Content uploaded by Keith Herman
Author content
All content in this area was uploaded by Keith Herman on Sep 02, 2016
Content may be subject to copyright.
School Psychology Quarterly
Improving Teacher Perceptions of Parent Involvement
Patterns: Findings From a Group Randomized Trial
Keith C. Herman and Wendy M. Reinke
Online First Publication, July 21, 2016. http://dx.doi.org/10.1037/spq0000169
CITATION
Herman, K. C., & Reinke, W. M. (2016, July 21). Improving Teacher Perceptions of Parent
Involvement Patterns: Findings From a Group Randomized Trial. School Psychology Quarterly.
Advance online publication. http://dx.doi.org/10.1037/spq0000169
Improving Teacher Perceptions of Parent Involvement Patterns:
Findings From a Group Randomized Trial
Keith C. Herman and Wendy M. Reinke
University of Missouri
For children with the most serious and persistent academic and behavior problems,
parent involvement in education, particularly teacher perceptions of involvement, is
essential to avert their expected long-term negative outcomes. Despite the widespread
interest in and perceived importance of parent involvement in education, however, few
experimental studies have evaluated programs and practices to promote it. In this group
randomized trial, we examined the effects of the Incredible Years Teacher Classroom
Management program (IY TCM) on teacher perceptions of contact and comfort with
parents. One hundred five classrooms with 1818 students were randomly assigned to an
IY TCM or to a control, business as usual condition. Measures of key constructs
included teacher ratings of parent and student behaviors, direct observations in the
classroom, and a standardized academic achievement test. Latent transition analysis
(LTA) was used to identify patterns of involvement over time and to determine if
intervention condition predicted postintervention patterns and transitions. Four patterns
of involvement were identified at baseline and at follow-up; parents of students with
academic and behavior problems were most likely to be in classes with the least
adaptive involvement patterns. Intervention status predicted group membership at
follow-up. Specifically, intervention classroom parents were significantly more likely
to transition to more adaptive teacher-rated parenting profiles at follow-up compared to
control classroom parents. This is the first randomized trial we are aware of that has
found that teacher training can alter teacher perceptions of parent involvement patterns.
Clinical implications for students with behavior and academic problems are discussed.
Keywords: parent involvement, teacher–parent relationships, student outcomes, LTA
Supplemental materials: http://dx.doi.org/10.1037/spq0000169.supp
For students with the most serious and per-
sistent academic and behavior problems, parent
involvement is essential to avert their expected
long-term negative outcomes (Wagner, Kutash,
Duchnowski, Epstein, & Sumi, 2005). Although
the pathway to serious antisocial behavior and
academic failure is often set in motion by coer-
cive parent– child interactions, the transition to
elementary school represents a crucial develop-
mental opportunity for interrupting the negative
cascade of social and academic outcomes that
typically occurs for these children (Patterson,
Reid, & Dishion, 1992). Children at high risk
for antisocial behavior often develop poor rela-
tions with teachers and receive less support and
instruction and more criticism in the classroom
(Arnold, Griffith, Ortiz, & Stowe, 1998; Camp-
bell & Ewing, 1990; Carr, Taylor, & Robinson,
1991). In turn, these children are often rejected
by their teachers and well-adjusted peers over
the course of elementary school and fail to
develop the “survival skills” necessary for aca-
demic and social success (Kellam, Rebok, Ia-
longo, & Mayer, 1994). Eventually, according
to Patterson et al. (1992), these children are
likely to drift into a deviant peer group where
Keith C. Herman and Wendy M. Reinke, Department of
Educational, School, & Counseling Psychology, University
of Missouri.
The research reported here was supported by the Institute
of Education Sciences, U.S. Department of Education,
through Grant R305A100342 to the University of Missouri
(PI: Wendy Reinke). The opinions expressed are those of
the authors and do not represent views of the Institute or the
U.S. Department of Education.
Correspondence concerning this article should be ad-
dressed to Keith C. Herman, Department of Educational,
School, & Counseling Psychology, University of Missouri,
Columbia, MO 65211. E-mail: hermanke@missouri.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
School Psychology Quarterly © 2016 American Psychological Association
2016, Vol. 31, No. 4, 000 1045-3830/16/$12.00 http://dx.doi.org/10.1037/spq0000169
1
antisocial behavior and academic failure are
reinforced.
Teachers of students with severe behavior
and academic problems often develop negative
perceptions not only of the students but also of
their families (Stormont, Herman, Reinke, Da-
vid, & Goel, 2013). Teacher perceptions are
especially important given that teacher beliefs
about parents, regardless of their accuracy, may
influence their interactions with parents and
their children (Henderson & Berla, 1997; Mc-
Coach et al., 2010; McDermott & Rothenberg,
2000). For instance, it is likely that teachers
interact differently with parents whom they per-
ceive as committed and interested in their
child’s education compared with parents whom
they perceive as less involved in learning (Hen-
derson & Berla, 1997; McDermott & Rothen-
berg, 2000). These perceptions in turn influence
parents’ ability, willingness, and motivation to
participate in education (see Herman et al.,
2012). Not surprisingly, families from disad-
vantaged backgrounds and those with children
with the highest service needs have the lowest
level of school participation (El Nokali, Bach-
man, & Votruba-Drzal, 2010; Park, Pullis,
Reilly, & Townsend, 1994).
Teacher Perceptions of Parent Involvement
Teacher perceptions of parent involvement
patterns have been found to be especially im-
portant predictors of student outcomes (Bakker,
Denessen, & Brus-Laeven, 2007; Barnard,
2004). For instance, using a sample of 1,165
students from the Chicago Longitudinal study,
Barnard (2004) found that only teacher ratings
of parent involvement in elementary school pre-
dicted student outcomes in high school. Control-
ling for background characteristics, elementary
students of higher versus lower participating par-
ents were less likely to drop out and more likely
to graduate from high school. Specifically, stu-
dents whose parents were rated by teachers as
having average or better participation for three
or more years were 96% more likely to graduate
from high school than students of parents never
rated as having average or better participation.
Parent ratings of their own involvement were
largely unrelated to student outcomes in high
school. Bakker et al. (2007) reported similar
findings about the predictive power of teacher
versus parent perceptions of involvement in an
elementary sample from the Netherlands. Other
authors have found teacher perceptions of par-
ent involvement in early grades predict student
achievement in later elementary school (Izzo et
al., 1999) and reading achievement, grade re-
tention, and special education status in eighth
grade (Miedel & Reynolds, 2000), even after
controlling for a host of known correlates of
academic success.
Teacher ratings of parent involvement com-
monly focus on the amount of contact parents
have with school personnel (e.g., attending
meetings, volunteering, and visiting the school).
A second equally important aspect of teacher
perceptions is their comfort with parents (Mill-
er-Johnson & Maumary-Gremaud, 1995). Com-
fort, sometimes called “bonding” or “alliance,”
refers to the quality of the relationship, not just
the frequency of contact. In particular, teacher
comfort with parents includes how aligned the
teacher perceives the parent’s goals and values
to be with his or her own (Miller-Johnson &
Maumary-Gremaud, 1995).
Although most studies have examined levels
of contact or comfort separately as individual
variables, their co-occurrence may provide even
more important information about relationship
patterns (McCoach et al., 2010). For instance,
some parents may be characterized as having
low contact with schools but still be perceived
as having a comfortable relationship. In fact, a
recent study (Stormont et al., 2013) found this
pattern of teacher-rated parent involvement to
be associated with the most favorable student
outcomes. On the other hand, some parents may
have frequent contact with teachers but be per-
ceived as intrusive and unhelpful. Thus, it is the
combination of contact and comfort, not either
in isolation, that may determine how adaptive
they are. Two recent studies have found that
teacher perceptions of the combination of these
two dimensions of contact and comfort are
strong correlates of student success (McCoach
et al., 2010; Stormont et al., 2013). For instance,
Stormont et al. (2013) found three profiles of
teacher-rated contact and comfort: high contact
and comfort, low contact/high comfort, and low
contact and comfort. Students in the latter group
had significantly higher levels of behavior prob-
lems at school and lower levels of academic
achievement.
2 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Pathways Between Parent Involvement and
Student Outcomes
What are the mechanisms by which teacher
perceptions of parent contact and comfort influ-
ence student outcomes? First, teacher bias may
undermine parent willingness and ability to sup-
port their child’s educational achievement and
social development. Much research has docu-
mented that parents are unlikely to seek services
through schools or mental health providers
when they feel judged (McKay, Atkins, Hawk-
ins, Brown, & Lynn, 2003; Nock & Kazdin,
2001). Biases and negative perceptions of
school staff may also make schools unwelcom-
ing to parents and will likely influence both
contact and comfort levels (Stormshak et al.,
2005). The parents of students with behavior
and/or academic problems and from culturally
diverse backgrounds tend to have the lowest
levels of school involvement (El Nokali et al.,
2010; Park, Pullis, Reilly, & Townsend, 1994)
which may contribute to a cycle of under-
achievement for these students.
A second, complementary mechanism of par-
ent involvement may be through its impact on
teacher and student interactions. In a longitudi-
nal study with 443 1st-grade students, Hughes
and Kwok (2007) found that teacher percep-
tions of parent involvement (a latent factor com-
posed of both contact and alliance) mediated the
effects of student background characteristics
(race, gender) on student engagement; engage-
ment in turn mediated the effects of teacher-
perceived involvement on student academic
achievement. Thus, teacher perceptions of both
dimensions of involvement may directly influ-
ence student engagement in learning which con-
tributes to student performance. A study with
894 3rd-grade students also found that parent
involvement predicted quality of student–
teacher interactions (Wyrick & Rudasill, 2009),
suggesting a plausible additional link in the
pathway to student outcomes. That is, positive
teacher perceptions of parent involvement may
lead to improved teacher-student interactions
and higher levels of student engagement, both
of which contribute to better student behaviors
and academic performance.
These findings suggest that altering teacher
perceptions of parent involvement patterns, es-
pecially perceived contact and comfort, may be
a key leverage point in fostering more positive
outcomes for at-risk students. Interventions de-
signed to change teacher attitudes and behaviors
toward parents hold promise for altering parent
participation patterns, which in turn could im-
prove student outcomes. Given that teachers
interact with dozens of parents and students
each year, teacher focused training that ad-
dresses parent involvement also has the advan-
tage of broader, class-wide impact than inter-
ventions delivered individually to parents. One
teacher focused intervention is the Incredible
Years Teacher Classroom Management (IY
TCM) program (Webster-Stratton & Reid,
2010).
IY TCM Program and Parent Involvement
IY TCM was developed as a companion to
the IY Parent intervention, the well-established
treatment for youth with conduct problems, to
address the common finding that parenting in-
terventions alone do not always improve stu-
dent behaviors at school (Taylor & Biglan,
1998; Webster-Stratton, 1990). IY TCM was
designed as a 6-day teacher training program
focused on improving teacher–student and
teacher–parent relationships and increasing
teacher use of effective classroom management
strategies. In particular, IY TCM has a primary
goal of encouraging teachers to examine their
assumptions and biases about parents and to
develop and/or repair relationships with chal-
lenging students and families. Although IY
TCM has been shown to improve teacher class-
room management behaviors and child out-
comes in a stand-alone intervention (Reinke,
Herman, & Dong, 2014; Reinke, Stormont,
Herman, & Newcomer, 2014) and as part of
multicomponent interventions (Webster-Strat-
ton, Reid, & Hammond, 2001, 2004; Webster-
Stratton, Reid, & Stoolmiller, 2008), no re-
search has reported its impact on parent
involvement.
This study examined the effects of the IY
TCM on teacher perceptions of contact and
comfort with parents. We believe this is the first
randomized trial to evaluate the effects of any
universal teacher training program on parent
involvement profiles. Using a person-centered
approach, we conducted latent profile analyses
(LPA) on teacher ratings of parent contact and
comfort at baseline and follow-up to identify
profiles of involvement. Given our definition of
3LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
teacher perceived parent involvement as a pat-
tern of co-occurring constructs, contact and
comfort, LPA was the optimal method for de-
fining these patterns. Similar to findings from
Stormont et al. (2013), which relied on the
baseline scores of the first cohort of participants
in the present trial, we hypothesized that three
profiles would emerge at both time points char-
acterized by the following: (a) high contact/high
comfort, (b) low contact/high comfort, and (c)
low contact/low comfort. To validate the pro-
files, we next examined their relations with doc-
umented correlates. Specifically, we hypothe-
sized that students of parents in the low contact
and low comfort profile would have signifi-
cantly higher observed and teacher-rated behav-
ior problems and lower academic performance
compared to other profiles.
Next, we conducted latent transition analyses
(LTA) to test the hypothesis that teachers who
received the training would be more likely to
shift into adaptive profiles at follow-up. LTA is
an advanced analytic method for examining
changes in profile membership (e.g., low con-
tact/high comfort, low contact/low comfort) de-
rived from LPAs at two time points (see
Thompson, Macy, & Fraser, 2011). LTA af-
forded the opportunity to examine intervention
effects on these patterns rather than on mean
level changes of singular variables. We ex-
pected teachers who received IY TCM would be
more likely to have parents shift into more
adaptive profile patterns (characterized by ei-
ther higher levels of contact and comfort or by
favorable student outcomes) at follow-up com-
pared to teachers in the business-as-usual con-
dition.
Method
Participants
Participants in this group randomized trial
were 105 teachers and 1818 students in Grades
kindergarten to third from nine urban schools
serving primarily African American students.
As indicated in the participant flowchart (see
Figure 1), the study had high rates of enrollment
for eligible teachers (96%) and students (84%).
All teacher participants and parents provided
written consent, and all students provided writ-
ten assent to participate in the study. Teachers
were randomly assigned within schools to re-
ceive IY TCM or to a wait-list, business-as-
usual control group. The majority of teacher
participants were female (97%) and White
(75%; 22% African American and 3% other).
The student sample included more males (52%)
and African Americans (76%; 22% White, and
2% other), and half the student sample qualified
for free or reduced lunch.
Measures
Student demographics. Free and reduced
lunch status (FRL), student race, and sex were
obtained from the school district for all partic-
ipating students. Students were coded as 1 if
they received FRL and 0 if not. Students were
coded as 1 if they were African American, and
0 if not, because of small cell size for students
in the other race category.
Parent contact and comfort. The Parent
Involvement Measure – Teacher (Conduct
Problems Prevention Research Group [CP-
PRG], 1991) is a 21-item measure that assesses
facets of parent involvement. This measure was
completed by the classroom teacher for each
student. Two subscales, Parent—Teacher Con-
tact and Parent Comfort, were included in the
latent profile analysis to explore specific sub-
types of parent involvement. The item re-
sponses ranged from 0 (not at all or never)to4
(very interested, very often, very comfortable, a
whole lot, more than once a week). Factor anal-
ysis on this measure documented strong support
for the comfort and contact factors (Miller-
Johnson & Maumary-Gremaud, 1995). Malone
(2000) found evidence that the measure can
discriminate between high and low risk sam-
ples. For the current control group sample, the
test–retest reliability from fall to spring for each
subscale were .57 (Parent–Teacher Contact) and
.68 (Parent Comfort), respectively.
Academic achievement. The Woodcock–
Johnson III Normative Update Tests of Achieve-
ment (WJ III ACH; Woodcock, McGrew, &
Mather, 2007) is an assessment of student aca-
demic achievement. The present study included
two subscales, Broad Reading and Broad Math.
The WJ III ACH is a well-established measure
with strong psychometrics. Teachers also rated
student academic competence using the Social
Competence Scale – Teacher version (T-
COMP; CPPRG, 1995). The item responses
range from 0 (almost never)to4(almost al-
4 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ways). Previous research has found adequate
psychometric properties for the T-COMP (CP-
PRG, 1995; Gouley, Brotman, Huang, & Shrout
2008).
Direct observations of students. Independent
observers conducted direct observations of
teacher and student behaviors using the Multi-
Option Observation System for Experimental
Studies (MOOSES; Tapp, 2004) interface for
handheld computers to gather real-time data
using the Brief Student–Teacher Classroom In-
teraction Observation code (ST-CIO; Reinke &
Newcomer, 2010). The ST-CIO simultaneously
collects counts/duration of teacher behaviors
(praise, reprimands/teaching) and student be-
haviors (disruptions/off task). Each student was
observed for 5 min during academic instruction
times (reading or math). Prior to data collection,
full-time research staff and graduate student
observers were trained for 2 weeks using videos
and practice sessions to reach 85% reliability
with a master coder. Reliability checks were
conducted on a random subset of 29% of all
observations, and observers received continuing
supervision to ensure against observer drift. The
mean percent agreement on the ST-CIO was
93% for end-of-the-year observations.
MOOSES utilizes a rigorous second-by-second
comparison of raters to determine reliability,
and an overall reliability of 80% is considered
acceptable; thus 93% is considered highly reli-
able (Tapp, 2004).
Teacher ratings of student disruptive be-
haviors and family problems. Teachers also
rated student disruptive behaviors and family
problems on the Teacher Observation of Class-
Assessed for eligibility
Teachers (n=109)
Students (n=2168)
Excluded (declined to participate)
Teachers (n=4)
Students (n=350)
Analyz ed
Teachers (n=53)
Students (n=896)
Excluded Students (n=5; missing x and y)
Lost to follow-up
Teachers (n=0)
Students (n=63; moved)
Allocated to IY TCM
Teachers (n=53)
Students (n=901)
Lost to follow-up
Teachers (n=0)
Students (n=57; moved)
Allocated to Wait-List Control
Teachers (n=52)
Students (n=917)
Analyz ed
Teachers (n=52)
Students (n=910)
Excluded Students (n=7; missing x and y)
Allocation
Analysis
Follow-Up
Classrooms Randomized
Teachers (n=105; 96%)
Students (n=1818; 84%)
Enrollment
Figure 1. IY TCM randomization participant flowchart. See the online article for the color
version of this figure.
5LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
room Adaptation-Checklist (TOCA-C; Koth,
Bradshaw, & Leaf, 2009) in October and April
of the academic year. The item responses
ranged from 1 (never)to6(almost always). The
TOCA-C is a reliable and valid measure. A
recent factor analytic study confirmed the factor
structure of the TOCA-C in our sample and
found it to have strong psychometric properties
in terms of goodness of fit and measurement
invariance (Wang et al., 2015).
Procedures
Teachers and students were recruited at the
beginning of the school year. All baseline as-
sessments occurred in late September and early
October, prior to the intervention. Postassess-
ments occurred at the end of the same school
year, in late April and early May.
In three sequential, annual cohorts of 15–20,
teachers in the IY TCM condition attended three
sets of two full-day group trainings in late Oc-
tober, December, and February. All trainings
were cofacilitated by two doctoral level IY
TCM group leaders who were supervised by the
program developer; one of these trainers also
served as a coach who supported teacher imple-
mentation following sessions.
IY TCM is a comprehensive curriculum for
improving teacher classroom management and
relationship skills. IY views solid relationships
with all students and parents as a necessary
element of successful classroom management.
In particular, the IY TCM logic model identifies
“positive parent–teacher partnerships” as one of
three program goals, specifies parent involve-
ment activities as a key modality or strategy,
and targets positive parent–teacher relations as a
proximal outcome of the program. Parental in-
volvement is discussed in every training work-
shop and includes strategies to assist teachers to
develop better relationships with parents and
create opportunities for parents to get involved
in a child’s education. The entire first full day of
training (seven hours) is focused on building
relationships with parents and students. Much
time during all training sessions is devoted to
observing video vignettes of actual teacher in-
teractions with students and parents. Each vi-
gnette provides a model for effective interaction
and also evokes discussion and insight about the
teacher’s beliefs, biases, and perceptions of
these interactions. The trainers prompt reflec-
tions with Socratic questioning about the videos
(“How are you feeling as the teacher in this
situation?” “How is the student/parent feeling?”
“What is the student/parent learning?” “How
would you respond in this situation?”) and fa-
cilitate group discussion. These conversations
spark role plays to practice challenging interac-
tions. In this way teachers can serve as models
for others and/or get feedback about improving
their interactions. Each section of the training
provides ample time for self-reflection; teachers
respond to a series of questions about their
current practices (e.g., what do I do to make
parents feel valued and welcome, what am I
doing that may lead parents to feel unvalued or
unwelcome) and what they can do differently.
Time is also allotted for teachers to develop
plans for improving their classroom ecology
and for developing behavior support plans for
responding to challenging students and parents.
These plans always include specific prompts for
how the teacher will engage parents (see Reinke
et al., 2014 for more details about these plans
and outcomes in this sample). Specific tools and
strategies are provided in the materials that each
teacher receives including letters that can be
sent home to facilitate home school communi-
cation. A coach meets with teachers on a regular
basis after the first workshop through the end of
the school year; the amount of time and focus of
these meetings is tailored to the teachers’ inter-
ests and needs. (For more information about cur-
riculum content see http://incredibleyears.com/
programs/teacher/classroom-mgt-curriculum).
Fidelity of implementation of the program
and teacher implementation skills were mon-
itored over the course of the year and are
documented in a separate report (see Reinke,
Herman, Stormont, Newcomer, & David,
2013). In summary, teachers in the interven-
tion were all exposed to the training work-
shops; nearly all teachers attended all six
workshops (attendance rate was 94%–100%
for each workshop), and the few teachers who
missed a workshop because of illness met
with a coach to review missed material.
Teachers rated the workshops with high sat-
isfaction and likelihood of recommending the
training to others (mean ratings of 6.44 – 6.75
on a scale from 1–7 with high scores indicat-
ing greater satisfaction). Teachers also re-
ceived a strong dose of coaching. The average
amount of time each teacher spent with a
6 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
coach between and after workshop sessions
was nearly six hours (358 min). Direct obser-
vations revealed that teachers in the interven-
tion condition used higher rates of proactive
teaching strategies compared to control teach-
ers at each follow-up point. Follow-up assess-
ments were collected in late April and May of
the same academic year.
Control
Teachers assigned to the wait-list control
condition continued their business as usual
teaching and pursuing professional develop-
ment opportunities during the study period.
They were offered IY TCM at the end of the
study after follow-up data had been collected.
Analysis Plan
Latent profile and transition analyses (LPA
and LTA, respectively) were used to examine
patterns of two indicators of teacher perceived
parent involvement, contact and comfort (Ny-
lund et al., 2005). First, we conducted uncondi-
tional LPAs at each time point to determine the
optimal number of profiles. Second, we exam-
ined independent correlates of identified classes
to confirm their distinctiveness and meaningful-
ness using the Auxiliary function in MPlus.
Third, we conducted an LTA to report transi-
tions between profiles over time. Finally, we
repeated the LTA with treatment condition as a
covariate to determine if treatment predicted
class membership at follow-up. Conditional
models also accounted for clustering of students
within classrooms and included student sex,
race, and lunch status as covariates. All analyses
were conducted using MPlus 6.1 (Muthén &
Muthén, 1998 –2010). In LPA, a combination of
statistical considerations and substantive theory
are used to decide on the best fitting model.
LPA decision rules are given in the supplemen-
tal materials. In LTA, we used the KNOWN-
CLASS command to identify treatment condi-
tion as a categorical latent variable for which
class membership was known (Muthén &
Muthén, 1998 –2010).
To accommodate for missing data, Mplus
software uses full information maximum likeli-
hood with the assumption that the data are miss-
ing at random (Little, 1995), a common ap-
proach employed within this analysis method
(Schafer & Graham, 2002). The minimum co-
variance coverage recommended for reliable
model convergence is 0.10 (Muthén & Muthén,
1998-2010). In this study, coverage ranged
from 0.924 to 0.987.
Results
Latent Profile Analysis of Baseline and
Follow-Up Parent Contact and Comfort
Teacher ratings of parent contact and com-
fort were entered as indicators in LPAs at
baseline and follow-up. LPA fit indices for
class solutions are summarized in Table 1.
We used these fit indices as well as theory and
prior literature to select the four class solution
as the best fitting model at both time points.
The four class solutions had significantly
lower values than the three class solutions,
and a parametric bootstrapping likelihood ra-
tio test in mixture analyses confirmed it as a
better fit than the three class model. The BIC
values for the four and five class solutions
Table 1
Model Fit Indices For 1–5 Class Solutions of Baseline and Follow-Up
Baseline Follow-Up
LC BIC aBIC VLMR Entropy BIC aBIC VLMR Entropy
1 7898.84 7886.14 — — 6479.47 6466.76 — —
2 7624.67 7602.44 .04 .74 6289.45 6267.21 .02 .67
3 7579.16 7547.39 .27 .62 6163.54 6131.77 .46 .73
a
47562.48 7521.17 .10 .63 6109.68 6068.39 .04 .72
5 7561.62 7510.79 .42 .67 6101.86 6051.03 .42 .72
Note. LC ⫽Latent Classes; BIC ⫽Bayesian Information Criterion; aBIC ⫽adjusted Bayesian Information Criterion;
VLMR ⫽Vuong-Lo-Mendell-Rubin.
a
Bold indicates equal selected class solution.
7LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
were comparably low, and the fifth class did
not provide additional information relevant to
the outcome questions of interest to the study.
That is, the fifth class that emerged in both
five class models represented a very small
percentage of parents (3%) and was not con-
ceptually relevant (i.e., the average ratings on
both indicators were very similar to those
observed for the fourth class); moreover, sub-
sequent analyses revealed the fifth class was
not distinguished by behavior or academic
correlates. Thus, we rejected the fifth class in
favor of the more parsimonious four class
solution. Figure 2 summarizes the prevalence
and characteristics of the classes at each time
point. The four class solution included three
classes that we hypothesized: (1) High Con-
tact/High Comfort, (2) Low Contact/Medium
Comfort, and (3) Low Contact/Low Comfort.
The fourth class was characterized by Me-
dium Contact/Medium Comfort at baseline
and Low Contact/High Comfort at follow-up.
For transition purposes, we considered these
fourth classes as comparable because they
represented a similar proportion of partici-
pants at each time point, and they were both
characterized by contact and comfort mean
ratings that fell between the High/High and
Low/Medium profiles at both time points.
Roughly half the sample fell into Low Con-
tact/Medium Comfort at both time points
(54.8% and 47.3%, respectively). The Low
Figure 2. Baseline and follow-up profiles of contact and comfort.
8 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Contact/Low Comfort (16.2% and 14.3%, re-
spectively) and High Contact/High Comfort
(3.6% and 2.1%, respectively) profiles were
less common.
Demographic Correlates of
Class Membership
Demographic variables were uniquely related
to several baseline and follow-up classes. At
baseline, the Low Contact/Low Comfort class
was significantly more likely to have FRL stu-
dents than the High Contact/High Comfort class
(OR: 2.79; 1.59 –4.42); girls were more likely to
appear in the Low Contact/Medium Comfort
(OR: 1.73; 1.16 –2.41) and Medium Contact/
Medium Comfort (OR: 2.14; 1.09 –4.23). At
baseline, the Low Contact/Low Comfort class
was significantly more likely to have FRL (OR:
3.35; 1.66 –6.75) and African American stu-
dents (OR: 13.51; 2.20 –83.33) than the Low
Contact/High Comfort class. At follow-up, Af-
rican American students were more likely to
appear in the Low Contact/Low Comfort class
compared to the High Contact/High Comfort
class (OR: 5.56; 1.15–27.03), and girls were
more likely to be in the Low Contact/Medium
Comfort class compared to the Low Contact/
High Comfort class (OR: 1.99; 1.32–3.01).
Academic Skills and Behaviors Associated
With Follow-Up Classes
We examined the association between classes
at follow-up and indicators of academic perfor-
mance, student behaviors, and family problems
(see Table 2). Equality tests of means across
classes using posterior probability-based multi-
ple imputations revealed a number of significant
differences among the parent involvement pro-
files. The Low Contact/Low Comfort class had
significantly lower academic performance com-
pared to the other three groups across all com-
parisons. Additionally, the Low Contact/High
Comfort class had the highest academic perfor-
mance on all academic measures, even in com-
parison to the High Contact/High Comfort
group on teacher ratings and WJ Reading
scores. Notably, the mean standard score on the
math and reading subtests of the WJ was 10
points lower for students in the Low Contact/
Low Comfort compared to the Low Contact/
High Comfort class. Additionally, independent
Table 2
Estimated Mean Scores On Family Problems and Student Achievement and Behaviors by Class at Follow-Up
Variables
1 Low Contact Low
Comfort M (SE)
2 Low Contact Medium
Comfort M (SE)
3 Low Contact High
Comfort M (SE)
4 High Contact High
Comfort M (SE) Significant class comparisons
Teacher report
Family Problems 3.80 (.09) 2.34 (.04) 1.74 (.04) 2.21 (.21) 1 vs. All
ⴱⴱⴱ
, 2 vs. 3
ⴱⴱⴱ
, 3 vs. 4
ⴱ
Disruptive 2.38 (.07) 1.87 (.03) 1.69 (.03) 2.41 (.19) 3 vs. All
ⴱⴱⴱ
, 2 vs. 1,4
ⴱⴱⴱ
Prosocial 4.03 (.08) 4.69 (.04) 5.12 (.04) 4.34 (.21) 3 vs. All
ⴱⴱⴱ
Academics (COMPS) 2.22 (.08) 3.39 (.05) 4.05 (.05) 3.44 (.22) 1 vs. All
ⴱⴱⴱ
, 2,3 vs. 4
ⴱⴱⴱ
Achievement
WJ Reading 92.42 (1.00) 99.51 (.52) 103.47 (.50) 100.80 (2.41) 1 vs. All
ⴱⴱⴱ
, 2 vs. 3
ⴱⴱⴱ
WJ Math 89.78 (1.12) 96.66 (.63) 100.17 (.65) 94.25 (2.74) 1,2 vs. 3
ⴱⴱⴱ
, 3 vs. 4
ⴱ
Direct observations
Off Task .095 (.01) .070 (.01) .059 (.01) .071 (.02) 1 vs. 3
ⴱⴱ
Disruptions .076 (.01) .064 (.01) .045 (.01) .101 (.03) 1,2 vs. 3
ⴱ
ⴱ
p⬍.05.
ⴱⴱ
pⱕ.01.
ⴱⴱⴱ
pⱕ.001.
9LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
direct observations of student behavior in the
classroom found that students of parents in the
Low Contact/Low Comfort class had signifi-
cantly higher rates of disruptive and off-task
behaviors compared to students of parents in the
Low Contact/High Comfort class. Teacher rat-
ings of student disruptive behaviors confirmed
these findings. Additionally, teachers rated fam-
ilies in the Low Contact/Low Comfort class as
having significantly more family problems than
the other classes. Across most comparisons, the
Low Contact/High Comfort profile emerged as
having the most positive child and family cor-
relates. The High Contact/High Comfort profile
was associated with some student academic and
behavior problems. As did Stormont et al.
(2013), we interpreted this finding as indicating
an adaptive pattern of parent involvement in
education, perhaps in response to student chal-
lenges with school, given that teachers were
perceiving these parents favorably (i.e., as
highly involved and comfortable).
LTA Intervention Effects
We first conducted an LTA without cova-
riates. (We provide joint and conditional
probabilities from these analyses in the sup-
plemental materials.) We then repeated the
LTA with intervention, sex, race, and lunch
status as predictors of baseline and follow-up
class membership. Intervention condition was
unrelated to baseline class membership, sug-
gesting that randomization resulted in base-
line equivalence between intervention condi-
tions in profile patterns. Thus, analyses
focused on intervention effects of follow-up
class membership.
As hypothesized, intervention predicted
follow-up class membership. Parents of inter-
vention teachers were more likely to be in the
High Contact/High Comfort class at fol-
low-up versus the Low Contact/Low Contact
(OR: 4.55; 1.09 –19.03) and Low Contact/
Medium Comfort classes (OR: 5.30;
1.21–22.36). Intervention teachers were also
more likely to rate parents as being in the
Low Contact/High Comfort class compared to
the Low Contact/Low Comfort (OR: 2.95;
1.40 –6.49) an d Low Contact/Medium Comfort
classes (OR: 3.43; 1.40 –8.43). One quarter of
the sample (459 parents) were in the most adap-
tive classes (High Contact/High Comfort and
Low Contact/High Comfort) at follow-up, and
77% of these parents were from the intervention
group.
Examining transitions among classes also
provided support for the treatment condition.
Only 7% of parents for control teachers in the
lower two adaptive classes at baseline transi-
tioned to a more adaptive class at follow-up,
compared to 39% of parents for intervention
teachers who transitioned to a more adaptive
class. Particularly striking, only 2% of control
parents in the Low/Low class at baseline tran-
sitioned to the most adaptive classes at fol-
low-up, compared to 25% for the intervention
group. Moreover, 80% of parents who transi-
tioned to the High Contact/High Comfort
class at follow-up were from the intervention
group, versus 20% from the control group.
Finally, we examined likelihood of pro-
gressing, digressing, or staying in a given
class between conditions (see Figure 3). We
defined Stayers as those with stable member-
ship in a class at both time points. Progressors
were those who moved from a less adaptive
class to a more adaptive class from baseline to
follow-up. Digressors were defined by move-
ment from a more adaptive class to a less
adaptive class. Given that Low Contact/High
Comfort and High Contact/High Comfort
both were characterized as types of adaptive
involvement (the former because students in
this class had the most favorable outcomes
and the latter because it represented the high-
est scores on both types of involvement), we
considered movements between these classes
as types of Stayers (i.e., it did not make
conceptual sense to characterize movement
between these classes either as a type of pro-
gression or digression). Chi-square analyses
indicated that progression was more likely in
the intervention group and digression was
more likely in the control group (
2
[2] ⫽
65.21; p⬍.0001;
c
⫽.19). Specifically, the
observed occurrence of progression was 38%
less than expected for the control group and
38.6% more than expected for the interven-
tion group. Counterfactual evidence was also
found in the observed frequency of digression
in the control group, which was 22.4% more
than expected and 22.8% less than expected
in the intervention condition (see Table 3).
10 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Discussion
This may be the first randomized trial to find
significant improvement in teacher perceptions
of parent educational involvement following a
universal teacher training intervention. Al-
though the primary outcome focused on teacher
perceptions of involvement only, existing evi-
dence suggests that such perceptions are a
meaningful educational outcome. First, prior
studies suggest that teacher perceptions of par-
ent involvement may provide the strongest pre-
dictors of future student success and thus are a
highly relevant target of intervention. The com-
pelling literature showing longitudinal, inde-
pendent links between teacher perceptions of
parent involvement and educational outcomes
through high school (see Bakker et al., 2007;
Barnard, 2004; Izzo et al., 1999; Miedel &
Reynolds, 2000) suggest that IY TCM’s signif-
icant impact on parent involvement profiles
may lead to long-term educational benefit for
students. Second, profiles of teacher-rated par-
ent involvement were associated with objective
educational outcomes in this study including
performance on standardized achievement tests,
teacher ratings of academic skills and behavior
problems, and independent observations of dis-
ruptive and off task classroom behaviors.
The primary focus of the study was on
using sophisticated modeling techniques to
Low/High
Low/Med
Low/High
High/High
Tx: 68
Ctrl
:
77
Tx: 305
Ctrl
:
440
Ctrl
:
22
Tx: 59
Tx: 12
Ctrl
:
3
Tx: 2
Ctrl: 1
Tx: 27
Ctrl: 2
Tx: 4
Ctrl: 1
Tx: 20
Ctrl: 48
Tx: 156
Ctrl: 43
Tx: 2
Ctrl: 0
Tx: 92
Ctrl: 33
Tx: 62
Ctrl: 108
Tx: 58
Ctrl: 93
Tx: 22
Ctrl: 35
Tx: 1
Ctrl: 0
Tx: 6
Ctrl: 4
Figure 3. Transition patterns for intervention (n⫽896) and control (n⫽908) participants.
Note: Circles indicate Stayers, those with stable membership within class over time. Forward
arrows (¡) indicate Progressors, those who moved from a less adaptive class to more
adaptive class. Back arrows (¢) indicate Digressors, those who moved from a more adaptive
class to a less adaptive class. Double arrows (¢¡) indicate transitions between the most
adaptive classes (shaded gray).
Table 3
Chi Square Analysis of Progressors, Stayers,
and Digressors
Transitions Control Treatment Totals
Progressors 95 209 304
Stayers 575 538 1113
Digressors 240 149 389
Totals 910 896 1806
Note.
2
(2) ⫽65.21; p⬍.002;
c
⫽.19.
11LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
determine if teacher ratings of parent involve-
ment could be altered by an intervention. The
person-centered analyses indicated that pro-
files of perceived contact and comfort provide
a valuable method for understanding parent
involvement and supporting student out-
comes. Low contact alone was not associated
with adverse student performance or out-
comes. The worst correlates were observed
for students with low contact and low comfort
parents, whereas, the low contact and high
comfort profile was associated with higher
academic achievement and lower behavior
problems. These findings imply that a dimen-
sional, profile approach adds a more nuanced
portrayal of involvement patterns than can be
captured by single indicators.
Notably, the low contact and low comfort
profile had a greater probability of being com-
posed of parents of African American stu-
dents, students from low income back-
grounds, and students with academic and
behavior problems. Other researchers have
documented that involvement, however it is
measured, tends to be lowest for families
from cultural minority groups and students at
risk for school failure (El Nokali et al., 2010;
Park et al., 1994; Stormont et al., 2013). That
students in intervention classrooms had a
greater probability of appearing in the most
adaptive profiles versus the least adaptive
profile (i.e., low contact and low comfort) at
follow-up suggests IY TCM may hold benefit
for improving educational outcomes for fam-
ilies and students with the greatest needs.
In line with Stormont et al. (2013), low con-
tact alone was not a risk for teacher-rated dis-
comfort or for negative student outcomes. This
implies that teachers base their discomfort on
something other than actual contact with par-
ents. Given the background characteristics of
students and parents in the low contact and low
comfort group, prime candidates to explain
teacher perceived discomfort with parents in-
clude racial or social economic differences and
teacher interpretation of students’ misbehavior
and/or skill deficits as a reflection of the parent
(see Stormont et al., 2013). Given the amount of
focus devoted to examining teacher bias and
building empathy for children and parents in IY
TCM, a tenable hypothesis is that the training
improves teacher perceptions of parent involve-
ment by altering what otherwise would remain
ingrained assumptions about parents who differ
from them and/or who have students with
school problems. Notably, a recent study found
that the effects of an intensive teacher–parent
consultation model on youth outcomes were
mediated by teacher perceptions of parent in-
volvement, but not parent perceptions of in-
volvement (Sheridan et al., 2012). This finding
is consistent with the idea that changing teacher
attitudes about parents contributes to student
outcomes, possibly by fostering improved rela-
tions with students and in turn increases in
student engagement (Hughes & Kwok, 2007).
Limitations and Future Directions
Unfortunately, the present study design did
not allow us to test mechanisms between inter-
vention, teacher perceptions of parents, and stu-
dent outcomes. IY TCM is a multifaceted inter-
vention so we cannot be sure the observed
changes in teacher perceptions occurred be-
cause of specific program components or be-
cause of the intervention package. Because IY
TCM provides explicit instruction and practice
to support improvements in teacher–parent in-
teractions, it is possible that teachers became
more skilled at involving parents in school,
which led to successful interactions and im-
proved perceptions of one another. It is also
possible that teacher-observed improvements in
student behaviors may have been the result of
the entire program package (including effective
classroom management behaviors), which led
them to develop more positive attitudes about
parents. One way to examine these alternate
hypotheses of program effects would be
through a dismantling study comparing the ef-
fects of the entire program to a condensed ver-
sion with only content related to parent–teacher
relationships. Additionally, future studies are
needed that include measures of parent–teacher
and teacher–student interactions at multiple
time points to provide more fine-tuned analyses
of when and how changes in teacher perceptions
contribute to student success. Future studies are
also needed to assess parent perceptions of their
own involvement and of teacher efforts in in-
volving them in school. Such studies would
help determine if improvement in teacher per-
ceptions of parents coincide with parent percep-
tions and whether any new perceptions lead to
changes in behavior.
12 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Implications
Findings from the present study suggested
that training teachers to assess their perceptions
of parents and students and equipping them with
skills to be more effective in parent interactions
holds promise for promoting parent involve-
ment in education on a broad scale. Improve-
ments in teacher-perceived parent involvement
is associated with positive educational and so-
cial outcomes for all students. Given that teach-
ers interact with entire classrooms of students
and their parents each year, improving teacher
skill in and awareness of promoting parent in-
volvement has the potential to foster positive
youth development more efficiently than work-
ing with one parent at a time. For youth at
highest risk for academic and behavior prob-
lems, working with parents alone may not be
enough to improve success in school settings for
these children (Taylor & Biglan, 1998; Web-
ster-Stratton, 1990). Yet it may not be realistic
for a school psychologist to work with each
child’s teacher to promote more effective parent
involvement strategies. Instead, universal
teacher training programs like IY TCM may
complement efforts to improve academic and
social outcomes for all students, including those
with the greatest need.
School psychologists implementing evi-
dence-based parenting programs for children
with behavior disorders are encouraged to con-
sult with the child’s teacher about classroom
management issues based on IY TCM princi-
ples. Equipping teachers with skills and adap-
tive cognitions in working with parents of chil-
dren with behavior problems using IY TCM
strategies may help overcome the common
challenges teachers experience in these relation-
ships. Prior studies have implemented IY TCM
with teachers of students with conduct problems
whose parents also received IY Parent and
found additive effects on student outcomes
(Webster-Stratton et al., 2004). The present
findings suggested that part of the benefit of the
IY TCM approach on student outcomes may
arise from improvements in teacher assump-
tions about parents and/or knowledge of strate-
gies to improve parent school involvement and
not simply changes in the classroom environ-
ment. IY TCM also provides school psycholo-
gists with a training model for addressing
teacher biases about students and parents that
can undermine academic success. Given the
well-established influence of teacher expec-
tancy on student educational success (see
Rosenthal, 1994), any strategy for fostering
adaptive expectations of students and parents
will likely go a long way toward helping stu-
dents with the most intense needs in schools to
be successful.
Conclusion
Parent involvement in education is a valued
outcome in itself. Teacher beliefs about and
attitudes toward parents likely affect their inter-
actions with students. Thus, altering these per-
ceptions to be more favorable can promote stu-
dent learning and development. The findings
from the present study suggested that IY TCM
provides one method for fostering greater parent
involvement in education that may benefit stu-
dents with academic and behavior problems.
References
Akaike, H. (1987). Factor analysis and AIC. Psy-
chometrika, 52, 317–332. http://dx.doi.org/10
.1007/bf02294359
Arnold, D. H., Griffith, J. R., Ortiz, C., & Stowe,
R. M. (1998). Day care interactions and teacher
perceptions as a function of teacher and child
ethnic group. Journal of Research in Childhood
Education, 12, 143–154. http://dx.doi.org/10.1080/
02568549809594880
Bakker, J., Denessen, E., & Brus-Laeven, M. (2007).
Socio-economic background, parental involve-
ment and teacher perceptions of these in relation to
pupil achievement. Educational Studies, 33, 177–
192. http://dx.doi.org/10.1080/030556906010
68345
Barnard, W. M. (2004). Parent involvement in ele-
mentary school and educational attainment. Chil-
dren and Youth Services Review, 26, 39 – 62. http://
dx.doi.org/10.1016/j.childyouth.2003.11.002
Campbell, S. B., & Ewing, L. J. (1990). Follow-up
of hard-to-manage preschoolers: Adjustment at
age 9 and predictors of continuing symptoms.
Journal of Child Psychology and Psychiatry, 31,
871– 889. http://dx.doi.org/10.1111/j.1469-7610
.1990.tb00831.x
Carr, E. G., Taylor, J. C., & Robinson, S. (1991). The
effects of severe behavior problems in children on
the teaching behavior of adults. Journal of Applied
Behavior Analysis, 24, 523–535. http://dx.doi.org/
10.1901/jaba.1991.24-523
Conduct Problems Prevention Research Group (CP-
PRG). (1991). Parent and Teacher Involvement
13LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Measure – Teacher. Available from the Fast Track
Project website: http://www.fasttrackproject.org
Conduct Problems Prevention Research Group (CP-
PRG). (1995). Teacher–Social Competence Scale.
Available from the Fast Track Project website:
http://www.fasttrackproject.org
El Nokali, N. E., Bachman, H. J., & Votruba-Drzal,
E. (2010). Parent involvement and children’s aca-
demic and social development in elementary
school. Child Development, 81, 988 –1005. http://
dx.doi.org/10.1111/j.1467-8624.2010.01447.x
Gouley, K. K., Brotman, L. M., Huang, K. Y., &
Shrout, P. E. (2008). Construct validation of the
social competence scale in preschool-age children.
Social Development, 17, 380 –398. http://dx.doi
.org/10.1111/j.1467-9507.2007.00430.x
Henderson, A. T., & Berla, A. (1997). A new gener-
ation of evidence: The family is critical to student
achievement. Washington, DC: National Commit-
tee for Citizens in Education.
Herman, K. C., Reinke, W. M., Bradshaw, C., Loch-
man, J., Borden, L., & Darney, D. (2012). Increas-
ing parental engagement in school-based preven-
tive interventions: The integration of the Family
Check-Up and the Parent Coping Power Program.
In M. Weist, N. Lever, C. Bradshaw, & J. Owens
(Eds.), Handbook of school mental health (2nd ed.,
pp. 223–236). New York, NY: Springer.
Hughes, J., & Kwok, O. M. (2007). Influence of
student–teacher and parent–teacher relationships
on lower achieving readers’ engagement and
achievement in the primary grades. Journal of
Educational Psychology, 99, 39 –51. http://dx.doi
.org/10.1037/00220663.99.1.39
Izzo, C. V., Weissberg, R. P., Kasprow, W. J., &
Fendrich, M. (1999). A longitudinal assessment of
teacher perceptions of parent involvement in chil-
dren’s education and school performance. Ameri-
can Journal of Community Psychology, 27, 817–
839. http://dx.doi.org/10.1023/A:1022262625984
Kellam, S. G., Rebok, G. W., Ialongo, N., & Mayer,
L. S. (1994). The course and malleability of ag-
gressive behavior from early first grade into mid-
dle school: Results of a developmental epidemio-
logically-based preventive trial. Journal of Child
Psychology and Psychiatry, 35, 259 –281. http://dx
.doi.org/10.1111/j.1469-7610.1994.tb01161.x
Koth, C. W., Bradshaw, C. P., & Leaf, P. J. (2009).
Teacher Observation of Classroom Adaptation –
Checklist: Development and factor structure. Mea-
surement and Evaluation in Counseling and De-
velopment, 42, 15–30. http://dx.doi.org/10.1177/
0748175609333560
Lo, Y., Mendall, N. R., & Rubin, D. B. (2001).
Testing the number of components in a normal
mixture. Biometrika, 88, 767–778. http://dx.doi
.org/10.1093/biomet/88.3.767
Little, R. J. (1995). Modeling the dropout mechanism
in repeated-measures studies. Journal of the Amer-
ican Statistical Association, 90, 1112–1121. http://
dx.doi.org/10.2307/2291350
Malone, P. S. (2000). Parent and Teacher Involve-
ment Measure – Teacher: Year 3 update. Available
from http://www.fasttrackproject.org/
McCoach, D., Goldstein, J., Behuniak, P., Reis,
S. M., Black, A. C., Sullivan, E. E., & Rambo, K.
(2010). Examining the unexpected: Outlier analy-
ses of factors affecting student achievement. Jour-
nal of Advanced Academics, 21, 426 – 468. http://
dx.doi.org/10.1177/1932202X1002100304
McDermott, P., & Rothenberg, J. (2000). Why urban
parents resist involvement in their children‘s
elementary education. Qualitative Report,
5(3&4). Retrieved from http://www.nova.edu/ssss/
QR/QR5-3/mcdermott.html
McKay, M. M., Atkins, M. S., Hawkins, T., Brown,
C., & Lynn, C. J. (2003). Inner-City African
American Parental Involvement in Children’s
Schooling: Racial Socialization and Social Support
from the Parent Community. American Journal of
Community Psychology, 32(1–2), 107–114. http://
dx.doi.org/10.1023/a:1025655109283
Miedel, W. T., & Reynolds, A. J. (2000). Parent
involvement in early intervention for disadvan-
taged children: Does it matter? Journal of School
Psychology, 37, 379 – 402. http://dx.doi.org/10
.1016/s0022-4405(99)00023-0
Miller-Johnson, S., & Maumary-Gremaud, A.
(1995). Parent–teacher involvement: Parent ver-
sion (Fast Track Project Technical Report. Year 2).
Durham, NC: Duke University.
Muthén, L. K., & Muthén, B. O. (1998 –2010). Mplus
user’s guide (6th ed.). Los Angeles, CA: Author.
Nock, M. K., & Kazdin, A. E. (2001). Parent expec-
tancies for child therapy: Assessment and relation
to participation in treatment. Journal of Child and
Family Studies, 10, 155–180. http://dx.doi.org/10
.1023/A:1016699424731
Nylund, K., Muthén, B., Bellmore, A., Nishina, A.,
Graham, S., & Juvoven, J. (2005, May). The state
of victimization during middle school: A latent
transition mixture model approach. Paper pre-
sented at the annual convention of the Society for
Prevention Research, Washington, DC.
Park, E., Pullis, M., Reilly, T., & Townsend, B.
(1994). Cultural biases in the identification of stu-
dents with behavioral disorders. In R. L. Peterson
& S. Ishii-Jordan (Eds.), Multicultural issues in the
education of students with behavioral disorders
(pp. 14 –26). Cambridge, MA: Brookline Books.
Patterson, G., Reid, J., & Dishion, T. (1992). Antiso-
cial boys. Eugene, OR: Castalia Publishing Com-
pany.
Ramaswamy, V., DeSarbo, W. S., Reibstein, D. J., &
Robinson, W. T. (1993). An empirical pooling
14 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
approach for estimating marketing mix elasticities
with PIMS data. Marketing Science, 12, 103–124.
http://dx.doi.org/10.1287/mksc.12.1.103
Reinke, W. M., Herman, K. C., & Dong, N. (2014,
March). Incredible Years Teacher Classroom
Management Program: Initial findings from a
group randomized trial. Paper presented at the
Society for Research on Educational Effectiveness
Spring Conference, Washington, DC.
Reinke, W. M., Herman, K. C., Stormont, M., New-
comer, L., & David, K. (2013). Illustrating the
multiple facets and levels of fidelity of implemen-
tation to a teacher classroom management inter-
vention. Administration and Policy in Mental
Health and Mental Health Services Research, 40,
494 –506. http://dx.doi.org/10.1007/s10488-013-
0496-2
Reinke, W. M., & Newcomer, L. (2010). Brief Class-
room Interaction Observation Revised (BCIO-R)
manual. Columbia, MO: University of Missouri.
Reinke, W. M., Stormont, M., Herman, K. C., &
Newcomer, L. (2014). Using coaching to support
teacher implementation of classroom-based inter-
ventions. Journal of Behavioral Education, 23(1),
150 –167.
Rosenthal, R. (1994). Interpersonal expectancy ef-
fects: A 30-year perspective. Current Directions in
Psychological Science, 3, 176 –179. http://dx.doi
.org/10.1111/1467-8721.ep10770698
Schafer, J. L., & Graham, J. W. (2002). Missing data:
Our view of the state of the art. Psychological
Methods, 7, 147–177. http://dx.doi.org/10.1037/
1082-989X.7.2.147
Schwartz, G. (1978). Estimating the dimensions of a
model. The Annals of Statistics, 6, 461– 464.
Sclove, S. L. (1987). Application of model-selection
criteria to some problems in multivariate analysis.
Psychometrika, 52, 333–343. http://dx.doi.org/10
.1007/bf02294360
Sheridan, S. M., Bovaird, J. A., Glover, T. A., Gar-
bacz, S. A., Witte, A., & Kwon, K. (2012). A
randomized trial examining the effects of conjoint
behavioral consultation and the mediating role of
the parent–teacher relationship. School Psychology
Review, 41, 23– 46.
Stormont, M., Herman, K. C., Reinke, W. M., David,
K., & Goel, N. (2013). Latent profile analysis of
teacher perceptions of parent contact and comfort.
School Psychology Quarterly, 28, 195–209. http://
dx.doi.org/10.1037/spq0000004
Stormshak, E. A., Dishion, T. J., Light, J., & Yasui,
M. (2005). Implementing family-centered inter-
ventions within the public middle school: Linking
service delivery to change in student problem be-
havior. Journal of Abnormal Child Psychology,
33, 723–733. http://dx.doi.org/10.1007/s10802-
005-7650-6
Taylor, T. K., & Biglan, A. (1998). Behavioral family
interventions for improving child-rearing: A re-
view of the literature for clinicians and policy
makers. Clinical Child and Family Psychology
Review, 1, 41– 60. http://dx.doi.org/10.1023/A:
1021848315541
Tapp, J. (2004). Multi-option observation system for
experimental studies (MOOSES). Retrieved from
http://kc.vanderbilt.edu/mooses/mooses.html
Thompson, A. M., Macy, R. J., & Fraser, M. W.
(2011). Assessing person-centered outcomes in
practice research: A latent transition profile frame-
work. Journal of Community Psychology, 39, 987–
1002. http://dx.doi.org/10.1002/jcop.20485
Vuong, Q. H. (1989). Likelihood ratio tests for model
selection and non-nested hypotheses. Economet-
rica, 57, 307–333. http://dx.doi.org/10.2307/
1912557
Wagner, M., Kutash, K., Duchnowski, A. J., Epstein,
M. H., & Sumi, W. C. (2005). A national picture of
the characteristics of students with emotional dis-
turbances receiving special education. Journal of
Emotional and Behavioral Disorders, 13, 79 –96.
http://dx.doi.org/10.1177/10634266050130020201
Wang, Z., Rohrer, D., Chuang, C. C., Fujiki, M.,
Herman, K., & Reinke, W. (2015). Five methods
to score the Teacher Observation of Classroom
Adaptation Checklist and to examine group differ-
ences. The Journal of Experimental Education,
83(1), 24–50.
Webster-Stratton, C. (1990). Long-term follow-up of
families with young conduct problem children:
From preschool to grade school. Journal of Clin-
ical Child Psychology, 19, 144 –149. http://dx.doi
.org/10.1207/s15374424jccp1902_6
Webster-Stratton, C., Reid, J. M., & Stoolmiller, M.
(2008). Preventing conduct problems and improv-
ing school readiness: Evaluation of the Incredible
Years Teacher and Child Training Programs in
high-risk schools. Journal of Child Psychology
and Psychiatry, 49, 471– 488. http://dx.doi.org/10
.1111/j.1469-7610.2007.01861.x
Webster-Stratton, C., Reid, M. J., & Hammond, M.
(2001). Preventing conduct problems, promoting
social competence: A parent and teacher training
partnership in head start. Journal of Clinical Child
Psychology, 30, 283–302. http://dx.doi.org/10
.1207/S15374424JCCP3003_2
Webster-Stratton, C., Reid, M. J., & Hammond, M.
(2004). Treating children with early-onset conduct
problems: Intervention outcomes for parent, child,
and teacher training. Journal of Clinical Child and
Adolescent Psychology, 33, 105–124. http://dx.doi
.org/10.1207/S15374424JCCP3301_11
Webster-Stratton, C., & Reid, M. J. (2010). The
incredible years parents, teachers and child train-
ing series: A multifaceted treatment approach for
15LTA OF PARENT INVOLVEMENT PATTERNS
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
young children with conduct problems. In J. R.
Weisz, & A. E. Kazdin (Eds.), Evidence based
psychotherapies for children and adolescents (2nd
ed.). New York, NY: Guilford Press.
Woodcock, R. W., McGrew, K. S., & Mather, N.
(2001, 2007). Woodcock-Johnson III Tests of
Achievement. Rolling Meadows, IL: Riverside
Publishing.
Wyrick, A. J., & Rudasill, K. M. (2009). Parent
involvement as a predictor of teacher– child rela-
tionship quality in third grade. Early Education
and Development, 20, 845– 864. http://dx.doi.org/
10.1080/10409280802582803
Received June 25, 2015
Revision received May 12, 2016
Accepted May 14, 2016 䡲
16 HERMAN AND REINKE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
A preview of this full-text is provided by American Psychological Association.
Content available from School Psychology Quarterly
This content is subject to copyright. Terms and conditions apply.