communication mechanism is the memorandum of understanding (MOU) which delineates the responsibilities of SROs so that both districts and LEAs are aligned (Valerio et al., 2024; Cray and Weiler, 2011; McKenna and White, 2020). However, in the case of guardians in Texas, there are no requirements for MOUs between districts and LEAs.

Success in crises is highly dependent upon the level of prior communication (Lopez, 2019) and trust that has been established between responders before the event (Kapucu and Hu, 2014). Unfortunately, this level of communication does not typically happen between districts and LEAs (Lopez, 2019). In a content analysis involving 13 states, non-LE security plans showed limited details about how information was shared between the schools and local LE (Weiler et al., 2020). However, recent research in Texas suggested communication between schools and LE was occurring, as 95.1% of districts informed LE about the presence of guardians in their districts (Texas School Safety Center, 2024; Vaughn et al., 2024). However, the specific information and how it was shared with LE was not revealed. This lack of communication could have catastrophic consequences during crises, as uncoordinated responders may misjudge situations and/or miscommunicate during critical moments.

Although the intention of the armed security personnel mandate in Texas was to provide support in the event of an active shooter (Walker and Sampson, 2018), it is unknown if guardians are being asked to fulfill this responsibility, what they are being asked to do once LE arrives on a scene, what their day-to-day responsibilities (outside of active threats) include and what kinds of information (if any) are being shared between districts and LEAs. The purpose of this study was to determine the scope of guardian responsibilities and gain a better understanding of the level of communication occurring between districts and LEAs.

The widespread use of guardians in Texas raises concerns that miscommunication between districts and LEAs could potentially lead to catastrophic misunderstandings, such as guardians being mistakenly perceived as threats. To address these issues, researchers developed a survey to gather insights from both districts and LEAs regarding the responsibilities of guardians during active threats, both before and after LE arrives on the scene, and how both entities expected guardians to respond. The survey also asked both districts and LEAs about what other responsibilities were part of guardian expectations. Researchers then conducted a matching pairs analysis between school districts and their responding LEA (hereinafter all references to a district’s LEA will refer to the primary responding LEA for that particular district) to assess whether their expectations aligned or misaligned regarding guardian responsibilities.

This study used two separate surveys to gather data from districts and their responding LEAs, although each survey had similar questions. Both surveys included one select-all-that-apply question and two open-ended questions. Surveys were designed to determine the responsibilities of guardians from both districts’ and LEAs’ perspectives and how much communication/interaction occurred between them.

Researchers utilized the TxSSC District Audit Report to identify districts that employed guardians (Texas School Safety Center, 2024). Of these districts, a list of matching LEAs was compiled. LEAs included city police departments, sheriff’s offices, constable agencies and district police departments. Since multiple LEAs may service the same district, LEAs in the sampling frame were prioritized by jurisdiction and contacted accordingly: school district police agencies were targeted first, followed by sheriff’s offices, constable agencies, and finally, city police departments. The targeted participants for this study included superintendents from the districts (n = 226) and the lead LEA officer (i.e. sheriff, constable, police chief, etc.) from the responding LEA (n = 195).

For the quantitative analysis, participants’ responses were included if they had matching responses from the district and the responding LEA on the one select all that apply survey question. Of the 139 districts that submitted the survey, 123 completed the select-all question. Of the 93 LE personnel who submitted the survey, 79 completed the select-all question. Of these submissions, there were 50 matching responses from the district and the LEA. These 50 matching pairs were used in the quantitative analysis.

Twenty-two percent (22%) of the total possible matched pairs were included in the study (50 of 226). As the study required responses from two different participants (i.e. matching pairs) for inclusion, it stands to reason that the response rate was lower because non-response by either participant of the pair resulted in an overall non-response. Response rates by individual entity who completed all of the select all that  apply questions were much higher, with 54% of districts (123/226) and 41% of LEAs (79/195) responding. It is important to note that for many specialized populations, there is often no well-established “acceptable” survey response rate. Police officer response rates are sometimes lower than other populations due to worries about anonymity and distrust of researchers, among other factors (Nix et al., 2019).

Even though nonresponses can lead to bias, response rates alone are not strong predictors of nonresponse bias (Groves and E., 2008; Yeager et al., 2011). The researchers in this study have no reason to believe responders in this sample were markedly different from non-responders, other than the possibility that some of the LEAs that were responsible for multiple districts may have experienced survey fatigue and only responded for one of the districts in their jurisdiction. Thirty-one (31) of the 195 LEAs that were contacted had jurisdiction over multiple school districts.

For the qualitative analysis, 65 matching pairs of districts and responding LEAs completed at least one of the two open-ended questions in the survey. For one question, 29 responses were removed because of nonresponse from one of the pair (total n = 36). On the other question, 30 responses were removed for nonresponse from one of the pair (n = 35).

Institutional review board approval was granted by the authors’ university to send an email invitation with a link to the survey and consent language to district superintendents (n = 226) and lead LEA officers (n = 195). The survey landing page reiterated consent and that participation in this research was voluntary. Surveys took approximately 15 min to complete, and responses were recorded in Qualtrics. In line with procedures to increase response rates as outlined by Dillman et al. (2014), reminder emails were sent every seven days for seven weeks. The initial and follow-up emails yielded overall response rates of 54.4% from districts and 40.5% from LEAs (reasonable rates for an email survey) (Daikeler et al., 2020; Dillman et al., 2014).

Participants indicated on a select all that apply question the responsibilities guardians had in their districts (from a list of eight common responsibilities of school security). Participants had the option of selecting “unknown” if the respondent was unaware whether a guardian held that responsibility or “other” if they participated in roles beyond those listed. Selection of a particular item indicated the respondent did believe the guardian fulfilled that responsibility (coded as 1), and non-selection of that item indicated they did not (coded as 0). Responses of “unknown” or “other” were not considered in this analysis. This coding strategy is supported by research indicating that non-selection of a select all option yields similar results to independent questions being asked in a binary (yes/no) manner (Lewis and Sauro, 2021).

Researchers conducted frequency analyses on matching school district and LEA data (n = 50) to examine trends in response patterns across school districts and LEAs. Then, the alignment between school districts and LEAs was evaluated for each response option. Alignment was defined as both the school district and their LEA selecting the same responsibility (aligning that the item was a responsibility) or both not selecting the same responsibility. In some cases, the matching pairs aligned in their selection of that responsibility on the survey, and that responsibility met the original intent of implementing guardians. In other cases, matching pairs aligned, but that responsibility is unspecified in whether it meets the original intent of implementing guardians.

Situations in which a district and LEA had different responses for the same item were characterized as misalignment. Misalignment occurred when the (a) district selected the response option for a given responsibility when the LEA did not, or (b) LEA selected the response option when the district did not. All examples of misalignment suggest potential miscommunication or a lack of shared knowledge between the two entities. Figure 1 includes details about what was determined to be alignment vs. misalignment.

Two tests were used to assess alignment. Cohen’s kappa (κ) was calculated for each district–LEA comparison to quantify the level of alignment beyond chance. κ values range from −1 to 1, where 0 indicates alignment no better than chance, and higher positive values indicate better alignment. κ reflects the overall consistency between district and LEA reports for each responsibility. To assess whether misalignment tended to favor one group over the other, an exact McNemar’s test (binomial) was conducted. This test examines pairs of misaligned responses (cases where districts’ and LEAs’ responses differed), allowing researchers to evaluate whether districts were more frequently selecting the response option when responding LEAs were not, or vice versa (i.e. whether misalignment is systematically biased in one direction). The exact McNemar’s test was selected over the traditional chi-square approximation because it is more appropriate when the number of discordant pairs is small (fewer than 20), which was the case for several responsibilities. Because Cohen’s κ adjusts for alignment expected by chance based on how often each group says “Yes” or “No” overall, it can suppress κ values when both groups frequently have aligned responses. For these responsibilities, high shared alignment was not considered attributable to chance but reflected meaningful similarities in district and LEA expectations. As a result, κ may underestimate practical alignment, while McNemar’s test more directly identifies whether misalignment, when present, is systematically directional.

For the qualitative analysis of open-ended responses, data from the two questions from matching pairs were independently coded. Coding was based on alignment or misalignment of district and LEA views of guardian responsibilities during an active shooter event (first question, n = 36) and once LE arrived (second question, n = 35). Researchers calculated the percentage of matching pairs that aligned and discussed how the open-ended responses between districts and LEAs were similar or dissimilar.

Alignment statistics indicated poor to slight alignment between district and LEA responses across responsibilities (κ values ranged from approximately −0.16 to 0.29), suggesting that even when districts and LEAs selected the same response option, the level of consistency beyond chance was quite low. As a result, researchers looked at the McNemar results to gain a better understanding of misalignment patterns.

The McNemar results revealed instances when the direction of misalignment was significantly different (Table 1). For example, if significantly more LEAs responded “Yes” when their corresponding districts responded “No” than vice versa, this resulted in a difference in direction (p̂ represents the probability of misalignment in this direction; moreover, 1- p̂ represents the probability of misalignment when LEAs responded “No” and districts responded “Yes”; a p̂ of 0.5 represents an equal split of LE responding “No” and districts responding “Yes” or LE responding “Yes” and districts responding “No”). McNemar tests revealed cases of misalignment across all eight responsibilities wherein a higher percentage of school districts indicated guardians were fulfilling the specific responsibility when LEA indicated they were not (p̂ < 0.5; Table 1).

Although there was misalignment across all eight responsibilities, some of the responsibilities are potentially more dangerous, as they put guardians in direct contact with LEAs on campuses (Figure 2). These responsibilities, referred to as high-stakes responsibilities, include responding to active threats, monitoring athletics and extracurricular events, patrolling school grounds and responding to active threats other than shootings. If guardians fulfill these responsibilities and LEAs are unaware, guardians may be mistaken as threats during emergency situations. Figure 2 shows the breakdown of the responses related to high-stakes responsibilities.

Data indicated 62% of matching pairs aligned in their responses regarding whether guardians would or would not respond to an active shooter event. However, 38% of matching pairs had misaligned responses. In addition, 6% of all districts and 36% of the LEAs indicated their guardians would not respond to an active shooter. These findings indicate districts and LEAs were sometimes on different pages about this responsibility, and guardians were sometimes not being asked to fulfill their original purpose of responding to active shooter threats.

Data from this study indicated 54% of matching pairs aligned in their perspectives about guardians monitoring athletics and extracurricular events (24% of pairs both selected this responsibility, and 30% of pairs did not). The remaining 46% of matching pairs were misaligned in their views. In 36% of those misaligned cases, districts reported guardians were at those events when LEAs did not. In most cases (66%), LEAs indicated guardians would not be monitoring athletics and extracurricular events. These findings indicate that in a high percentage of cases, LEAs believed guardians were not monitoring these events when they were, potentially leaving guardians to be perceived as threats if the LEAs were called to respond.

Only 22% of matching pairs both selected that patrolling school grounds was part of the responsibilities. A larger 42% of matching pairs were misaligned in their perspectives on this responsibility, with 32% of districts indicating their guardians were patrolling when their responding LEA did not report the same. In fact, a large majority of LEAs across all matching pairs (68%) indicated guardians were not patrolling school grounds. These findings indicate that in some cases, LEAs believed guardians did not patrol school grounds when they did. These misaligned situations could lead to a guardian being perceived as a threat as they patrol school grounds.

Finally, data revealed that only 40% of matching pairs both selected that responding to other active threats was part of the responsibilities of the guardians. The majority (54%) of matching pairs were misaligned in their understanding of this responsibility. Most districts indicated their guardians would respond to other active threats (78%), while only 56% of responding LEAs indicated guardians would respond to other active threats. If non-LE security are fulfilling these roles and the LEAs are unaware, guardians may be mistaken as threats.

To further examine the differences between the matching district and LEA responses, both respondents provided open-ended answers about their perceptions of the guardian responsibilities. For this analysis, two open-ended responses were coded based on the percentages of matching pairs that aligned or misaligned regarding guardian responsibilities. The first question, “During the initial response to an active shooter, what is expected of the guardian?” (n = 36 matched pairs) had 47.2% of the district and responding LEAs with aligned responses. For example, one district indicated “engage the threat ASAP”, while the responding LEA indicated “to respond and neutralize the threat.” Conversely, 52.8% of the matching pairs indicated misalignment in responses. For example, one LEA indicated the responsibility of the guardian was “the safeguarding of children”, whereas the accompanying school district indicated the guardian should “get armed and go to the shooter to take out the active threat.” In another example, the LEA wrote “stop the killing”, while the accompanying school district indicated “take care of the students in their care. If no students in their care, then assess the situation.” Overall, the percentages found in the qualitative findings trended in a similar manner to the quantitative McNemar results, with school districts and LE often differing in terms of their expectations of responsibilities.

The second open-ended question analyzed was, “Upon arrival of LE in response to an active shooter, what is expected of the guardian?” (n = 35 matched pairs). Slightly more than half (51.4%) of the district and LEA participants provided aligned responses. For example, one district indicated guardians should, “turn over to LE,” while the matching LEA indicated, “release duties to LE, clearly identify themselves as guardians and disarm ASAP.” Conversely, 48.6% of the matching pairs were notably misaligned in their responses. For example, one LEA indicated the expectation of the guardian was to “put down the weapon immediately and allow LE to take over,” whereas the matching district indicated the guardian should “stay in the classroom.” Another example can be found where the LEA wrote, “take a protective position and evacuate students and teachers,” while the matching district indicated, “they are to meet with LE.” These differences reflect inconsistent domain orientation, where districts and LEAs are not fully aligned toward the same operational goals during crisis response.

The purpose of this research was to assess the level of collaboration occurring between districts and LEAs regarding guardian responsibilities. Findings revealed significant misalignment, with at least 36% misalignment on guardian responsibilities between districts and LEAs based on quantitative data. This discrepancy was particularly evident in high-stakes responsibilities such as responding to active shooter events, patrolling school grounds and monitoring extracurricular activities. Qualitative data further emphasized this misalignment, with matching pairs consistently misaligning around 50% of the time in regard to perceived guardian roles. Such inconsistencies highlight the potential for confusion, miscommunication and compromised response coordination during emergencies. Without a unified understanding of responsibilities, there is an increased risk of operational challenges, including delayed responses, unnecessary casualties or harm to guardians by responding officers.

Legislation in Texas allows flexibility in Guardian Plans to accommodate diverse factors such as district size, LEA response time, and available security personnel. While this flexibility is beneficial, the absence of clear communication between districts, LEAs and guardians can result in incongruent domain orientation and unintended outcomes. For instance, differences in interpretation regarding whether guardians should “engage the shooter” or “protect the children” highlighted the discord between organizations and the necessity for precise role definitions. To minimize these risks, open and ongoing dialogue among all stakeholders is essential to ensure expectations are clearly outlined, responsibilities are understood and operational coordination is effective. The use of guardians as armed personnel in schools is an emerging practice (National Conference of State Legislatures, 2023). Given this growing trend, it is imperative for districts and LEAs to collaboratively establish “best practices” tailored to their needs. These practices should not impose a one-size-fits-all approach but rather provide a structured framework that aligns guardian responsibilities with district-specific needs. Importantly, expecting guardians—who are private citizens—to fully replicate LE tactics is impractical (Blair et al., 2013; Harris and Lurigio, 2012; Martaindale and Blair, 2019). Instead, all relevant parties (e.g. school boards, school administration, police, guardians, etc.) should be involved in robust collaboration to gain relevant insight and determine the responsibilities their guardian(s) need to fulfill.

For example, when determining if guardians should respond to an active shooter event, stakeholders should assess multiple situational factors, including but not limited to the background and training of the guardian, the primary role of the guardian at the school (e.g. teacher, administrator, or other staff), the distance between buildings, the availability of other armed security personnel on campus (e.g. other guardians, SROs), and the response time of external LEA. Beyond active shooter response, districts must establish guidelines for ongoing guardian training, collaboration with LE and participation in relevant drills and emergency exercises. These considerations ensure that guardian responsibilities are tailored to the specific needs of each district while maintaining alignment with broader LE protocols.

To operationalize these efforts, districts should document guardian responsibilities within formal written plans, such as Guardian Plans or Emergency Operations Plans. Written plans should be clear, consistent, practiced, and refined regularly (U.S. Department of Education Office of Elementary and Secondary Education and Office of Safe and Healthy Students, 2013). The process of developing these plans should include input from relevant stakeholders, including boards of trustees, superintendents, LE, administration, guardians themselves and other stakeholders (e.g. LE officers) who might encounter guardians during emergency situations. Once these responsibilities are defined, they must be clearly communicated to all stakeholders through formalized and documented training programs. Structured training programs that cover specific guardian responsibilities included in the written plans will ensure that all stakeholders know and understand what guardians will be doing in all situations, thereby preventing inconsistencies in expectations. Ensuring that all parties are working from the same guidelines and have had opportunities to study and learn those guidelines, regardless of district-specific circumstances, is critical for effective emergency response.

Moreover, collaboration should not end with the creation of written policies and implementation of introductory training programs. Districts should establish structured communication protocols, including scheduled meetings between districts, LEAs and other relevant stakeholders to review and update guardian policies, as well as reinforce stakeholders’ awareness and create institutional memory. Important stakeholders include not only guardians and primary LEAs but also other individuals who may encounter guardians during emergency situations. Sharing this information in a private, face-to-face setting, such as a district’s safety and security meeting, can help ensure mutual understanding and expectations. After the initial meeting, all parties should establish ongoing sharing practices, including determining the frequency and location of future meetings and identifying attendees. These future meetings should include reviewing details of written plans, discussing modifications, developing actionable steps to implement any modifications, and making other adjustments to ensure clarity and consistency.

When shaping future policy, it is crucial to provide flexibility that considers the unique situations of each district (e.g. size, proximity to responding LEA, access to other types of security personnel, etc.). Legislative flexibility allows districts to align guardian responsibilities to specific needs. However, this flexibility in policy should be balanced with structured collaboration, communication and training. Only through clear communication can guardians and LE work together effectively during emergencies, reducing the likelihood of miscommunication.

The sample contact list was limited to Texas and was based on districts that reported using guardians before the mandate for armed security on all Texas campuses took effect on September 1, 2023 (although data were collected after the mandate took effect). Although it is probable additional districts adopted guardians before the current survey was distributed, the sampling frame for this study was sufficient to provide a foundational understanding of how guardians were being used in districts. Furthermore, although the current study only included data from districts and their responding LEAs (thereby excluding some data from individual districts and LEAs that were not matching), the matching pairs data were important to gain a fundamental understanding of the level of communication occurring between districts and responding LEAs regarding guardian responsibilities.

An additional limitation involved not knowing, for certain, who completed the survey. The surveys were initially sent to verified email addresses of the district superintendent (school survey) and the lead LE official of the responding LEA (LEA survey). The recruitment email instructed the recipient to respond to the survey or forward it to the individual with the most knowledge about guardian responsibilities in the district. So, in certain circumstances, completion of the survey may have been delegated to someone else at the school district and/or LEA who was considered by the initial survey recipient to be better equipped to respond to the survey. It is important to note that with this limitation, results could differ across school personnel and/or LE officers depending on who is asked about guardian responsibilities and their role in the district or LEA. For this study, we have no reason to believe the initial survey was sent to anyone other than whoever was most qualified to respond. Nevertheless, if either superintendents or lead LE officials delegated the survey response to someone with limited knowledge, that situation reinforces the need for more intentional communication between all stakeholders regarding guardian responsibilities.

Finally, qualitative responses may not capture the full scope of guardian roles in active shooter events, as respondents may have listed only select duties they deemed relevant. As a result, some misalignments could stem from incomplete information rather than actual differences in expectations.

In terms of practical applications, future studies should investigate the creation and development of resources to facilitate meaningful conversations between districts and LEAs. These supports should consider the unique situations of districts and include evidence-based strategies for customizing the expectations and responsibilities of non-LE security personnel to meet the district’s needs while considering other variables such as school size and proximity to responding LEAs. Future studies should also consider if and/or how these resources and conversations they initiate affect the quality of written documentation and the shared understanding of guardian responsibilities by all applicable stakeholders. Research that occurs “on the ground” (e.g. case studies) in school districts may also be important to determine the effectiveness of all changes resulting from the collaborative process. This future research should also gain the perspectives of the guardians on campus, as it is important to identify where guardians are receiving their understanding of their roles given misalignment between districts and responding LEAs. While this study focused on representatives who were most knowledgeable about how guardians are implemented on their campuses, other district and LE personnel may be less informed. Future research should therefore examine how knowledge about guardians is communicated, taught, and reinforced across all school and LE roles. This can be explored through mixed-methods and qualitative research approaches to capture levels of awareness across schools and LEAs, thereby better understanding, across roles, the implementation of guardians and dissemination of guardian-related information as interagency collaboration evolves.

Another important opportunity for further investigation involves examining guardian training and school-based emergency exercises. Other results from the current study that were outside the scope of this article revealed most LEAs were not involved in guardian training. Better collaboration in the future may lead to districts and LEAs prioritizing training together, but the effectiveness of that training needs to be studied. Furthermore, it will be important to monitor how increased training affects execution of emergency drills and stakeholder confidence regarding emergency preparedness.

Finally, it is important to expand the scope of the current study to include additional schools in Texas and other states that use non-LE security. As previously mentioned, there are other states that allow school employees to have firearms on campuses. Only Texas and Florida have more well-defined policies, yet there is limited evaluation on how effectively these policies are implemented. Research including more of these states will identify additional focus areas and ultimately provide more generalizable results, leading to better-informed policies and improved practices.

This research underscores the importance of clear collaboration and communication between districts and LEAs in defining the responsibilities and expectations of non-LE security. The flexibility provided by legislation in Texas allows districts to tailor non-LE (guardian) responsibilities to their unique needs, but without structured and ongoing dialogue, this flexibility could lead to unintended outcomes. The misalignments between districts and LEAs found in this research highlight the need for formalized processes and shared strategies to ensure all stakeholders are on the same page. Other states considering similar legislation should integrate flexible but clear guidelines for guardians to mitigate potential risks, ensuring written plans are both adaptable to the environment and communicated with consistency. Moving forward, the development of evidence-based practices addressing these concerns will be essential to enhancing the safety and efficacy of armed employee plans across other states and educational environments.