What are the patterns of office discipline referrals across
grade levels?
Evaluation Brief
Claudia Vincent, Robert Horner, and Seth May
Schools use
office discipline referral (ODR) data to design and evaluate maximally
effective and efficient behavioral support practices for their entire student
populations. Research exists that suggests specific ODR patterns across grade
levels that could indicate when support needs are greatest. Based on ODR data
from one elementary-middle school with grades K-6, Putnam, Luiselli, Handler, and
Jefferson (2003) found that the frequency of ODR increases during the
elementary grades. Similarly, Tidwell, Flannery, and Lewis-Palmer (2003)
examined ODR data from 16 elementary schools and found an increase in ODR rates
from Grade 1 to Grade 5. Focusing both on frequency and type of behavioral
violations in middle school grades, Tobin and Sugai (1999) showed that
frequency of ODR for aggression related behaviors in Grade 6 predicted
frequency of ODR for similar behaviors in the later middle school grades, and
high ODR frequency in the later middle school grades predicted suspensions in
Grade 9. In addition, Tobin, Sugai, and Colvin (1996) demonstrated that the
type of behavior students engage in during Grade 6 predicts the type of
behavior they engage in during Grade 8. Data from the longitudinal High School
& Beyond dataset indicated that prior discipline problems in general and
suspension and truancy in particular predicted dropping out of high school
(Wehlage & Rutter, 1986). More recent high school data indicate that the
drop-out rate of 16-24 year olds decreased from 14.1% in 1980 to 8.7% in 2007
(U.S. Department of Education, 2009). Students dropping out of school in 9th and 10th grades will likely impact the frequency of ODR in 11th and 12th grade. These findings suggest that patterns of overall ODR
frequencies as well as ODR frequencies by type of problem behavior across grade
levels might offer insights into how to allocate resources and expertise to
meet the behavioral support needs of students in varying grade levels. Within
this context, our goal was to examine if the overall ODR frequency patterns
across grade levels that have been documented in the literature can be
replicated with a large national sample of schools using the School-wide
Information System (SWIS; May et al., 2006) to record ODR.
SWIS is a
web-based ODR data collection system that allows schools to record the
occurrence of a number of behavioral violations and subsequent administrative
decisions. Two levels of behavioral violations exist in SWIS: minor violations
and major violations. Minor violations include low-intensity defiance,
low-intensity disruption, inappropriate language, and inappropriate physical
contact. Major violations include abusive language, insubordination, sustained
disruption, and fighting. SWIS users are also encouraged to record their
schools’ overall enrollment to facilitate calculations of ODR rates across
schools with varying student numbers. SWIS does not record enrollment by grade
level. However, the National Center on Education Statistics (NCES) provides this
information for most public schools in the United States.
Method
Our
sample included all schools that (a) were located in the United States, (b)
were not labeled as “alternative,” and (c) used SWIS to record ODR during the
most recent year for which NCES grade level enrollment data were available
(2006-2007). Our dataset included only major ODR by grade level, because many
SWIS user do not consistently record minor ODR and major ODR tend to generate
administrative decisions that can be of significant consequence to the student.
A total
of 2214 schools met all inclusion criteria. These 2214 schools were located in 44
states; Illinois was most frequently represented with 247 (11.2%) schools,
followed by Maryland with 236 (10.7%) schools and Oregon with 223 (10.1%) schools.
Table 1 provides an overview of the sample by school level.
Table 1:
Overview of sample by school level
| |
Number |
Percent |
K-6 |
1437 |
64.9 |
6-9 |
430 |
19.4 |
9-12 |
138 |
6.2 |
K8-12 |
209 |
9.4 |
Total |
2214 |
100 |
We
aggregated major ODR recorded in SWIS by grade level as well as enrollment by
grade level recorded through NCES. It is important to note that some schools
that are labeled “K-6” might only have Grades 1-5, while some schools that are
labeled “6-9” might have Grades 4-9. Therefore, sample sizes for each grade
level are not identical with the sample sizes for each school level. Table 2
provides an overview of the sample sizes by grade level, total enrollment by
grade level, total major ODR by grade level, mean grade level enrollments and
standard deviations, and mean grade level major ODR and standard deviations.
Table 2: Sample size by grade level, total
enrollment by grade level, mean grade level enrollment and standard deviations,
total major ODR by grade level, and mean grade level ODR and standard
deviations.
Sample size by grade level |
Total Enrollment by grade level |
Total major ODR by grade level |
Mean (SD) grade level enrollment |
Mean (SD) major ODR by grade level |
Grade 1 (n = 1452)* |
112,210 |
55,307 |
77.28 (40.73) |
37.44 (51.23) |
Grade 2 (n = 1456) |
108,485 |
58,366 |
74.51 (37.91) |
39.48 (55.56) |
Grade 3 (n = 1458) |
108,020 |
61,930 |
74.09 (38.72) |
41.95 (50.68) |
Grade 4 (n = 1448) |
107,180 |
69,253 |
74.02 (40.50) |
47.25 (57.14) |
Grade 5 (n = 1406) |
104,827 |
77,135 |
74.56 (45.13) |
54.42 (66.17) |
Grade 6 (n = 909) |
118,850 |
139,572 |
130.75 (107.83) |
152.31 (216.36) |
Grade 7 (n = 617) |
113,954 |
184,159 |
184.69 (119.31) |
296.56 (366.01) |
Grade 8 (n = 614) |
113,930 |
184,502 |
185.55 (120.51) |
302.15 (342.26) |
Grade 9 (n = 201) |
45,558 |
84,166 |
226.66 (215.30) |
419.53 (618.55) |
Grade 10 (n = 192) |
38,384 |
57,065 |
199.92 (188.36) |
296.66 (405.78) |
Grade 11 (n = 188) |
35,768 |
41,713 |
190.26 (185.69) |
221.63 (328.15) |
Grade 12 (n = 189) |
32,710 |
28,148 |
173.07 (172.49) |
148.71 (216.55) |
*n indicates the number of schools in
the dataset with the given grade level.
For example, a total of 1452 schools in
our dataset had Grade 1. Within those 1452 schools, a total of 112,210 first
graders were enrolled. The 1452 schools recorded 55,307 major ODR for first
graders. The mean number of first graders in the 1452 schools was 77.28, with a
standard deviation of 40.73. The mean number of major ODR recorded for first
graders in the 1452 schools was 37.44, with a standard deviation of 51.23. We included means and standard
deviations of enrollment to get a sense of the overall density of students by
grade level within a school and the variability of it within our dataset. We
included means and standard deviations of major ODR to get a sense of the
average frequency of ODR by grade level within a school and the variability of
it in our dataset.
To
evaluate frequency patterns of ODR across grade levels, we calculated for each
grade level the number of major ODR per 100 students enrolled by dividing the
number of major ODR for a grade level by the enrollment for that grade level and
multiplying by 100. Major ODR per 100 students represents the number of ODR
given by grade level. One student might receive multiple ODR. In addition, we
calculated the number of students who received a major ODR per 100 students
enrolled by dividing the number of students per grade level who received a
major ODR by the number of students enrolled at that grade level and
multiplying by 100. Students with major ODR per 100 students enrolled
represents the number of unique students per grade level who received a major
ODR.
Results
Overall, results replicated the findings that
have been documented in the research literature. Figure 1 illustrates major
ODR/100 students by grade level. The frequency of major ODR increased steadily
from Grade 1 to Grade 5, increased quite dramatically from Grade 5 to Grade 6
and from Grade 6 to Grade 7, peaked in Grade 9, and then declined quite rapidly
throughout the remaining high school grades. The middle school grades (6-9) clearly
emerged as a period where student behavior support needs are particularly high.
These high support needs might be due to students’ entry into adolescence and
concomitant tendency to challenge adult authority represented by
institutionalized behavioral rules. At the same time, it is interesting to note
that the mean number of students enrolled in Grade 6 (130.75) was almost twice
as large as the mean number of students enrolled in Grade 5 (74.56), and that
mean enrollment increased steadily until it peaked in Grade 9 (226.66) (see
Table 2). This suggests that overall school size might make effective delivery
of behavior support more challenging at a developmental period when it appears
most critical.
|
Figure 1: Major ODR per100 students across
grade levels. Horizontal line indicates average across all grade levels.
Given that the literature suggests that students
who experience behavioral challenges in prior years tend to drop out in high
school (Wehlage & Rutter, 1986), the steady decline in ODR frequency after
Grade 9 is of concern, because it might be a reflection of behavioral support
systems not meeting the needs of those students who have significant support
needs. On the other hand, the
decline in ODR after grade 10 might also be partially due to students’
increasing maturity. It is also worth noting that the frequency of ODR per 100
students exceeds 100 from Grade 6 through Grade 11. This might suggest that
more students in those grade levels are likely to receive multiple ODR than
students in the other grade levels.
The number of unique students who received a
major ODR per 100 students enrolled by grade level is illustrated in Figure 2. The overall trends are similar to the frequency of major ODR
given across grade levels. The number of students who received a major ODR rose
steadily from Grade 1 to Grade 5, increased more sharply from Grade 5 to Grade
6 and from Grade 6 to Grade 7, peaked at Grade 9, and then declined steadily
through Grade 12. This overall pattern is likely due to the same speculations
offered above. During the middle school years (Grades 6-9) more students
receive one or more major ODR than during the elementary school years. The
number of students who receive one or more major ODR declines in the high school
years, possibly because many students with behavioral challenges opted to drop
out of school, or because students are beginning to mature into young adults.
|
Figure 2: Students with major ODR per
100 students enrolled across grade levels. Horizontal line indicates average
across all grade levels.
Discussion
The patterns of ODR frequency in our dataset closely matched
those reported in the literature. These patterns suggest that the behavioral
support needs of school populations increase steadily with grade level until a peak
is reached at Grade 9, when frequency of ODR begins to decline. From a
developmental perspective, these patterns are to be expected. From the
perspective of behavior support delivery, these patterns suggest that support
demand increases with increasing grade level. If one assumes that the decline
in ODR frequency after grade 9 is primarily due to students with high support
needs leaving school, then one could conclude that behavior support delivery might
have to be strengthened in the middle school years. However, the decline in ODR
frequency after grade 9 could also be due to students’ increasing maturity, which
might make behavioral violations as a way to challenge adult authority less
common. In general, our findings might allow schools to allocate resources
within their building to meet the demands of students at varying grade levels
in the most effective and efficient manner.
References
May,
S., Ard, W., Todd, A. W., Horner, R. H., Glasgow, A., Sugai, G., et al. (2006). School-wide information system.
Eugene: Educational and Community Supports, University of Oregon.
Putnam, R.F., Luiselli, J.K., Handler,
M.W., & Jefferson, G.L. (2003). Evaluating student discipline practices in
a public school through behavioral assessment of office referrals. Behavior Modification 27(4), 505-523.
Tidwell, A., Flannery, K.B., &
Lewis-Palmer, T. (2003). A description of elementary classroom discipline
referral patterns. Preventing School
Failure 48(1), 18-26.
Tobin, T.J. & Sugai, G.M. (1999).
Using sixth-grade school records to predict school violence, chronic discipline
problems, and high school outcomes. Journal
of Emotional and Behavioral Disorders 7(1), 40-53.
Tobin, T.J., Sugai, G., & Colvin,
G. (1996). Patterns in middle school discipline records. Journal of Emotional and Behavioral Disorders 4(2), 82-94.
Wehlage, G.G. & Rutter, R.A.
(1986). Dropping out: How much do school contribute to the problem? Teachers College Records 87(3), 374-392.
U.S. Department of Education, National
Center for Education Statistics. (2009). The Condition of Education 2009 (NCES 2009-081). Available at http://nces.ed.gov/FastFacts/display.asp?id=16