Skip Navigation
IES    
NCES
National Center for
Education Statistics
Menu
Contact
Tools
ED.gov
Newsflash
Search NCES
Annual Reports and Information Staff (Annual Reports)

  • Preprimary, Elementary,
    and Secondary Education

  • Postsecondary
    Education

  • Population Characteristics
    and Economic Outcomes

  • International
    Comparisons

  • School Crime
    and Safety

  • Education Across
    America

  • Spotlight

Preprimary, Elementary, and Secondary Education

Safety and Security Practices at Public Schools

Last Updated: July 2024
|
This indicator also appears under School Crime and Safety.
Download PDF
Download PDF
Between 2009–10 and 2019–20, the percentage of public schools that reported having one or more security staff present at school at least once a week increased from 43 to 65 percent, although the percentage in 2021–22 (61 percent) was lower than in 2019–20.
Schools use a variety of practices and procedures to promote the safety of students, faculty, and staff.1 The School Survey on Crime and Safety (SSOCS) collects data on school safety and security practices by asking public school principals about their school’s use of safety and security measures,2 as well as whether their school had written procedures for responding to selected scenarios and whether it had emergency drills for students. SSOCS also asks schools about the presence of security staff and the availability of trainings for classroom teachers or aides on school safety and discipline provided by the school or school district.

Select a subgroup characteristic from the drop-down menu below to view relevant text and figures.

All available findings Level of institution Locale Racial composition Size Socioeconomic status (SES) Time series
Figure 1. Percentage of public schools that used selected safety and security measures: School years 2009–10 and 2021–22
Hover, click, and tap to see more for all figures on this page.
Users can select any combination of available years and categories in any figure format (i.e., bar or table). The initial selection of years and categories may not be optimal in different views.
Bar | Table
A confidence interval is a range of values that describes the uncertainty surrounding an estimate. Throughout the Condition of Education, confidence intervals are calculated as the estimate +/- the margin of error, based on a 95 percent level of confidence. This means that there is 95 percent certainty that the range includes the true or actual value of the statistic.
Confidence Interval
Bar | Table
Users can select years at irregular intervals. However, as a result, the distance between the data points will not be proportional to the number of years between them.
X
Embed this figure

— Not available.

1 Prior to 2017–18, the examples of controlled access to buildings included only “locked or monitored doors” and did not include loading docks.

2 Prior to 2015–16, the questionnaire asked only if visitors were required “to sign or check in” and did not include the requirement to wear badges.

3 Prior to 2017–18, the questionnaire asked about prohibiting the “use of cell phones and text messaging devices during school hours.” It did not refer to “nonacademic” use or “smartphones.”

4 For example, a system for reporting threats through online submission, telephone hotline, or written submission via drop box.

NOTE: To estimate the margin of error, the standard error is scaled based on the desired level of confidence in the estimate. Throughout the Condition of Education, margins of error are produced based on a 95 percent level of confidence. Margin of error is calculated as 1.96*standard error. Responses were provided by the principal or the person most knowledgeable about crime and safety issues at the school. This figure only includes the safety and security measures that more than 50 percent of public schools reported using in 2021–22. Figures are plotted based on unrounded data.

SOURCE: U.S. Department of Education, National Center for Education Statistics, 2009–10 and 2021–22 School Survey on Crime and Safety (SSOCS), 2010 and 2022. See Digest of Education Statistics 2023, table 233.50.

In 2021–22, SSOCS asked public schools about 20 safety and security measures. Of these, 10 were reported as being used by more than half of schools. These commonly used measures were generally related to either controlling access to the school during school hours or using communication systems and technology. The most commonly reported safety and security measures (reported by more than 90 percent of schools) were
  • controlling access to school buildings during school hours (97 percent);
  • requiring visitors to sign or check in and wear badges (97 percent); and
  • using security cameras to monitor the school (93 percent).
The 10 safety measures reported in SSOCS by less than half of schools tended to focus on student conduct (for more information, see Digest of Education Statistics, table 233.65). For example, the least commonly reported measures (reported by less than 10 percent of schools) were
  • metal detector checks on students every day (2 percent);
  • requiring clear book bags or banning book bags on school grounds (4 percent); and
  • random metal detector checks on students (6 percent).
Of the 20 safety and security measures asked about in 2021–22, all but 2 were also asked about in 2009–10. Of the measures asked about in both survey years, 10 were reported by a higher percentage of public schools in 2021–22 than in 2009–10 (including 7 of the 10 measures reported by more than half of schools). The largest increases over this period were for
  • using security cameras to monitor the school (from 61 to 93 percent); and
  • providing a structured anonymous threat reporting system3 (from 36 to 62 percent).
Of the 18 measures with data for both 2009–10 and 2021–22, six were reported by a lower percentage of public schools in 2021–22 than in 2009–10 (including 2 of the 10 measures reported by more than half of schools). The largest reductions over this period were for
  • enforcing a strict dress code (decreased from 57 to 37 percent); and
  • prohibiting nonacademic use of cell phones or smartphones4 (from 91 to 76 percent), although there was no consistent trend throughout the period.
[Time series ]
Public schools’ use of various safety and security measures differed by school characteristics during the 2021–22 school year. For example, greater percentages of elementary schools and middle schools than of secondary/high schools reported
  • controlling access to school buildings (98 and 97 percent vs. 94 percent);
  • requiring faculty and staff to wear badges or picture IDs (75 and 71 percent vs. 60 percent); and
  • requiring students to wear uniforms (18 and 16 percent vs. 9 percent).
In contrast, greater percentages of secondary/high schools and middle schools than of elementary schools reported
  • using security cameras to monitor the school (97 and 96 percent vs. 91 percent);
  • conducting random sweeps for contraband (54 and 39 percent vs. 8 percent);
  • enforcing a strict dress code (38 and 49 percent vs. 32 percent);
  • requiring students to wear badges or picture IDs (15 and 14 percent vs. 6 percent);
  • conducting random metal detector checks (14 and 10 percent vs. 2 percent);
  • conducting daily metal detector checks (6 and 4 percent vs. less than 1 percent); and
  • requiring book bags to be clear or banning them (5 and 6 percent vs. 2 percent).
[Level of institution ]
Public schools’ use of various safety and security measures also differed by the racial/ethnic composition of their student body. For instance, schools where 25 percent or less of the students were students of color5 reported the following safety and security measures at lower rates than did schools with higher percentages of students of color (26 to 50 percent, 51 to 75 percent, or 76 to 100 percent):
  • controlling access to school grounds (47 percent vs. 60, 60, and 79 percent, respectively)
  • requiring students to wear badges or picture IDs (3 percent vs. 8, 14, and 15 percent, respectively)
  • requiring students to wear uniforms (2 percent vs. 8, 12, and 43 percent, respectively)
In contrast, schools where 25 percent or less of the students were students of color reported conducting random sweeps for contraband (30 percent) at a higher rate than did schools with higher percentages of students of color (18, 21, and 20 percent, respectively). [Racial composition]
Figure 2. Percentage of public schools with a written plan for procedures to be performed in selected scenarios: School year 2021–22
Hover, click, and tap to see more for all figures on this page.
Users can select any combination of available years and categories in any figure format (i.e., bar or table). The initial selection of years and categories may not be optimal in different views.
Bar | Table
A confidence interval is a range of values that describes the uncertainty surrounding an estimate. Throughout the Condition of Education, confidence intervals are calculated as the estimate +/- the margin of error, based on a 95 percent level of confidence. This means that there is 95 percent certainty that the range includes the true or actual value of the statistic.
Confidence Interval
Bar | Table
Users can select years at irregular intervals. However, as a result, the distance between the data points will not be proportional to the number of years between them.
X
Embed this figure

1 For example, earthquakes or tornadoes.

2 In 2021–22, respondents were asked if they had a “formal plan in place to prepare for and respond to multi-country or worldwide pandemic diseases (such as the coronavirus pandemic, COVID-19).”

3 For example, release of mustard gas, anthrax, smallpox, or radioactive materials.

NOTE: To estimate the margin of error, the standard error is scaled based on the desired level of confidence in the estimate. Throughout the Condition of Education, margins of error are produced based on a 95 percent level of confidence. Margin of error is calculated as 1.96*standard error. Responses were provided by the principal or the person most knowledgeable about crime and safety issues at the school. Figures are plotted based on unrounded data.

SOURCE: U.S. Department of Education, National Center for Education Statistics, 2021–22 School Survey on Crime and Safety (SSOCS), 2022. See Digest of Education Statistics 2023, table 233.65.

Another aspect of school safety and security is ensuring that plans are in place to be enacted in the event of specific scenarios. In 2021–22, more than 90 percent of public schools had a written plan for procedures to be performed in the event of each of the following scenarios:
  • active shooter (96 percent)
  • natural disasters6 (96 percent)
  • suicide threat or incident (94 percent)
  • pandemic disease (92 percent)
  • bomb threats or incidents (92 percent)
From 2009–10 to 2021–22, the percentage of schools that had written plans increased for active shooter scenarios (from 84 to 96 percent) and suicide threats or incidents (from 75 to 94 percent). There was no consistent trend in the percentage of public schools that had a written plan for pandemic disease from 2009–10 to 2021–22, but the percentage was higher in 2021–22 (92 percent) than in 2019–20 (52 percent), the first school year impacted by the coronavirus pandemic.7, 8 [Time series ]
In 2021–22, schools were also asked whether they had drilled students during the current school year on the use of selected emergency procedures. About 96 percent of public schools had drilled students on a lockdown procedure,9 93 percent had drilled students on a shelter-in-place procedure,10 and 93 percent had drilled students on an evacuation procedure.11
Presence of Security Staff and Officers Carrying Firearms
Figure 3. Percentage of public schools with one or more security staff present at least once a week, by selected school characteristics: School year 2021–22
Hover, click, and tap to see more for all figures on this page.
Users can select any combination of available years and categories in any figure format (i.e., bar or table). The initial selection of years and categories may not be optimal in different views.
Bar | Table
A confidence interval is a range of values that describes the uncertainty surrounding an estimate. Throughout the Condition of Education, confidence intervals are calculated as the estimate +/- the margin of error, based on a 95 percent level of confidence. This means that there is 95 percent certainty that the range includes the true or actual value of the statistic.
Confidence Interval
Bar | Table
Users can select years at irregular intervals. However, as a result, the distance between the data points will not be proportional to the number of years between them.
X
Embed this figure

1 Students of color include those who are American Indian/Alaska Native, Asian, Black, Hispanic, Pacific Islander, or of Two or more races.

NOTE: To estimate the margin of error, the standard error is scaled based on the desired level of confidence in the estimate. Throughout the Condition of Education, margins of error are produced based on a 95 percent level of confidence. Margin of error is calculated as 1.96*standard error. Responses were provided by the principal or the person most knowledgeable about crime and safety issues at the school. Security staff include security officers, security personnel, School Resource Officers (SROs), and sworn law enforcement officers who are not SROs. “Security officers” and “security personnel” do not include law enforcement. SROs include all career law enforcement officers with arrest authority who have specialized training and are assigned to work in collaboration with school organizations. Figures are plotted based on unrounded data.

SOURCE: U.S. Department of Education, National Center for Education Statistics, 2021–22 School Survey on Crime and Safety (SSOCS), 2022. See Digest of Education Statistics 2023, table 233.70.

Schools were asked to report whether they had any security staff—including security officers, security personnel, School Resource Officers (SROs), and sworn law enforcement officers who are not SROs—present at their school at least once a week.12 Between 2009–10 and 2019–20, the percentage of public schools that reported having one or more security staff present at school at least once a week increased from 43 to 65 percent, although the percentage in 2021–22 (61 percent) was lower than in 2019–20. [Time series ]
In 2021–22, the percentage of public schools that reported having one or more security staff present at least once a week was
  • higher for secondary/high schools (82 percent) and middle schools (78 percent) than for elementary schools (49 percent) and combined/other schools (38 percent);
  • generally higher for schools with larger enrollment sizes, ranging from 92 percent for schools with 1,000 or more students enrolled to 48 percent for schools with fewer than 300 students enrolled;
  • higher for schools in suburban areas than for schools in cities (64 vs. 56 percent); and
  • higher for schools where 51 to 75 percent of their students were eligible for free or reduced-price lunch13 (66 percent) and 76 percent or more of their students were eligible (62 percent) than for schools where 25 percent or less of their students were eligible (55 percent).
[Locale ] [Level of institution ] [Socioeconomic status (SES) ] [Size]
In 2021–22, about 45 percent of public schools reported having sworn law enforcement officers who routinely carried a firearm. This percentage was lower than the percentage in 2019–20 (51 percent). The percentage of schools that reported having sworn law enforcement officers who routinely carried a firearm differed by school characteristics. For example, in 2021–22, this percentage was
  • higher for secondary/high schools (63 percent) and middle schools (62 percent) than for elementary schools (34 percent) and combined/other schools (24 percent);
  • higher for schools in rural areas (55 percent) and towns (54 percent) than for schools in suburban areas (45 percent) and cities (30 percent); and
  • lower for schools where 76 percent or more of their students were students of color (29 percent) than for schools with lower percentages of students of color (ranging from 46 to 55 percent).
[Time series ] [Locale ] [Level of institution ] [Racial composition]
Figure 4. Percentage of public schools providing training for classroom teachers or aides in specific safety and discipline topics: School years 2019–20 and 2021–22
Hover, click, and tap to see more for all figures on this page.
Users can select any combination of available years and categories in any figure format (i.e., bar or table). The initial selection of years and categories may not be optimal in different views.
Bar | Table
A confidence interval is a range of values that describes the uncertainty surrounding an estimate. Throughout the Condition of Education, confidence intervals are calculated as the estimate +/- the margin of error, based on a 95 percent level of confidence. This means that there is 95 percent certainty that the range includes the true or actual value of the statistic.
Confidence Interval
Bar | Table
Users can select years at irregular intervals. However, as a result, the distance between the data points will not be proportional to the number of years between them.
X
Embed this figure

1 Includes those related to cyberbullying, bullying, violence, and alcohol and/or drug use. Schools that reported providing trainings on two or more of these topics were counted once.

2 Includes early warning signs of student violent behavior; physical, social, and verbal bullying behaviors; signs of self-harm or suicidal tendencies; and signs of students using/abusing alcohol and/or drugs. Schools that reported providing trainings on two or more of these topics were counted once.

3 This item on the questionnaire provided the following examples of mental health disorders: depression, mood disorders, and ADHD. The questionnaire defined mental health disorders as “collectively, all diagnosable mental health disorders or health conditions that are characterized by alterations in thinking, mood, or behavior (or some combination thereof) associated with distress and/or impaired functioning.”

NOTE: To estimate the margin of error, the standard error is scaled based on the desired level of confidence in the estimate. Throughout the Condition of Education, margins of error are produced based on a 95 percent level of confidence. Margin of error is calculated as 1.96*standard error. Includes trainings provided by the school or school district. Responses were provided by the principal or the person most knowledgeable about crime and safety issues at the school. Figures are plotted based on unrounded data.

SOURCE: U.S. Department of Education, National Center for Education Statistics, 2019–20 and 2021–22 School Survey on Crime and Safety (SSOCS), 2020 and 2022. See Digest of Education Statistics 2021, table 233.67b and Digest of Education Statistics 2023, table 233.67b.

In 2021–22, most public schools and school districts provided training on school safety and security to classroom teachers and aides:
  • 93 percent provided trainings on safety procedures (e.g., how to handle emergencies).
  • 88 percent provided trainings on schoolwide discipline policies and practices related to at least one of the following issues: cyberbullying, bullying, violence, and alcohol and/or drug use.
  • 87 percent provided trainings on recognizing at least one of the types of student behaviors or tendencies related to potential safety or discipline problems.14
  • 81 percent provided trainings on classroom management for teachers
In addition to these trainings on procedures, policies, and recognizing risks, a majority of schools also reported providing trainings on intervention strategies to help inform teachers about how to appropriately intervene in various safety-related scenarios involving students. Specifically,
  • 84 percent of schools reported providing training on positive behavioral intervention strategies;
  • 74 percent reported providing training on crisis prevention and intervention; and
  • 67 percent reported providing training on intervention and referral strategies for students displaying signs of mental health disorders.15
The percentage of public schools that reported each of these various trainings on school safety and security were lower in 2021–22 than in 2019–20. For example, 67 percent of schools reported providing training on intervention and referral strategies for students displaying signs of mental health disorders in 2021–22, compared with 72 percent in 2019–20. [Time series ]

1 For general technical notes related to data analysis, data interpretation, rounding, and other considerations, please refer to the Reader’s Guide.

2 In addition to data from the School Survey on Crime and Safety (SSOCS), this indicator uses 2013–14 data from the Fast Response Survey System (FRSS) for select trend analyses. The 2013–14 FRSS survey was designed to allow comparisons with SSOCS data. However, the mode of the 2013–14 FRSS survey differed from that of SSOCS, which evolved over time. Specifically, all respondents to the 2013–14 survey could choose either to complete the survey on paper (and mail it back) or to complete the survey online. All respondents to SSOCS had only the option of completing a paper survey prior to 2017–18. In 2017–18, SSOCS experimented with offering an online option to some respondents. In 2019–20 and 2021–22, SSOCS switched to using primarily an online survey instrument. The 2013–14 FRSS survey also relied on a smaller sample than SSOCS. The FRSS survey’s smaller sample size and difference in survey administration may have impacted the 2013–14 results.

3 For example, a system for reporting threats through online submission, telephone hotline, or written submission via drop box.

4 Prior to 2017–18, the questionnaire asked about prohibiting the “use of cell phones and text messaging devices during school hours.” It did not refer to “nonacademic” use or “smartphones.”

5 Students of color include those who are American Indian/Alaska Native, Asian, Black, Hispanic, Pacific Islander, or of Two or more races.

6 For example, earthquakes or tornadoes.

7 Prior to 2017–18, respondents were asked if they had a written plan that describes procedures to be performed in the event of a “pandemic flu.” In 2017–18 and 2019–20, respondents were asked if they had a written plan that describes procedures to be performed in the event of a “pandemic disease.” In 2021–22, respondents were asked if they had a “formal plan in place to prepare for and respond to multi-country or worldwide pandemic diseases (such as the coronavirus pandemic, COVID-19).”

8 The coronavirus pandemic affected the 2019–20 data collection activities. The change to virtual schooling and the adjusted school year may have impacted the data collected by the School Survey on Crime and Safety. Readers should use caution when comparing 2019–20 estimates with those from other years. For more information, see Crime, Violence, Discipline, and Safety in U.S. Public Schools in 2019–20: Findings From the School Survey on Crime and Safety (NCES 2022-029).

9 Defined for respondents as “a procedure that involves securing school buildings and grounds during incidents that pose an immediate threat of violence in or around the school.”

10 Defined for respondents as “a procedure that requires all students and staff to remain indoors because it is safer inside the building or a room than outside. Depending on the threat or hazard, students and staff may be required to move to rooms that can be sealed (such as in the event of a chemical or biological hazard) or that are without windows, or to a weather shelter (such as in the event of a tornado).”

11 Defined for respondents as “a procedure that requires all students and staff to leave the building. The evacuation plan may encompass relocation procedures and include backup buildings to serve as emergency shelters. Evacuation also includes ‘reverse evacuation,’ a procedure for schools to return students to the building quickly if an incident occurs while students are outside.”

12 “Security officers” and “security personnel” do not include law enforcement. SROs include all career law enforcement officers with arrest authority who have specialized training and are assigned to work in collaboration with school organizations.

13 The percentage of students eligible for free or reduced-price lunch (FRPL) programs is a proxy measure for school poverty. For more information on eligibility for FRPL and its relationship to poverty, see the NCES blog post “Free or reduced price lunch: A proxy for poverty?”

14 Includes early warning signs of student violent behavior; physical, social, and verbal bullying behaviors; signs of self-harm or suicidal tendencies; and signs of students using/abusing alcohol and/or drugs.

15 This item on the questionnaire provided the following examples of mental health disorders: depression, mood disorders, and ADHD. The questionnaire defined mental health disorders as “collectively, all diagnosable mental health disorders or health conditions that are characterized by alterations in thinking, mood, or behavior (or some combination thereof) associated with distress and/or impaired functioning.”

Supplemental Information

CLOSE
CLOSE
Table icon
Table 233.50 (Digest 2023): Percentage of public schools with various safety and security measures: Selected school years, 1999-2000 through 2021-22;
Table icon
Table 233.60 (Digest 2023): Percentage of public schools with various safety and security measures, by selected school characteristics: School year 2021-22;
Table icon
Table 233.65 (Digest 2023): Percentage of public schools with a written plan for procedures to be performed in selected scenarios and percentage that have drilled students on the use of selected emergency procedures, by selected school characteristics: Selected school years, 2003-04 through 2021-22;
Table icon
Table 233.67b (Digest 2023): Percentage of public schools providing training for classroom teachers or aides in specific safety and discipline topics, by safety and discipline training topic and selected school characteristics: School year 2021-22;
Table icon
Table 233.70 (Digest 2023): Percentage of public schools with security staff present at least once a week, and percentage with security staff routinely carrying a firearm, by selected school characteristics: Selected school years, 2005-06 through 2021-22;
Table icon
Table 233.67b (Digest 2021): Percentage of public schools providing training for classroom teachers or aides in specific safety and discipline topics, by safety and discipline training topic and selected school characteristics: 2019-20
CLOSE
Bullying (School Survey on Crime and Safety) Combined/other school Cyberbullying Elementary school Free or reduced-price lunch Locale codes Mental health disorder (School Survey on Crime and Safety) Middle school Public school or institution School Resource Officer (SRO) Secondary/high school
CLOSE
PDF icon
PDF icon
PDF icon
Previous versions of this indicator available in the Indicators of School Crime and Safety reports.
CLOSE

Suggested Citation

National Center for Education Statistics. (2024). Safety and Security Practices at Public Schools. Condition of Education. U.S. Department of Education, Institute of Education Sciences. Retrieved [date], from https://nces.ed.gov/programs/coe/indicator/a19.
IES    
NCES
National Center for
Education Statistics
Explore the Institute of Education Sciences