Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013 Copyright @ by LDW 2013

Self-Regulated Assignment Attack Strategy:

Evaluating the Effects of a Classroom-level
Intervention on Student Management of
Curricular Activities in a Resource Context
Bryan M. Ness
University of New Hampshire

McKay Moore Sohlberg

University of Oregon

The purpose of this study was to evaluate the impact of a classroom-based

strategy instruction package grounded in self-regulated learning. The Self-
Regulated Assignment Attack Strategy (SAAS) targeted self-regulation of
assignment management and related academic-behavioral variables for
6th grade students in resource support classrooms. SAAS was implemented
by a special education teacher and two educational assistants in three
separate classes, and this study examined implementation and efficacy
using direct observation of student self-regulatory behaviors. Utilizing a
multiple baseline, across-participants research design, the study revealed
positive effects of SAAS on “assignment attack” and teacher-reported stu-
dent behavior during resource support classes. The results demonstrated
that the effect was maintained after external supports were faded and
suggest improved assignment attack behavior may have contributed to
improved assignment completion. A discussion of the results is provided,
addressing the implementation of SAAS and considerations for educators
interested in SRL strategy instruction in resource settings.

Background

Self-regulated Learning
Self-regulation refers to the capacity for modifying behavior according to
internally defined standards. Self-regulation involves self-evaluation of cognitive and
motivational processes that facilitate behavioral modifications based on contextual
demands (Bandura, 1986). Examining self-regulation in educational context refers
to studying processes that allow for cognitive flexibility while students are engaged in
academic work (Boekaerts, Maes, & Karoly, 2005). Requisite in self-regulated learn-
ing (SRL) is metacognitive knowledge; students are aware of what they know or do
not know and have the ability to change their behavior based on that knowledge
(Dinsmore, Alexander, & Loughlin, 2008). Intact metacognition enables students to
apply, monitor, and adapt self-regulated learning strategies (Boekaerts, 1997).
SRL depends largely on a student’s ability to learn and apply strategies
(Boekaerts & Cascallar, 2006). Self-regulatory strategies support learning by help-
ing students balance task demands, motivational constraints, and cognitive processes
(Boekaerts & Corno, 2005). For example, self-monitoring is a cognitive strategy that
promotes awareness of engagement and performance during a learning task (Harris,
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

Friedlander, Saddler, Frizzelle, & Graham, 2005). Self-monitoring enables students to

evaluate productivity and accuracy in relation to task demands (Reid, 1996). Moti-
vational constraints such as success expectancy and intrinsic interests are known to
impact student success. Self-regulated learners manage motivational factors by em-
ploying strategies such as goal setting, planning, and self-rewarding (Metallidou &
Vlachou, 2010).

Self-regulated Learning and Struggling Students

Students who struggle academically often report lower metacognitive strat-
egy use and lower perceived ability to self-regulate their learning compared to peers
(Klassen, Krawchuk, Lynch, & Rajani, 2008). Students with high incidence disabili-
ties, such as attention deficit/hyperactivity disorder (ADHD) and learning disabilities
(LD), are less skilled at self-monitoring (Harris et al., 2005) and strategic planning
(Sikora, Haley, Edwards, & Butler, 2002) compared to their peers. These traits con-
tribute to unproductive and ineffective learning behaviors (Wolters, 2003).
Impairments in self-regulation are often manifested by difficulty organiz-
ing, initiating, and remaining engaged in academic work in the classroom and at
home. Bryan, Burstein, and Bryan (2001) published a review of assignment manage-
ment and completion problems for students identified with learning disability (LD).
Their summary revealed that students with LD spend more time on assignments
due to poor organization and time management compared with peers. In addition,
students who struggle academically lose materials and forget assignments more fre-
quently, resulting in less time engaged in academics (Bryan, Burstein, & Bryan, 2001).
Soderlund and Bursuck (1995) examined assignment completion problems in the
classroom and at home for students identified with behavior disorders. Their results
indicated that assignment management difficulties included distractibility, procras-
tination, and difficulty starting new tasks, as measured by teacher and parent report
(Soderlund, Bursuck, Polloway, & Foley, 1995).

Academic Interventions Grounded in Self-Regulated Learning

SRL behaviors predict overall academic achievement and concept mastery
(Wolters, 1999). Therefore, it is important for educators to understand interventions
that support development of SRL. There is a need to equip educators with instruc-
tional models that are both theoretically sound and easy to implement in complex
educational settings (Schunk, 2008).
Instructional strategies directed at self-regulation skills seek to bolster meta-
cognitive control, motivational control, and academic strategy use (Meltzer, Katzir,
Miller, Reddy, & Roditi, 2004). For example, teaching self-monitoring in classroom
learning contexts has been shown to improve classroom preparation (Gureasko-
Moore, DuPaul, & White, 2007), on-task behavior (Harris et al., 2005), and academic
productivity (Uberti, Mastropieri, & Scruggs, 2004) for students with high incidence
disabilities. Similarly, interventions targeting motivational barriers often incorporate
goal setting and self-efficacy enhancements, and have been associated with improved
reading, writing, and classroom behavior (Schunk & Zimmerman, 2007; Wehmey-
er, Yeager, Bolding, Agran, & Hughes, 2003). Finally, strategy instruction targeting
planning, organization, and self-monitoring has been shown to improve at-home as-

36
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

signment management and completion (Hughes, Ruhl, Schumaker, & Deshler, 2002;
Trammel, Schloss, & Alper, 1994).
A robust body of research supports the effectiveness of self-regulation strat-
egy instruction, however, many of these strategies are implemented using individual
instruction. Individualized strategy instruction may not always be feasible or efficient
in some educational contexts (Boekaerts & Corno, 2005). Behavior support and aca-
demic interventions (e.g., reading remediation curricula) are frequently applied at
classroom or school-wide levels and integrated into the general educational milieu
(e.g., Sugai et al., 2000). SRL strategy instruction has penetrated classroom peda-
gogy in general education, which is thought to benefit all learners, including students
with disabilities (Mastropieri & Scruggs, 2010; Paris & Paris, 2001). However, there
remains a need to demonstrate how SRL strategy instruction may benefit struggling
learners who require more extensive intervention than is typically offered in the class-
room. This is especially true for interventions delivered to groups of students with
disabilities in resource support settings where additional academic and assignment
management instruction are often provided.
Meltzer et al. (2004) developed and implemented a strategy training cur-
riculum with a large sample of students with and without LD. Classroom teachers
were trained to deliver the curriculum, which included self-regulatory strategies cov-
ering the spectrum of content areas and learning strategies (i.e., literacy skills, test
taking, study skills). Daily strategy instruction was provided in the classrooms for six
months. All students improved strategy use and reduced difficulty in reading, writ-
ing, and spelling as measured by teachers, and the effect was larger for students with
LD (Meltzer et al., 2004).
Ness et al. (2010) examined the effects of teaching students to self-monitor
assignment management in a resource setting using a classroom-based organization-
al system. The intervention featured a homework “database” posted in a prominent
place within the classroom. Students learned to record new assignments on home-
work “tickets” and post them on a billboard in the resource room. Students used the
tickets to remember and prioritize unfinished assignments, and using single subject
methodology, the researchers demonstrated a positive effect of the strategy on stu-
dent assignment recall, initiation, and overall task engagement (Ness, Sohlberg, &
Albin, 2010).
The present study sought to extend Ness et al. (2010) by evaluating the ef-
fects of a strategy targeting student “assignment attack” in three different classrooms.
While Ness et al. (2010) demonstrated a positive effect in one classroom, it would
be useful to know if a different assignment attack strategy can produce similar ef-
fects when implemented simultaneously in different contexts. The purpose of this
study was to investigate whether one strategy implemented in three different resource
contexts would produce contextually relevant changes in individual student behav-
ior. A strategy instruction package based on SRL principles was designed and deliv-
ered to groups of students in resource settings. The package, called “Self-regulated
Assignment Attack Strategy” (SAAS), was developed in collaboration with a special
education teacher to explicitly teach students self-regulatory skills related to assign-
ment completion (i.e., planning, organization, and self-monitoring) in the resource
context. The strategy was taught at the group level, so procedural parsimony and

37
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

instructional efficiency were essential considerations in how SAAS was constructed.

Additionally, this study sought to examine the maintenance of student behavior once
SAAS was faded.
Our specific research questions were as follows:(1) What are the effects of
the SAAS across students in different classrooms as measured by “assignment attack”
ratings, teacher-reported assignment completion behavior, and assignment comple-
tion rate; (2) Will SAAS produce an effect that will be maintained once the key inter-
vention supports have been faded?
We hypothesized that students who implemented SAAS would demonstrate
improved self-regulation, as measured by direct observation of a cluster of student
behaviors during in-class assignment completion and teacher report of behavior in
the classroom. We also hypothesized that the effects would be maintained once exter-
nal supports provided by SAAS were faded.

Method

Setting and Participant Selection

Participants were sampled from a middle school in a mid-sized Western city
that housed three different resource support classes for sixth graders. There were ap-
proximately twelve students in the second period resource, and there were six students
and three students in each third period resource class. The school was selected because
the sixth grade special education teacher asked the first author to conduct a strategy
instruction study in her classroom similar to work the researchers had conducted in
a different school. She learned of this work through email communication with other
special education teachers in the district. The special education teacher, who had six
years of teaching experience, all in special education and in the same middle school,
taught the second period resource support class. She supervised educational assistants
facilitating the third period resource support classes. University IRB and school dis-
trict approval were secured prior to the study. Teacher and parental informed consent
were obtained in writing, and student assent was determined verbally.
The participants (N = 3) for this study were selected based on the follow-
ing inclusion criteria: (1) enrolled in sixth grade; (2) qualified for special education
services; (3) enrolled in general education classes, but spent a portion of their day
in a resource room for help managing out-of-class assignments; (4) low academic
achievement as measured by a current grade of either a D or F in at least one class
prior to the intervention. These criteria were generally satisfied by all students in the
resource classes, since academic difficulty was a pre-requisite for resource support.
One student from each class was nominated by the special educator as rep-
resentative of students with poor organization and initiation of assignments, and
these students were selected for data analysis. The nature of the primary outcome
variable (i.e., “Assignment Attack”) precluded simultaneous measurement of each
student in the classroom. It was determined that the duration and complexity of ob-
servational data collection could result in measurement error. Therefore, one student,
deemed as representative of the majority of students by the special education teacher,
was selected from each class to demonstrate the impact of the intervention in three
different classes.

38
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

Adam.
Adam was a twelve-year-old boy who qualified for special education based
on identifications of specific learning disability and speech-language impairment.
His full scale IQ was 94 according to the Reynolds Intellectual Assessment Scale
(RIAS; Reynolds & Kamphaus, 2003) and he qualified for the district free/reduced
lunch program. His academic supports and accommodations included a behavioral
support plan, reduction of large assignments, preferential seating, extended time
for reading and writing assignments, allowing healthy snacks during class, morning
check-in, teacher check for comprehension, and increasing active participation.

Brett.
Brett was a twelve-year-old boy who qualified for special education based
on an identification of emotional-behavior disturbance. His full scale IQ was 86 as
measured by the Wechsler Intelligence Scale for Children 4th edition (Wechsler et al.,
2003) and he did not qualify for free/reduced lunch. Brett’s other academic supports
and accommodations included providing immediate teacher feedback, repeating
directions, preferential seating, re-teaching expected behaviors, and teaching self-
monitoring.

Christina.
Christina was a 12-year-old girl who received study supports due to an iden-
tification of specific learning disability. Her learning disability was secondary to a
diagnosis of phenylketonuria (PKU), which is an inherited metabolic disorder associ-
ated with impaired conversion of the amino acid phenylalanine. Christina’s full scale
IQ was 108 as measured by the Wechsler Abbreviated Scale of Intelligence (Wechsler
et al., 1999), and she did not qualify for free/reduced lunch. Other academic supports
and accommodations included teacher check for comprehension, breaking down
lengthy assignments/projects, reducing workload, providing oral directions, sitting
in close proximity to the teacher, teacher support with the homework planner, and
extended time for assignments.

Dependent Variables
SRL can be difficult to measure since cognitive and motivational strategies
are unique to the individual and tailored for the task at hand (Boekaerts & Corno,
2005). Traditionally, educators and researchers have used questionnaires to measure
self-reported SRL constructs such as learning strategy use, motivational strategies,
and environmental modifications (e.g., Cleary, 2006). This approach is efficient but
does not assess the use of strategies in natural contexts, which is why some researchers
have recommended using direct observation as a method of assessing SRL (Boekaerts
& Corno, 2005). For this study, the researchers utilized both observation and ques-
tionnaires to measure constructs of interest.

Self-regulation behavior during assignment completion.

A rating scale was adopted to quantify observable behavior indicative of
self-regulation in the resource setting. The primary outcome measure of self-regula-
tion was a slightly modified version of the assignment attack scale used in Ness et al.

39
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

(2010). The rating scale quantified four observable student behaviors that reflected
SRL in the classroom: (1) recalling assignment details; (2) gathering necessary mate-
rials; (3) initiating work; (4) task engagement. Each of the four domains were rated
using a five-point scale, so the range of composite assignment attack scores was 4 to
20. Scoring criteria were defined for each domain. For example, with the “recalling as-
signments” domain, the lowest score (1) was operationalized as, “Teacher told student
assignment or showed student the pink sheet/folder. Student had no recollection of
assignment and acts like they have no work.” The highest score (5) meant, “Student
required no support for knowing subject and assignment details.” The operational
definitions for each domain were slightly modified from Ness et al. (2010) to match
vocabulary and contextual variables present in the participating school (e.g., refer-
ence to the “pink sheet” above). The rating scale has not been evaluated for psycho-
metric properties; however, it is theoretically grounded in SRL and produced reliable
results in previous work (Ness et al., 2010).
Assignment completion behavior was measured using teacher-reported
Homework Problem Checklist (HPC; Anesko, 1987). The teachers used the HPC
to evaluate student behavior while working on assignments in the resource setting.
The questionnaire consists of 20 items, and uses a 0-3 Likert-type rating scale. The
HPC sampled teacher’s perception of student behaviors during assignment comple-
tion in the resource room (e.g., “Refuses to do homework assignment”; “Whines
or complains about homework”), homework performance (e.g., “Hurries through
homework and makes careless mistakes”; “Produces messy or sloppy homework”),
and homework management (e.g., “Fails to bring home assignment and necessary
materials”; “Forgets to bring assignments back to class”). While the scale has not been
normed, previous studies have demonstrated the internal consistency (Cronbach al-
pha = .91) and discriminative validity of the HPC (e.g., Anesko, 1987).

Assignment completion rate.

Assignment completion rate was measured to assess the generalized impact
of SAAS in general education courses. Assignment completion rates for math and
social studies were calculated based on the proportion of assignments completed di-
vided by the total number of assignments in a one-month period near the end of the
first and third academic grading periods. The special education teacher assembled
assignment completion data using school-wide electronic grading software. The
math and social studies teachers entered the data, and the special education teacher
provided completion rates to the first author.

Intervention Package: Self-Regulated Assignment Attack Strategy (SAAS)

Using SRL as an intervention framework, SAAS was developed to teach stu-
dents to self-regulate in-class assignment management. The strategy was conceptual-
ized to support the planning, performance, and self-evaluation of academic work,
and featured a four-step sequence specific to the resource context: (1)Planner/Pink
Sheet Check; (2) Pick an Assignment; (3) Prepare Materials; (4) Proceed. SAAS fea-
tured an implementation approach utilizing external aids and purposeful self-regula-
tion instruction to bolster strategic assignment management.

40
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

The 4 P’s Checklist: External Aid for Planning, Performance, and Self-evaluation.
External supports are effective for modeling self-regulated learning process-
es, such as self-monitoring and self-evaluation (e.g., Harris et al., 2005). The external
support for SAAS was a checklist given to students during the resource period (see
Figure 1). The checklist was used to teach the four-step sequence described above.
Students were instructed to review the “4 P’s” checklist at the beginning of class to
support goal setting and organization. Students were then prompted to refer to the “4
P’s” during class to promote self-monitoring of assignment engagement. Finally, stu-
dents were asked to self-evaluate their performance using the “How Did I do?” check-
list corresponding to the “4 P’s” at the end of the resource block. The self-evaluative
portion of SAAS was used to model the cyclical nature of planning, performing, and
self-evaluating learning typified by

Figure 1. The Four P’s Checklist

Plan to fade external supports.

The components of SAAS were intended to facilitate internalization of self-
regulatory processes. First, SAAS was designed as a first letter mnemonic to help stu-
dents remember key behaviors. Since all four behaviors started with the letter “p”, the
checklist was referred to as the “4 P’s”. Second, the “4 P’s” checklist was systematically
faded until students were instructed to say the four steps “in your head” without the
support of a checklist. Fading external supports using self-instruction, or “self-talk”, is
an effective approach for helping struggling students develop self-regulated learning

41
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

practices (Johnson, Graham, & Harris, 1997). The checklist depicted in Figure 1 was
replaced with a version that included only the self-evaluation table and the letter “P”
on each row. Finally, the first-letter-only checklist was replaced by teacher prompts
for the students to use self-instruction.

Model self-regulation through systematic instruction.

SAAS instruction was scripted to promote consistent and effective imple-
mentation. Three instructional goals were explicitly addressed in the script. First,
the teacher was prompted to teach the strategy using direct instruction principles
(i.e., model, guided practice, and evaluate learning) known to facilitate self-regulat-
ed learning for students with disabilities (Butler, 2003). Specifically, the teacher was
prompted to explain the four steps of the checklist, model the procedure, and assess
students’ understanding through questions (e.g., “Pretend I’m a student in 2nd (3rd)
period and the teacher just told me to take out my homework. The first thing I do is
a Planner Check.”) Second, the teacher was prompted to lead students in a discus-
sion about their experiences completing assignments and their need to manage this
academic skill more independently in the future (e.g., “Keeping up on work not only
helps you do better in school and helps lots of students feel better about how they
do in school. Which assignments are easiest to remember? What assignments do you
like best?”) Establishing personal relevance is a powerful predictor in student accep-
tance and use of self-regulation strategies in classrooms (Schunk, 1998). Third, the
teacher was prompted to model self-regulation processes using self-appraisal state-
ments during strategy demonstration (e.g., “While thinking about how you did today,
ask yourselves, ‘Is there anything I could have done differently so Ms. … did not have
to tell me what to do?’”) Teacher modeling of self-appraisal statements is an effective
method to teach SRL strategies (Paris & Winograd, 2001).
To promote consistent instruction across resource classes, the special edu-
cation teacher delivered the scripted instruction described above. Initial instruction
required two days in each class and was conducted at the beginning of each inter-
vention phase (described below). Subsequent to the initial instructional period, the
special education teacher implemented SAAS in Adam’s class, while educational as-
sistants implemented SAAS in Brett and Christina’s classes. SAAS following the initial
instruction period included using the 4 P’s checklist and teacher modeling of self-
regulatory strategies as described above.

Treatment fidelity.
SAAS was designed to target a discrete constellation of behaviors indica-
tive of poor self-regulation. Since the strategy would be implemented in different
classrooms, the researchers collaborated with the special education teacher to maxi-
mize consistent implementation across contexts. Treatment fidelity was measured by
both the first author and research assistant in each classroom during the intervention
and fade experimental phases. Treatment fidelity refers to the accuracy and consis-
tency with which independent variables are delivered (Gresham, MacMillan, Beebe-
Frankenberger, & Bocian, 2000). Following Gresham et al. (2000) recommendations,
both fidelity of intervention delivery (“treatment delivery”) and fidelity of student
implementation (“treatment adherence”) were measured. Intervention delivery and

42
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

adherence fidelity data were measured once a week during each phase. The outcome
variable for both was total percentage.

Social validity.
Perception of SAAS as a socially valid intervention was measured to exam-
ine relationships between fidelity, student outcomes, and replication likelihood (Fos-
ter & Mash, 1999). Teacher perception of SAAS acceptability was measured using
the Treatment Evaluation Inventory-Short Form (TEA-SF; Kelly, 1989). The TEA-SF
consisted of four items, and the teacher rated each item using a five level scale. Also, a
fifth item, an open ended question, was included, asking for teacher input in making
the intervention more acceptable.

Research Design
A multiple baseline across participants design was employed to evaluate
the functional relation between SAAS and changes in student behavior over time.
Single subject designs have been shown to be powerful research tools for evaluating
clinical practice (Horner et al., 2005). Multiple baseline designs, in particular, are
useful for evaluating interventions, enabling both within and across participant data
analysis using individuals as their own controls (Kennedy, 2005). The replication of
effects across different participants at different points in time can be interpreted as
experimental control in multiple baseline studies (Horner et al., 2005). The multiple
baseline design was selected to evaluate SAAS because the nature of the dependent
variable necessitated more in-depth data collection than would be possible using ag-
gregated, classroom-level data.
The research design consisted of four phases per participant: Baseline, Inter-
vention, Fade, and Maintenance. The purpose of the baseline phase was to measure
assignment attack behavior under normal classroom conditions. To adequately de-
fine a stable, predictable pattern of assignment attack for each student, baseline data
were collected until future behavior level and trend could be predicted (Horner et al.,
2005). During the intervention phase, the teacher introduced SAAS into the normal
classroom routine. Assignment attack data were recorded until a clear change in be-
havior level, trend, and/or variability was evident. The decision of whether there was
an intervention effect was based on evidence that assignment attack levels in the inter-
vention phase were outside the range of predicted levels demonstrated in the baseline.
At that point, SAAS was introduced to the next classroom with assignment attack data
recorded for the participating student. When there was evidence assignment attack
data had changed, the external supports were modified during the fade phase. External
supports were then removed, which constituted the maintenance phase.

Data Collection and Analysis

Assignment attack data were collected on a daily basis by the first author and
an undergraduate research assistant. Once the special education teacher prompted
the students to begin working on assignments, assignment attack observation be-
gan. As soon as the student demonstrated behaviors on the rating scale (i.e., recalling
assignment, gathering materials, and initiating work), scores were assigned to each
domain. At the end of the period, the whole-class engagement score was recorded.

43
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

HPC and assignment completion rates were collected pre- and post-intervention to
evaluate the generalized impact of SAAS. Since students often worked on assignments
that could not be completed during one resource period, daily assignment comple-
tion was not recorded. Given the small sample size, descriptive analysis of HPC raw
scores and average completion rate were conducted within participants.
Inter-rater reliability refers to the stability of scores across raters (Prima-
vera, Allison, & V, 1997). Inter-rater reliability for assignment attack was measured
using the exact percentage of agreement. Agreement was calculated by dividing the
number of agreements (i.e., between the first author and research assistant) by the
total number of observations. The first author and research assistant observed stu-
dents simultaneously in the classrooms and agreement was assessed following the
observation. Exact agreement meant the same cumulative total, plus or minus two
points (e.g., scores of 15 and 17 would be considered in agreement). Agreement was
also calculated for each of the four domains constituting the cumulative assignment
attack score. Exact agreement was defined as the same domain score, plus or minus
one point.
Reliability training was completed during ten in-class observations with stu-
dents not included in final data collection. Exact agreement for assignment attack
scores reached 90%. Subsequently, inter-observer agreement was measured for 28%
of the assignment attack observations, which were distributed across participants
during each of the four experimental phases. Agreement for cumulative assignment
attack was 89%. Agreement for each of the assignment attack domains was as follows:
“Recall” was 93%; “Gather Materials” was 85%; “Initiation” was 85%; and “Whole-
class Engagement” was 88%.

Results
This study was designed to evaluate the effects of SAAS on self-regulated as-
signment management in a resource setting for adolescents who qualified for special
education. SAAS was implemented as prescribed by the special education teacher
across three classes. Initial instruction required approximately 10-15 minutes per day.
SAAS was applied to Adam’s classroom first followed by Brett’s and Christina’s classes
respectively. Subsequently, the special education teacher provided two additional in-
struction days for each group, once to introduce fading of the 4P’s checklist and once
to eliminate external supports. Approximately 60 minutes of instruction were pro-
vided for each class across all phases of the study. Data were obtained to answer re-
search questions pertaining to student assignment attack, assignment completion be-
havior, assignment completion rate, and student and teacher strategy implementation.

Assignment Attack
In-class assignment attack behavior was recorded over the course of thirteen
academic weeks, not counting two weeks for winter break and one week for spring
break. A total of 97 data points were collected across the three participating students:
31 for Adam, 32 for Brett, and 34 for Christina. The assignment attack data are pre-
sented in Figure 2.

44
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

Figure 2. Assignment Attack Data

Adam and Brett revealed a clear change in level across phases with high av-
erage assignment attack during the intervention, fade, and maintenance phases rela-
tive to baseline. Christina’s data revealed a high degree of variability and rising trend
during the baseline phase, which confounded interpretation of the intervention ef-
fect. Replication of treatment effects was evident across participants at three different
points in time, although Christina’s data suggest other variables also impacted her
assignment attack data.
Adam’s baseline data revealed a low and stable baseline with assignment
attack scores ranging between four and eight (M = 5.8). Assignment attack in the
intervention and fade phases was characterized by considerable variability, ranging
between six and twenty, with a slightly delayed but distinct increase in level (M =
14.1). One data point late in the intervention phase was low (6), but this should be in-
terpreted with caution since he was involved in a fight before the class. Adam’s assign-
ment attack scores sustained the high level during the maintenance phase (M = 15.3).
Brett’s assignment attack data were slightly higher in level at baseline (M =

45
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

8.4) compared to Adam’s baseline. His data were also more variable, ranging between
five and twelve, and there was a slight downward trend at baseline. There was some
data overlap between Brett’s baseline, intervention, and fade phases reflecting a high
degree of variability, but there was a rapid and large increase in level for the interven-
tion and fade phases (M = 14.8). Brett’s assignment attack level remained high during
the maintenance phase (M = 17).
Christina’s data reflected a high degree of variability in the baseline phase,
with assignment attack scores ranging from four to eighteen. Also, Christina’s assign-
ment attack began to trend up starting at the low on day nine. Despite this trend,
her mean assignment attack score during baseline was 8.6, similar to Brett’s baseline
average. Christina’s intervention and fade assignment attack levels were consistently
high (M = 16.1) with very little variability compared to baseline, with scores ranging
from twelve to nineteen.
To support visual analysis of assignment attack data, an overall effect size
was calculated. Effect sizes are useful for interpreting the magnitude of a treatment
effect, particularly for group research designs. However, effect size calculations in
single subject research is complicated by dependence on visual analysis and a lack of
clear interpretation guidelines (Parker & Hagan-Burke, 2007b). Parker et al. (2007a)
propose Percentage of All Non-Overlapping Data (PAND) as a suitable effect size
calculation in single subject research, as it involves visual analysis of overlapping
data points and enables researchers to directly calculate the Phi coefficient (Parker &
Hagan-Burke, 2007a). Another benefit of PAND is that Phi can be used for effect size
and power analysis, using the same guidelines as Pearson’s r correlation coefficient
(Cohen, 1992; Parker & Hagan-Burke, 2007b).
PAND was calculated by first counting overlapping data points in the base-
line and intervention phases (i.e., phases A and B) according to the procedures de-
scribed in Parker et al. (2007a) for multiple baseline designs. This revealed 3 over-
lapping data points for Adam, 1 overlapping data point for Brett, and 3 overlapping
data points for Christina, yielding a total of 7/61 (11.4%) overlapping data points.
Second, Phi was calculated by setting up a balanced contingency table (i.e., a 2 X 2
matrix), cross-tabulating the proportion of overlapping and non-overlapping data
in the baseline and intervention phases. The ratio of non-overlapping data on a per-
centage basis in the intervention phases (28.3/34) was subtracted from the ratio of
overlapping data in the baseline phases (5.7/66) resulting in a Phi = 0.75, suggesting
a large magnitude of change across participants (Cohen, 1992).

Assignment Completion Rate

The average assignment completion rate increased or was unchanged across
participants with the exception of Adam, whose completion rate in social studies de-
creased. The average initial assignment completion rate in math for all three students
was 88.3%, and the final average assignment completion rate for math was 89.3%.
Adam’s math completion rate increased 3% while Brett’s and Christina’s comple-
tion rates were unchanged. The average initial assignment completion rate in social
studies was 78.3%, and the final average social studies completion rate was 87.3%,
representing an average 9% increase. Brett and Christina’s social studies assignment
completion rates increased 12% and 25% respectively, but Adam’s final assignment

46
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

completion rate in social studies dropped 10%. In summary, the general trend for as-
signment completion rate was positive, as indicated by average percentage improve-
ment. However, there was variability across students and across classes, indicating the
overall effect was minimal.

Teacher-perceived Assignment Completion Behavior

The post-intervention HPC data revealed an improvement during in-class
assignment completion behavior. For this measure, lower scores indicated fewer per-
ceived problem behaviors and higher perceived efficiency during assignment comple-
tion. The data are presented in Table 1. It should be noted that different teachers
completed Christina’s pre- and post-intervention HPC questionnaires, which could
have affected the reliability of the scores. In summary, these data revealed improve-
ment in teacher-perceived student behavior.

Table 1. Homework Problem Checklist Data

Students HPC-Teacher
(n = 3) (max = 60)
Initial Final Change
Adam 44 42 -2
Brett 32 14 -18
Christina 25 14# -9
Average 33.7 28 -9.7
# Completed by a different teacher.

Treatment Fidelity
Treatment delivery for the special education teacher (Adam’s class) ranged
from 80% to 100% (M = 97%). The range of delivery fidelity for the educational as-
sistant in Brett’s class was 80% to 100% (M = 96.8%). The treatment delivery fidelity
mean for the other educational assistant (Christina’s class) was 100%.
Fidelity of treatment adherence ranges and average percentages were calcu-
lated for each student: Adam was 33% to 100% (M = 77.7%); Brett was 33% to 100%
(M = 89.9%); and Christina was 100%. There were three instances of 33% treatment
adherence between Adam and Brett. The respective teachers were provided with feed-
back following each of these three sessions to point out the neglected elements. It
should be noted that these three instances were in the first two weeks of the respective
intervention phases and researcher feedback resulted in subsequent 100% adherence.
These data imply a high degree of teacher implementation and student adherence.

Social Validity
The questionnaire data indicated strong teacher endorsement of SAAS as
an effective intervention. The responses indicated generally positive responses about
the usability of the strategy (3/4 “agree” responses), and the teacher reported that she

47
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

intended to repeat SAAS at the beginning of next school year writing, “…I am excited
to implement it (the 4P’s) from the beginning of the year.” She shared that she plans
to extend the strategy by reinforcing student self-reflection by posting the 4P’s check-
list in a prominent spot in the classroom and asking students to write self-reflective
statements at the end of class.

Discussion
The goal of this study was to evaluate the effects of a strategy instruction
package on assignment initiation, engagement, and behavior for middle school stu-
dents in a resource support setting. We hypothesized that students who implemented
SAAS would demonstrate behaviors consistent with self-regulation, as measured by
assignment attack observation and teacher report of behavior, and that the effects
would be maintained once external supports were faded. The results supported our
hypotheses and revealed positive outcomes on assignment attack and teacher-report-
ed behavior following implementation of SAAS, but the generalized impact of these
improvements on assignment completion was minimal. We demonstrated that im-
proved student behaviors can be maintained after external supports are faded.

Improved Assignment Attack

The primary contribution of this study was the demonstration of improved
self-regulatory behavior and maintenance of the effect following implementation of
SAAS. Assignment attack data for Adam and Brett revealed a positive effect of SAAS
on assignment attack. The assignment attack measure was designed to capture behav-
ior indicative of self-regulation during assignment management. The overall impact
of improved assignment attack was more independent management of in-class as-
signments, as reflected in teacher-reported change in behavior.
Christina’s assignment attack level was similarly high with little variation
following SAAS implementation. However, the variability and upward baseline trend
limit the degree to which this change was attributable to SAAS. One plausible expla-
nation for Christina’s improved baseline data is a change in her resource class. Her
group was rearranged following the start of the second academic quarter. Elective
class scheduling and changes in some students’ individualized education plans result-
ed in Christina’s study support class size decreasing from six students to two or three
students, depending on the day. This change in class size had the effect of reducing
distractions and increasing the educational assistant’s availability to provide struc-
tured learning, thereby reducing demands to self-regulate assignment attack. One
solution to this problem for related studies would be selecting classes with larger,
more stable student enrollment.
In previous research (Ness et al., 2010), it was evident that observable
self-regulatory behavior could be improved (i.e., become less teacher-dependent)
by teaching discrete strategies in a resource setting. Findings from this study pro-
vide a replication and extension of that effect by demonstrating improved assign-
ment attack in students across three different classrooms. The present study provides
evidence supporting the efficacy of incorporating self-regulation strategies into the
classroom milieu, and raises questions that may be addressed in future research. Ad-
ditional studies are necessary to demonstrate the utility of SAAS beyond improving

48
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

assignment attack behavior. Self-regulated learning is a broad construct, theorized

to explain student motivation and cognitive control across academic tasks. Future
research activities may include expanding the intervention curriculum and corre-
sponding measures to capture the generalized impact of improved self-regulation.

Fidelity of Implementation and Acceptability in Natural Context

In their review of assessment and intervention principles grounded in self-
regulated learning theory, Beokaerts and Corno (2005) illustrate how skill acquisition
occurs as self-regulation skills are embedded in the classroom. A powerful benefit
of classroom-based strategy instruction is peer and teacher modeling of the target
skill (M. Boekaerts & Corno, 2005). This study demonstrated that the SAAS inter-
vention package can be implemented by a busy middle school teacher with high fi-
delity and ease. The demonstration of implementation fidelity is critical for evalu-
ating classroom-based interventions since naturalistic school contexts are nuanced
and may contain elements that render carefully designed experimental instructional
techniques invalid. Exploratory work preceding this study demonstrated that more
complex, individually administered assignment management strategies are ineffec-
tive due to students’ inability to manage the requisite components. The SAAS model
may provide educational professionals a framework for conceptualizing and teaching
self-regulation-based strategies that transcend contextual barriers.

Measuring the Impact of Assignment Attack

The assignment attack rating scale was sensitive to changes in student behav-
iors consistent with increased self-regulation in the resource context. The challenge
in SRL measurement, however, lies in understanding how processes relate directly to
academic success (Schunk, 2008). For this study, it was hypothesized that improved
assignment attack would impact academic success through increased assignment
completion. The assignment completion results were highly variable, culminating
in relatively low overall improvement in completion rates. An alternative measure
of academic achievement was considered – namely changes in assignment accuracy.
It is possible that SAAS contributed to more accurate assignment completion, par-
ticularly in response to self-evaluation and progress monitoring prompts. However,
assignment accuracy may be influenced by many factors outside the scope of SAAS,
including quality and quantity of classroom attendance (Valiente, Lemery-Chalfant,
Swanson, & Reiser, 2008). Additionally, students undertook many assignments dur-
ing the resource class that could not be completed in one class, which complicated
assignment completion measurement. More sensitive measures are required to un-
derstand the link between improved assignment attack and academic achievement
following implementation of SAAS.

Limitations
There are several limitations of this study that should be considered when
interpreting the data and evaluating the external validity of the results. First, the de-
pendent variables were measured by the researchers and interventionists. As this was
not a blinded study, the threat of measurement bias is present despite a scientific
ethic to remain objective. Second, the assignment attack rating scale is sensitive to

49
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

changes in student behavior and yields reliable data. However, the validity and inter-
nal consistency of the measure are unknown, which affects the interpretability of the
data. Finally, although this study presents an effect that was replicated in different set-
tings for different students, the generalization of the findings is limited by the small
sample size and should be restricted to students and settings closely matching those
described in this study.

Summary
This study evaluated the impact of a SRL strategy instruction package on
assignment attack for middle school students in resource support classrooms. The
intervention was implemented as prescribed, validating the utility of the instructional
approach. The results indicated positive effects on student assignment attack and as-
signment completion behavior, although the generalized effect on assignment com-
pletion rate was not clear. While additional research is necessary to examine the overall
impact of improved assignment attack, the findings demonstrate that classroom-based
strategies can improve self-regulation for struggling students in resource settings.

References

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Engle-
wood Cliffs, NJ: Prentice-Hall.
Boekaerts, M., & Cascallar, E. (2006). How far have we moved toward the integration of theory
and practice in self-regulation? Education Pscychology Review, 18, 199-210.
Boekaerts, M., & Corno, L. (2005). Self-regulation in the classroom: A perspective on assess-
ment and intervention. Applied Psychology: An International Review, 54(2), 199-231.
Boekaerts, M., Maes, S., & Karoly, P. (2005). Self-regulation across domains of applied psychol-
ogy: Is there and emerging consensus? Applied Psychology: An International Review,
54(2), 149-154.
Boekaerts, M. (1997). Self-regulated learning: A new concept embraced by researchers, policy
makers, educators, teachers, and students. Learning and Instruction, 7, 161-186.
Bryan, T., Burstein, K., & Bryan, J. (2001). Students with learning disabilities: Homework prob-
lems and promising practices. Educational Psychologist, 36(3), 167-180.
Butler, D. (2003). Structuring instruction to promote self-regulated learning by adolescents
and adults with learning disabilities. Exceptionality, 11(1), 39-60.
Cleary, T. (2006). The development and validation of the self-regulation strategy inventory -
self-report. Journal of School Psychology, 44, 307-322.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.
Dinsmore, D., Alexander, P., & Loughlin, S. (2008). Focusing the conceptual lens on metacog-
nition, self-regulation, and self-regulated learning. Education Pscychology Review, 20,
391-409.
Foster, S., & Mash, E. (1999). Assessing social validity in clinical treatment research: Issues and
procedures. Journal of Consulting and Clinical Psychology, 67(3), 308-319.
Gresham, F., MacMillan, D., Beebe-Frankenberger, M., & Bocian, K. (2000). Treatment integ-
rity in learning disabilities intervention research: Do we really know how treatments
are implemented? Learning Disabilities Research & Practice, 15, 198-205.
Gureasko-Moore, S., DuPaul, G., & White, G. (2007). Self-management of classroom prepared-
ness and homework: Effects on school functioning of adolescents with attention
deficit hyperactivity disorder. School Psychology Review, 36(4), 647-664.

50
 Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

Harris, K., Friedlander, B., Saddler, B., Frizzelle, R., & Graham, S. (2005). Self-monitoring of
attention versus self-monitoring of academic performance: Effects among students
with ADHD in the general education classroom. The Journal of Special Education,
39, 145-156.
Horner, R., Carr, E., Halle, J., McGee, G., Odom, S., & Wolery, M. (2005). The use of single-
subject research to identify evidence-based practice in special education. Exceptional
Children, 71(2), 165-179.
Hughes, C. A., Ruhl, K. L., Schumaker, J. B., & Deshler, D. D. (2002). Effects of instruction in
an assignment completion strategy on the homework performance of students with
learning disabilities in general education classes. Learning Disabilities: Research &
Practice, 17(1), 1-18.
Johnson, L., Graham, S., & Harris, K. R. (1997). The effects of goal setting and self-instruction
on learning a reading comprehension strategy: A study of students with learning
disabilities. Journal of Learning Disabilities, 30(1), 80-91.
Kennedy, C. (2005). Single-case designs for educational research. New York: Allyn and Bacon.
Klassen, R., Krawchuk, L., Lynch, S., & Rajani, S. (2008). Procrastination and motivation of
undergraduates with learning disabilities: A mixed methods inquiry. Learning Dis-
abilities Research & Practice, 23(3), 137-147.
Mastropieri, M., & Scruggs, T. (2010). The inclusive classroom: Strategies for effective differenti-
ated instruction (4th ed.). New Jersey: Merrill.
Meltzer, L., Katzir, T., Miller, L., Reddy, R., & Roditi, B. (2004). Academic self-perceptions,
effort, and strategy use in students with learning disabilities: Changes over time.
Learning Disabilities Research & Practice, 19, 99-108.
Metallidou, P., & Vlachou, A. (2010). Children’s self-regulated learning profile in language and
mathematics: The role of task value beliefs. Psychology in the Schools, 47, 776-788.
Ness, B., Sohlberg, M., & Albin, R. (2010). Evaluation of a second tier classroom-based assign-
ment completion strategy for middle school students in a resource context. Remedial
& Special Education, 32, 406-416.
Paris, S., & Paris, A. (2001). Classroom applications of research on self-regulated learning. Edu-
cational Psychologist, 36, 89-101.
Paris, S., & Winograd, P. (2001). The role of self-regulated learning in contextual teaching:
Principles and practices for teacher preparation. Center for the Improvement of Early
Reading Achievement, , 1/14/2009.
Parker, R., & Hagan-Burke, S. (2007a). Percentage of all non-overlapping data (PAND): An
alternative to PND. The Journal of Special Education, 40, 194-204.
Parker, R., & Hagan-Burke, S. (2007b). Useful effect size interpretations for single case research.
Behavior Therapy, 38, 95-105.
Primavera, L., Allison, D., & V, A. (1997). Measurement of dependent variables. In Design and
analysis of single-case research (pp. 41-89). Mahwah, NJ: Lawrence Erlbaum Associates.
Reid, R. (1996). Research in self-monitoring with students with learning disabilities: The pres-
ent, the prospects, the pitfalls. Journal of Learning Disabilities, 29(3), 317-331.
Schunk, D. (1998). Teaching elementary school students to self-regulate practice of mathemat-
ical skills with modeling. In D. Schunk, & B. Zimmerman (Eds.), Self-regulated learn-
ing: From teaching to self-reflective practice (pp. 137-159). New York: Guilford Press.
Schunk, D. (2008). Metacognition, self-regulation and self-regulated learning: Research rec-
ommendations. Educational Psychology Review, 20, 463-467.
Schunk, D., & Zimmerman, B. (2007). Influencing children’s self-efficacy and self-regulation
of reading and writing through modeling. Reading & Writing Quarterly, 23, 7-25.
Sikora, D. M., Haley, P., Edwards, J., & Butler, R. W. (2002). Tower of london test performance in
children with poor arithmetic skills. Developmental Neuropsychology, 21(3), 243-254.

51
Learning Disabilities: A Contemporary Journal 11(1), 35-52, 2013

Soderlund, J., Bursuck, B., Polloway, E. A., & Foley, R. A. (1995). A comparison of the home-
work problems of secondary school students with behavior disorders and nondis-
abled peers. Journal of Emotional and Behavioral Disorders, 3(3), 150-155.
Sugai, G., Horner, R., Dunlap, G., Hieneman, M., Lewis, T., Nelson, C. M., . . . Ruef, M. (2000).
Applying positive behavior support and functional behavioral assessment in schools.
Journal of Positive Behavior Interventions, 2(3), 131-143.
Trammel, D., Schloss, P., & Alper, S. (1994). Using self-recording, evaluation, and graphing
to increase completion of homework assignments. Journal of Learning Disabilities,
27, 75-81.
Uberti, H., Mastropieri, M. A., & Scruggs, T. E. (2004). Check if off: Individualizing a math
algorithm for students with disabilities via self-monitoring checklists. Intervention
in School and Clinic, 39, 269-275.
Valiente, C., Lemery-Chalfant, K., Swanson, J., & Reiser, M. (2008). Prediction of children’s
academic competence from their effortful control, relationships, and classroom par-
ticipation. Journal of Educational Psychology, 100, 67-77.
Wehmeyer, M., Yeager, D., Bolding, N., Agran, M., & Hughes, C. (2003). The effects of self-
regulation strategies on goal attainment for students with developmental disabilities
in general education classrooms. Journal of Developmental and Physical Disabilities,
15, 79-91.
Wolters, C. (1999). The relation between high school students’ motivational regulation and
their use of learning strategies, effort, and classroom performance. Learning and In-
dividual Differences, 11, 281-299.
Wolters, C. (2003). Understanding procrastination from a self-regulated learning perspective.
Journal of Educational Psychology, 95, 179-187.
Zimmerman, B. (1990). Self-regulated learning and academic achievement: an overview. Edu-
cational Psychologist, 25, 3-17.

52
Copyright of Learning Disabilities -- A Contemporary Journal is the property of Learning
Disabilities Worldwide and its content may not be copied or emailed to multiple sites or
posted to a listserv without the copyright holder's express written permission. However, users
may print, download, or email articles for individual use.