Influence of School Level Socioeconomic Status and Racial
Diversity on Schoolwide Positive Behavior Support Implementation
Research Statement
This evaluation brief examines the capacity of schools of varying levels of
socioeconomic and racial diversity to implement Tier I (Universal) of schoolwide
positive behavior support (SWPBS) with integrity. The three main evaluation questions addressed
are: (a) “What percentage of schools are able to achieve 80%-80% implementation
status on the School-wide Evaluation Tool (SET) within 1 year?” (b) “Is school
socioeconomic status associated with implementation effectiveness?” and (c) “Is
school racial diversity associated with implementation outcomes?”
Method
The data sources for this evaluation brief are: (a) School-wide
Evaluation Tool (Sugai, Lewis-Palmer, Todd & Horner, 2001) data submitted through
thehttp://www.pbssurveys.org website and
(b) the Common Core of Data provided by the U.S. Department of Education’s
National Center for Education Statistics (NCES). For the purpose of this analysis, we examined the
first post-implementation SET schools’ submitted to PBS Surveys between 2005-06
and 2007-08. Preschools, private schools, and alternative schools were excluded
from analyses.
Sample
A total of 890 schools nested within 299 districts in 20
states across the United States and were included in the analysis. A
majority of schools were elementary level schools (n = 606), followed by middle (n = 205) and high school levels (n =
73). An additional six schools served non-traditional grade level spans (e.g.,
6-10 or K-12). Approximately 33% of
schools were located an urban locale (n =
296), 28% were suburban (n = 250),
and 39% were located in a rural locale (n = 344). Average student enrollment for
elementary schools was 452.95 (SD=195.59),
average middle school enrolment was 651 students (SD=271.68), and average high school enrollment was 1120.05 students
(SD=808.17). The average number of
full-time classroom teachers was 30.52 (SD=14.48)
at the elementary level, 43.37 (SD=19.26)
at the middle school level, and 67.98 (SD=51.31)
at the high school level.
Results
What is the Probability a School will Achieve 80%-80% on the SET within
1 Year?
Results
from this sample indicate that, on average, approximately 61% of schools who receive training in tier I of SWPBS will achieve 80%-80%
implementation status on the SET within one year. While additional schools
achieved the implementation criteria after one year, we focused this analysis
only on the SET scores after one year of implementation.
Socioeconomic Status of Student Population
The socioeconomic status (SES) of
the student population, as measured by the percentage of students qualifying
for free or reduced price lunch (FRL), was not significantly associated with 80%-80%
attainment within 1 year (χ2 (4) = 8.18, p =0.09).
Note. Two schools were excluded from analyses due to
missing data.
As
figure 1 illustrates, there was very little difference between very high (less
than 10% FRL) and very low (more than 75% FRL) SES schools with regards to
implementation status at the end of one year. In fact, approximately 59% of very
high SES and 61% of very low SES schools attained 80%-80% status within one
year. Examining the pattern of implementation data across schools did not
reveal noteworthy trends. Sixty-five percent of schools with 11-25% FRL, 56% of
schools with 26-50% FRL, and 67% of schools with 51-75% FRL all attained 80%-80%
within one year.
Figure
1. Number and Proportion of Schools
Attaining 80%-80% Status within One Year
Racial Diversity of Student Population
The diversity of student population,
as measured by the percentage of racial minority students enrolled, was significantly
associated with 80%-80% attainment within 1 year (χ2 (2) = 8.53, p <.05). However, the relation between student racial
diversity and probability of attaining 80%-80% status was not linear, and schools
with higher levels of student racial diversity did not differ substantially
from their low diversity counterparts.
Note: Low
Diversity = less than 25% minority enrollment, Medium Diversity = 25-50%
enrollment, and High Diversity =50% or more minority enrollment. One school was
excluded from analyses due to missing data.
As figure 2 illustrates, Medium Diversity schools with
minority enrollments between 25-50% had the highest percentage of schools
attaining 80%-80% status within 1 year (70%). In comparison, 57% of Low Diversity and 59% of
High Diversity schools attained 80%-80% status within 1 year.
Figure
2. Proportion of Schools Attaining 80%-80% Status by Level of
Student Racial Diversity
Summary of Findings
Results
from these preliminary analyses suggest that a majority of schools (>60%) initiating
implementation of SWPBS will successfully attain 80%-80% on the SET within 1 year. Although, a sizeable
minority (40%) may require a longer period of time to meet this benchmark, this
finding provides further evidence of the feasibility of SWPBS implementation in
a very large and diverse array of public school settings.
A second noteworthy finding was that
the socioeconomic status of the student population was not significantly
associated first year implementation outcomes. In fact, very low SES schools were
almost as equally likely to attain 80%-80% status within one year as their very
high SES counterparts. Although very low SES schools were somewhat
under-represented in this sample, the hypothesis that socioeconomic status
significantly advantages (or disadvantages) schools’ implementation efforts was not supported.
Finally, although the level of racial
diversity among the student population was significantly associated with first
year implementation outcomes, the relation between diversity and SWPBS implementation
is complex. The overall pattern in the data observed suggested that medium
diversity schools were most likely to attain 80%-80% in one year. The reason
why medium diversity schools would outperform their low or high diversity counterparts
is an interesting question worthy of further consideration. Although this
finding may be an artifact of this particular sample, higher-order interactions
(e.g., diversity x locale), or unknown differences within the school or
community of medium diversity schools may explain these effects. However, the
hypothesis that the level of racial diversity among the student population
significantly advantages (or disadvantages) schools’ implementation efforts in
a linear way was not supported.
Limitations
Several plausible threats to the validity
of inferences and generalizability of findings are worth noting. First, because schools comprising this sample self-selected to
submit their data to PBS Surveys, selection bias is possible. The
results obtained from this sample may or may not be consistent with similar
studies utilizing a prospective random sampling frame. Second, SET data
collection occurred under uncontrolled (presumably) naturalistic conditions. Error
related to inappropriate SET administration or scoring is possible. Although
the sample size for this study was large, not all demographic groups were
equally represented. In particular, very affluent schools (less than 10% FRL)
were under-represented, and elementary schools were over represented within
this particular sample. Categorical binning schemes used to create SES and
diversity levels are consistent with those employed by the National Center on
Education Statistics Digest of Education Statistics series (see Snyder, Dillow
& Hoffman, 2009), however, studies utilizing different cut-points for
creating these groups may obtain different results. Finally, the results of
this study pertain to implementation outcomes at the end of one year utilizing
a single implementation outcome measure. Results of studies examining implementation
outcomes for longer durations of time or using different SWPBS implementation
measures may vary.
Future Directions
Implementation of comprehensive
Schoolwide interventions such as SWPBS is a complex multi-stage process, often
requiring the coordination multiple systems and supports. A considerable amount
of research to date has focused on processes predictive of implementation
success (see Fixsen, Naoom, Blase, Friedman and Wallace, 2005), however, examination
of the ways in which fixed school level variables such as SES and diversity interact
with implementation demands is an important area for future research. Future research
may wish to examine how other fixed variables such as grade level, school
locale, or enrollment size may (or may not) influence the probability of
implementation success. In all likelihood, these variables can (and probably
do) interact, and the implementation challenges faced by schools with low
levels of diversity and high levels of poverty, as is common in many rural schools,
may be quite different from those faced by highly diverse urban schools with
similar levels of poverty. Future exploration of these potential interactions
may greatly enhance our understanding of the nature and intensity of support
different schools require to achieve implementation
success. Finally, this particular analysis was limited to the examination of
implementation outcomes within a relatively short duration of time. A more extended longitudinal study examining
the predictors of implementation success and sustainability among schools with
varying levels of SES and student diversity may help to clarify whether the
pattern of implementation results observed in this study are maintained over
time.
References
Department of
Education, National Center for Education Statistics (2006). Public Elementary/Secondary School Universe
Survey Data, 2005-06 [Database]. Retrieved August 20, 2008, and
available from Common Core of Data, National Center for Education Statistics
Web site, http://nces.ed.gov/ccd/
Fixsen, D. L.,
Naoom, S. F., Blase, K. A., Friedman, R. M., & Wallace, F. (2005). Implementation Research: A Synthesis of the
Literature. Tampa, FL: Florida Mental Health Institute, The National
Implementation Research Network (FMHI Publication #231).
Snyder, T.D., Dillow, S.A., & Hoffman, C.M. (2009). Digest of Education
Statistics 2008 (NCES 2009-020). National Center for Education
Statistics, Institute of Education Sciences, U.S. Department of Education.
Washington, D.C.
Sugai, G., Lwis-Palmer, T., Todd, A., & Horner, R.H. (2001). School-wide evaluation
tool. Eugene: University of Oregon.