ArticlePDF Available



Abstract and Figures

The purpose of this study was to investigate whether participating in science, technology, engineering, and mathematics (STEM) project-based learning (PBL) activities effected students who had varied performance levels and to what extent students’ individual factors influenced their mathematics achievement. STEM PBL has been a critical challenge to be embedded in schools, thus the effect of STEM PBL should to be examined. Teachers in 3 high schools attended sustained professional developments provided by 1 STEM center based in a Southwestern university and were required to implement STEM PBLs once in every 6 weeks for 3 years (2008 through 2010). The participants were 836 high school students in these 3 schools who took the Texas Assessment of Knowledge and Skills (TAKS) test and had scores at least in the initial year. Hierarchical linear modeling was used to analyze the data using student’s mathematics TAKS scores and demographic information for the longitudinal study. STEM PBL instruction influenced student achievement in mathematics by both student demographic backgrounds and performance levels. Low performing students showed statistically significantly higher growth rates on mathematics scores than high and middle performing students over the 3 years. In addition, student’s ethnicity and economic status were good predictors of academic achievement. Results of the present study implied that STEM PBLs in schools benefitted low performing students to a greater extent and decreased the achievement gap.
Content may be subject to copyright.
Received: 15 December 2012; Accepted: 13 February 2014
ABSTRACT. The purpose of this study was to investigate whether participating in
science, technology, engineering, and mathematics (STEM) project-based learning (PBL)
activities effected students who had varied performance levels and to what extent
studentsindividual factors influenced their mathematics achievement. STEM PBL has
been a critical challenge to be embedded in schools, thus the effect of STEM PBL should
to be examined. Teachers in 3 high schools attended sustained professional developments
provided by 1 STEM center based in a Southwestern university and were required to
implement STEM PBLs once in every 6 weeks for 3 years (2008 through 2010). The
participants were 836 high school students in these 3 schools who took the Texas
Assessment of Knowledge and Skills (TAKS) test and had scores at least in the initial
year. Hierarchical linear modeling was used to analyze the data using students
mathematics TAKS scores and demographic information for the longitudinal study.
STEM PBL instruction influenced student achievement in mathematics by both student
demographic backgrounds and performance levels. Low performing students showed
statistically significantly higher growth rates on mathematics scores than high and middle
performing students over the 3 years. In addition, students ethnicity and economic status
were good predictors of academic achievement. Results of the present study implied that
STEM PBLs in schools benefitted low performing students to a greater extent and
decreased the achievement gap.
KEY WORDS: academic achievement, education, longitudinal HLM, performance level,
project-based learning, STEM
Science, technology, engineering, and mathematics (STEM) education
has been discussed as a critical issue inside and outside of schools, and a
large share of funds have been invested to encourage students and to
increase educatorsinterests and efforts in STEM fields. Teachers in
traditional classrooms did not encourage students interest in STEM fields
Electronic supplementary material The online version of this article (doi:10.1007/s10763-014-
9526-0) contains supplementary material, which is available to authorized users.
International Journal of Science and Mathematics Education 2014
#National Science Council, Taiwan 2014
(Butzin, 2001; Dominguez & Jaime, 2010; Sahin, 2009). Moreover, the
lecture type classrooms did not facilitate students improvement in critical
thinking and problem solving skills (Rabe-Hemp, Woollen & Humiston,
2009; Tiwari, Lai, So & Yuen, 2006). STEM project-based learning (PBL)
is a current instructional strategy that is student driven, interdisciplinary,
collaborative, and technology based. According to the report from the
Federal Inventory of STEM Education Fast-Track Action Committee and
Committee on STEM Education National Science and Technology Council
(2011), of the total 3.4 billion dollars spent by US Federal agencies on
STEM education, about 1.1 billion dollars was invested in K-12, and
hundreds of programs were implemented within the boundaries of STEM
education. Compared to the amount of investment, however, the effect of
STEM education on K-12 education has not been studied using advanced
and multifaceted methodologies to investigate the practical impacts in
Students may exhibit differential achievement within the same learning
environment. The most appropriate learning environment can differ for
each student by characteristic. For example, female and male students
who were taught by the same teacher with the same textbook showed
varied achievement scores (Benbow, 2012; Matteucci & Mignani, 2011).
Furthermore, homogeneous student groups favored higher achievers,
whereas heterogeneous grouping was more effective for low achievers
(Cheng, Lam & Chan, 2008; Hooper & Hannafin, 1988; Robinson,
1990). No learning environment can be guaranteed as the best milieu for
every student without considering other complicated and possibly
confounding factors. STEM PBL has been employed in K-12 classrooms
to empower students to participate in interdisciplinary and collaborative
activities. However, there are no studies that examined how classroom
teachers who implement STEM PBL affect various student groups. The
main purpose of this research was to investigate the impact of STEM PBL
on students academic achievement when considering student level
Student Factor
Student achievement is influenced by individual factors. A students
gender, ethnicity, SES, and language proficiency were indicated as critical
factors affecting academic achievement (Konstantopoulos, 2009;
Lubienski, 2002; Ma & Klinger, 2000; Shores, Shannon & Smith, 2010;
Tate, 1997). According to large-scale studies, diverse, complex, and
varied combinations of gender, ethnicity, SES, and language proficiency
led to differential impact on achievement (Konstantopoulos, 2009;Ma&
Klinger, 2000; Tate, 1997). For example, studentsscores indicated
important differences by gender; however, the difference in mathematics
was smaller than in other subjects, i.e., science, reading, and writing
(Konstantopoulos, 2009; Ma & Klinger, 2000; Shores et al., 2010). SES
was a critical predictor of mathematics achievement even if other student
and school variables were controlled (Ma & Klinger, 2000). The
influence of ethnicity on broad outcome variables of achievement varied
according to study design and objectives (Capraro, 2001; Ma & Klinger,
2000). In addition, the impact of language proficiency on mathematics
achievement varied according to students ethnicity (Tate, 1997). The
gender factor showed a larger difference for Black, Pakistani, and
Bangladeshi students than for White children (Hansen & Jones, 2011).
Diverse levels of achievement among students exist in a classroom,
and teachers change their instructional approaches based on their beliefs,
attitudes, and expectations of studentsability levels (Richardson &
Fallona, 2010). Student achievement level was one critical factor
teachersused when deciding an instructional method, which in turn has
been shown to impact achievement for those students in subsequent years
(McKown & Weinstein, 2008). For example, a student-directed and self-
regulated learning environment where the teacher acted as a guide to
assist the studentslearning process was shown to be more effective for
students who had higher achievement (Yoon, 2009). On the other hand,
low achievers exhibited less of a desire for learning, self-control, and self-
management indicative of insufficient readiness for self-directed learning;
therefore, teachers were advised to be more deeply involved in the
learning processes for low achievers (Abraham, Fisher, Kamath, Izzati,
Habila & Atikah, 2011). Moreover, problem solving combined with a
computer adventure game intervention was also shown to be an effective
method for improving low achieversmathematics scores (Kajamies,
Vauras & Kinnunen, 2010). The interactive and stimulated components of
the game intervention were more appropriate and effective with low
Case by case, high and low achievers responded diversely to different
instructional approaches. For example, high, average, and low mathemat-
ics achievers displayed no meaningful differences in achieving benefits
when using a graphing calculator (Tan, 2012). However, low achievers
demonstrated more improvements than high achievers in solving
problems and comprehending ecological concepts when they were
engaged in peer discussions (Rivard, 2004).
Learning Environment Factor: STEM PBL
STEM PBL has positively influenced studentsnon-academic perfor-
mances. Students who have experienced STEM PBL showed positive
attitudes toward learning itself, team communication, and collabora-
tive behavior (Dominguez & Jaime, 2010; Johnson, Johnson &
Holubec, 1998; Kaldi, Filippatou & Govaris, 2011;vanRooij,2009;
Veenman, Kenter & Post, 2000). Furthermore, STEM PBL was
examined with respect to increasing studentsinterest, self-confidence,
and self-efficacy (Baran & Maskan, 2010), which was highly related
to the components of STEM BPL such as collaborations in group
work and contextual problems reflecting studentsreal world
experiences. In addition, students who studied in STEM PBL
classrooms were less likely to drop out of courses and school
(Dominguez & Jaime, 2010). As an exceptional case, Kaldi et al.
(2011) indicated that students had a difficult time and received
negative feedback from students of different ethnicities during group
work. For example, some students involved in group work from
Romania and Roma had a difficult time working together, even
though Greek primary schools have had multi-ethnic classes since the
Compared to the studies on the impact of STEM PBL on students
attitude and perspective on learning, few studies have investigated the
effect of STEM PBL on the improvement of student achievement.
Baran and Maskan (2010) examined the effect of STEM PBL at the
university level and presented positive effect sizes and statistically
significant differences between experimental and control groups. One
interesting result from their study was that studentsscores were
statistically significantly different for comprehension, but not for
knowledge and application. In addition, diverse components of STEM
PBL were pointed out to improve studentsacademic achievement.
Kaldietal.(2011) found that hands-on activities and field-based
contexts were the primary reasons for positive effects for students in
content knowledge and attitude toward learning. Furthermore, stu-
dents encouraged through STEM PBL type factors were required to
solve problems embedded in the project which improved their
problem solving skills (Barron et al., 1998;Boaler,1997).
Therefore, it is essential to develop an intervention that positively
influences attitude and perspective on learning when designing an
intervention especially in light of other student factors.
Two critical factors influence student mathematics achievement: (a)
learning environment factor (i.e. STEM PBL) and (b) student level
factors. This study was designed to provide insights into how STEM PBL
interacts with student level factors.
PBL is a developed instructional strategy grounded in the Kilpatrick
(1918) and Dewey (1938)sproject method. Two educational theorists,
Kilpatrick (1918)andDewey(1938), defined students as active
investigators, not passive recipients in the learning process, representing
radical constructivism in mathematics and science education.
Empowering students to be active learning constructors was the primary
feature of PBL as well as the project method. PBL has been developed
from the project method and is referred to in the boundary of the reform
movement in the early part of the twenty-first century. More recently,
PBL has been more generally implemented into K-12 education than
before to encourage studentsdeep understanding. Deep understanding
occurs when students are provided scaffolds and formative assessment
within social structures (Barron et al., 1998). STEM PBL referred to in
this study was based on the same theoretical background with PBL and
the interdisciplinary feature combining science, technology, engineering,
and mathematics was added.
STEM PBL is an instructional approach embedded in classrooms for
STEM education. STEM PBL is grounded in the theoretical background
of constructivism where students are engaged in the diverse components
of problem solving, interdisciplinary curriculum, open-ended questions,
hands-on activities, group work, and interactive group activities (Capraro
& Slough, 2008; Clark & Ernst, 2007; Dolmans, Grave, Wolfhagen &
Vleuten, 2005). For example, in STEM PBL classrooms, students are
required to solve problems and engage in ill-defined tasks within the
boundary of a well-defined outcome collaborating with other group
members. Effective STEM PBL should be interdisciplinary and contain
diverse content objectives within the context of hands on activities to
produce an artifact (Capraro & Slough, 2008). STEM PBL classrooms are
more student-centered, where the teacher is expected to play a role as a
guide (Clark & Ernst, 2007). STEM PBL is a new teaching strategy and
learning environment for teachers as well as students. This teaching
strategy can have profound effects while being implemented in
classrooms. Therefore, studies to evaluate the effects of implementing
STEM PBLs in schools for educators and teachers are necessary.
The impact of PBL features (i.e. student-centered, formative assess-
ment, and community-based learning environments) on diverse students
has been explored in previous studies. For example, low achievers could
be motivated through PBL as compared to high achievers with critical
thinking and group interactions (Horan, Lavaroni & Beldon, 1996). By
gender, female students preferred PBL type activities and demonstrated
higher achievement (Boaler, 1997). However, what has not been
examined is how STEM PBL features (i.e. interdisciplinary, technology-
based, and engineering-based activities) impacted students who are
involved in varied factors (Thomas, 2000). We assumed that the impact
of STEM PBL should be varied concerning studentsindividual factors
(see Appendix A). To provide more effective instruction, the impact of
STEM PBL should be evaluated with consideration toward individual
student factors. Very little information is available on the role of student
factors on learning during STEM PBL instruction. Specifically, no studies
were concerned with how STEM PBL works for students having varied
academic achievement performances. The present study offers profound
information about the effects of implementing STEM PBLs on mathe-
matics achievement while considering students diverse personal factors.
The purpose of this study was to investigate whether a pedagogical
strategy using STEM PBL, demonstrated differential effects on mathe-
matics achievement for students with varied performance levels (i.e. high,
middle, and low), and to what extent did studentsindividual factors (i.e.
gender, ethnicity, economic disability, English as a second language
(ESL), special education, gifted, and at-risk) influenced mathematics
achievement accompanied by their performance impact within the STEM
PBL instruction.
The participants were diverse students (N
= 836, N
= 533, and
= 485) enrolled in three small, urban, low socio-economic high
schools from 2008 to 2010. In the present study, students who took the
Texas Assessment of Knowledge and Skills (TAKS) test in 2008 were
selected, because students performance level in 2008 was the main
predictor in this study. Based on the demographics of the three schools in
2008, 412 students (49.3 %) were male. Largest majority of students were
Hispanic (n= 453, 54.2 %) and African American (n= 314, 37.6 %).
Additionally, there were 69 White and Asian students (8.25 %). We
focused on the analysis to examine the differences between Hispanic and
other students (i.e. African American, White and Asian), because
Hispanic students were the major population and have been underrepre-
sented in STEM subjects in this particular district. About 6.1 % and 2.3 %
of students were categorized as ESL and special education, respectively.
Approximately 85 % of students was eligible for free or reduced meals
under the National School Lunch and Child Nutrition Program, which
was regarded as an index of economic status. In addition, 518 (62 %)
students were categorized as at-risk.The Texas Education Agencys
[TEA] (2011) definition of at-risk included students who underperformed
on the state test, had limited English proficiency, or were in the care of a
state agency.
These teacher professional developments in STEM PBL were designed
and implemented under a state-wide project to improve students
readiness for postsecondary majors and professions especially with low-
income and low-performing students in STEM fields. The teachers in this
study attended a sustained period (30 sessions, 7 h per session) of PD
provided by one STEM center over a 3-year period. Study teachers were
required to teach one STEM PBL each 6 weeks. Teachers designed
STEM PBL lesson plans and cooperated with content specialists at the
STEM center thereafter to modify their lesson plans to enable the most
effective STEM PBLs enacted for students. Students who were selected
for this study participated in STEM PBL for 3 years in both their
mathematics and science classrooms. Therefore, two STEM PBL
activities were enacted with students every 6 weeks. STEM PBL lessons
were fundamentally based on mathematics and science content; but also
included technology and engineering content. The tasks of STEM PBL
implied real world contexts and students needed to apply their content
knowledge of subjects to achieve the well-defined outcome. The STEM
PBL projects were commonly implemented within groups, and students
more often communicated with their peers and the teacher.
For a comparative HLM model, 1,054 students in two high schools
(non-STEM PBL schools), who have not been educated using STEM
PBLs through the Texas High School Project, were selected as a control
group based on a propensity matching technique (Dehejia & Wahba,
2002). In Texas, STEM education for secondary students was initiated as
the Texas High School Project in 2007, and 51 high schools were
involved in this project. We selected two schools among non-STEM PBL
schools as a control group and examined whether students in STEM PBL
and non-STEM PBL schools had different base line mathematics
achievement (i.e. 2009).
Data Collection
The data for this study were studentsmathematics scores from the state
accountability assessment, TAKS, which provided empirical data (2008
to 2010). The employed analytic approach included controlled covariates
(i.e. studentsgender, ethnicity, economic status, ESL, special education,
and/or at-risk status) that may influence their achievement scores in
exploring STEM PBL across years. Students performance levels were the
main predictor for the outcome variable, studentsscores in 2010.
Reliability coefficients were used descriptively to evaluate to what
extent [we can] say that the data are consistent(Huck, 2008, p. 76). The
provided reliability for TAKS assessments ranged from 0.87 to .90
(reliability of TAKS-M assessments ranged from .82 to .88; TEA, 2008;
Zucker, 2003).
Data Analysis
Two methods were utilized to investigate the impact of STEM PBL on
students who had varied prior mathematics achievement: descriptive
statistics and longitudinal Hierarchical Linear Model (HLM). First,
descriptive statistics, including frequency, mean, standard deviation, and
correlation coefficients, were used to examine each variable. In addition,
skewness and kurtosis of the dependent variables were reported to
evaluate whether they were univariate normal.
Second, a longitudinal HLM analysis examined the two-level data
using SPSS version 21.0. Considering the 3-year longitudinal data, a
growth model was designed with a two-level hierarchy: time and student
level. At the time level, studentsmathematics scores were coded into
three time series. At the student level, students were divided into three
groups (i.e. high, middle and low achievers) according to their 2008
TAKS mathematics performance level. The 2010 TAKS mathematics
scores were used to measure student achievement. Lastly, effect sizes (i.e.
Hedgesg) were employed to contextualize the magnitude of differences
in means.
Grouping students into three groups was critical because the results
from the longitudinal analyses could possibly differ based on grouping
strategy. Navarro et al. (2012) used the normal distribution and standard
deviation (SD; i.e. Group 1 GXσ,XσGroup 2 X+σ,X
+σGGroup 3). In addition, Zady, Portes & Ochs (2003) employed some
specific scores to divide groups (i.e. low achievers 50 and high
achievers 70). A cumulative percentile approach was also employed to
assign students into several groups (Post et al., 2010; Stockdale &
Williams, 2004). In the current study, students were assigned into three
groups by the criteria offered by the test provider, TEA. The TEA
described three performance levels to divide students into groups (i.e.
did not meet the standard,”“met standard,and commended
performance). Accompanied by these descriptors of three performance
levels, TEA provided specific scores indicating each group. Based on the
2008 TAKS raw scores in mathematics, a score of less than 31 out of 52
indicated that students did not meet the standard, a score of 31 to 44 met
the standard, and students scoring 45 or above were commended
performance. Did not meet the standardmeant unsatisfactory
performance; below state passing standard; insufficient understanding of
the mathematics TEKS curriculum(TEA, 2009, p. 13), whereas met the
standardindicates satisfactory performance; at or above state passing
standard; sufficient understanding of the mathematics TEKS curriculum
(TEA, 2009, p. 13). Lastly, commended performancewas equated to
high academic achievement; considerably above state passing standard;
through understanding of the mathematics TEKS curriculum(TEA,
2009, p.13). Thus, these three student groups were regarded in this study
as low, middle, and high performance groups for convenience.
Depending on the main interests associated with the research questions,
we decided the reference groups of each predictor and covariates, and
coded them as 1. For the predictor variables, three performance levels
were the main research interests and the analyses were run twice (i.e. first
analysis contained the low performance group and second included the
middle performing students) as the reference groups. For the covariates,
student groups who were female, economically disadvantaged, ESL,
special education learners, and at-risk were considered as the reference
Benefits of Longitudinal HLM
To effectively investigate studentsindividual changes in mathematics
scores influenced by STEM PBL, we employed HLM as an analytic
approach (Hox, 2002; Raudenbush & Bryk, 2002). Longitudinal HLM is
a multi-level analytic approach, which regards individuals as the
second level and time points nested to an individual as the first level.
It was useless to compare student arithmetic mean scores at the
second and third time points, because students were at a different
level using base line data. Therefore, a longitudinal HLM analysis
was appropriate to answer the research questions because it was able
to provide studentsgrowthratesacross3yearsaswellasthe
longitudinal trajectory of studentsscores (Randenbush & Bryk,
2002). Moreover, the covariate variables were controlled in the
longitudinal HLM enabling us to more likely obtain the effects of
STEM PBL on studentsmathematical achievement. Even though we
could not control every factor, critical factors influencing students
achievement based on the literature review could be considered when
obtaining the results. This was the benefit of longitudinal HLM
within the limitations of a quasi-experimental design (Hox, 2002;
Randenbush & Bryk, 2002).
Longitudinal HLM has several benefits. First, it enables researchers
to have a larger number in their sample size than other quantitative
methodologies (e.g. ANOVA, ANCOVA, and MANOVA). This is
because it allows for having a different number of participants for
each time point. In other words, it is not necessary for each
individual to have the same number of time points in the longitudinal
HLM analysis and missing data, except for explanatory variables,
does not need to be excluded from the analysis (Hox, 2002).
Therefore, the numbers of students were different across years.
Another benefit of longitudinal HLM was the ability to have more
accurate estimates compared to other analyses. A traditional regres-
sion approach when used to analyze student level would inflate
standard errors and result in an inaccurate estimation of regression
coefficients (Cheng et al., 2008). That is, variables (dependent)
among student levels could be explained better by employing nested
data within HLM (Cheng et al., 2008).
Overview of Longitudinal HLM Models
Four models were designed and run to determine intra-class correlation
(ICC) and the percentage of explained variance by adding more
controlled covariates. The first model was designed to estimate the ICC,
which is a statistical measure related to the extent of how much
individually nested groups resembled each other. In the present study,
ICC indicated how strongly each individuals scores for 3 years were
correlated. The first model equations were:
ACHIEVEMENT ti ¼π0iþeti
π0i¼β00 þr0i
= student is TAKS mathematics score in year t
(2008, 2009, or 2010); π
= estimated score for student iacross years
(intercept); β
= grand mean of studentsscores in 2008 to 2010;
= deviation from that mean for student iin year t; and r
= random
effect for the intercept.
The second model was to investigate the effect of STEM PBL across
the years (2008 through 2010) without any predictors and covariates. The
second model equations were:
ACHIEVEMENT ti ¼π0iþπ1iYEAR2008ðÞ
ti þeti
π0i¼β00 þr0i
π1i¼β10 þr1i
where, π
= estimated score for student iin 2008 (intercept); β
= mean
of studentsscores in 2008; π
= estimated rate of (linear) change in score
for student ifrom 2008 to 2010 (slope); β
= average slope across
students; e
= within-person error of prediction (residual) for student i;
= random effect for the slope.
In the third model, the studentsperformance levels were included to
examine the effect of STEM PBL lessons on the improvement in
mathematics scores by the different performance levels. Students
individual factors were not yet considered in running the analysis. The
second level equations in the third model were:
π0i¼β00 þβ01 Performance12
ðÞþβ02 Performance13
π1i¼β10 þβ11 Performance12
ðÞþβ12 Performance13
where, β
= mean of studentsscores in 2008; β
= difference in average
intercept between middle and low performance groups; β
= difference in
average intercept between high and low performance groups;
= difference in average slope between middle and low performance
= difference in average slope between high and low
performance groups; r
= random effect for the intercept; r
= random
effect for the slope.
The fourth model included studentsindividual factors (i.e. gender,
ethnicity, economic disabilities, ESL, special education, at-risk) as
covariate variables and the predictor (i.e. performance levels). The second
level equations contained changes like those below:
π0i¼β00 þβ01 GenderðÞþβ02 EthnicityðÞþβ03 EcoDðÞþβ04ESLðÞþ
π1i¼β11 þβ11 GenderðÞþβ12 EthnicityðÞþβ13 EcoDðÞþβ14 ESLðÞþ
β15 SEðÞþβ16 AtRiskðÞþβ17 Performance12
ðÞþβ18 Performance13
where, EcoD = economic status; SE = special education; β
4, 5, 6, 7,or 8) = difference between groups (female vs. male, Hispanic vs.
others, economic disabled vs. others, special education vs. others, at risk
vs. others, middle vs. low performance groups, and high vs. low
performance groups, respectively) in average intercept; β
4, 5, 6, 7,or 8) = difference between groups (female vs. male, Hispanic vs.
others, economic disabled vs. others, special education vs. others, at risk
vs. others, middle vs. low performance groups, and high vs. low
performance groups, respectively) in average slope. The independent
variables of individual factors were controlled in this analysis to examine
the pure effect of STEM PBL on student academic achievement, rather
than considered them as interesting focal variables. However, conditional
second level equations in the third and fourth models still enabled the
researchers to examine the group differences associated with individual
student factors.
After four linear models fitted the collected data, 2 × restricted log
likelihood (2LL) was utilized to compare both the fixed effect and the
variance component estimates and to examine which model should be
selected (Raftery, 1996). 2LLs of four models were reported and smaller
values of 2LL indicated better-fit models.
Finally, to determine the effectiveness of STEM PBL on low
performing, economically disadvantaged, Hispanic and at-risk students,
we ran the comparative HLM analysis with 836 students in STEM PBL
schools and 1,054 students in non-STEM PBL schools. Two models were
developed: (a) STEM PBL and non-STEM PBL schools combination
model (i.e. fifth model), and (b) non-STEM PBL schools individual
model (i.e. sixth model). In the fifth model, students in STEM PBL and
non-STEM PBL schools were grouped into six groups (i.e. STEM PBL
and low achievers, STEM PBL and middle achievers, STEM PBL and
high achievers, non-STEM PBL and low achievers, non-STEM PBL and
middle achievers, and non-STEM PBL and high achievers) and their
growth rates in mathematics achievement were estimated by controlling
gender, ethnicity, economic disadvantage, special education, ESL, at-risk
covariates. In addition, non-STEM PBL studentsscores were separately
analyzed based on the fourth model to compare those in STEM PBL
schools and to explicate the impact of economic disadvantage, ethnicity,
and at-risk factors on student mathematics achievement with the STEM
PBL learning factor.
Descriptive Summaries and Correlation Coefficients
Descriptive statistics were employed to illustrate the distribution of the
participants across variables used in the study. Descriptive statistics
including frequency, mean, and standard deviation were reported (see
Appendix B). From the descriptive statistics, it was apparent that
economically disadvantaged, ESL, special education, non-gifted, and at-
risk students performed below their counterparts, whereas performance
was equal by ethnicity. In addition, student characteristics were varied
across three performance levels. There were 505 students who did not
meet the standard, 264 who met the standard, and 67 who had
commended performance. First, the low and middle performance groups
consisted of almost an even ratio of gender (malefemale = 1:1.02,
1:1.13, respectively) with less male students. In the high achievement
group, however, male students (n= 38) outnumbered female (n= 29).
The percentage of Hispanic students ranged from 50.9 % to 60.2 % across
the three performance groups. On the other hand, low, middle, and high
performance groups represented varied distributions of economic status
and at-risk students. When considering economic status, more than 80 %
of the students in low and middle performance groups as compared to
61 % of the students in the high group were economically disadvantaged.
For the at-risk variable, more than 80 % of students in the low
performance group, 40 % in the middle performance group, and less
than 6 % in the high performance group were at-risk. Only about 4 % of
the students in the low performance group were classified as special
education with no students in the high and middle performance groups
containing students in that category. About 9 % and 3 % of students in the
low and middle performance groups were ESL and there were no ESL
students in high performance group.
Bivariate correlations among the variables were calculated to examine
to what extent, student individual factors were related (Table 1). Student
scores were correlated positively with ethnicity, whereas negatively with
economically disadvantaged, ESL, special education, and at-risk charac-
Bivariate correlations for student-level variables
Correlations Gender Grade Ethnicity Economic Disadvantaged ESL Special Education At-Risk Achievement
Gender 1 .023 .080** .025 .029 .046* .013 .003
Grade 1 .033 .013 .047* .006 .054* .378**
Ethnicity ––1.132** .112** .014 .205** .057*
Economic Disadvantaged ––– 1 .046* .028 .212** .145**
ESL ––– – 1.038 .168** .139**
Special Education ––– – 1 .106** .147**
At-Risk ––– – 1.449**
Achievement ––– – 1
Skewness 0.029 0.363 0.106 1.843 3.693 6.188 0.664 0.413
Kurtosis 2.001 1.431 0.235 1.400 11.653 36.326 1.560 0.046
*pG.05; **pG.01
teristics. Being Hispanic was correlated slightly with higher scores.
Students who were classified as economically disadvantaged, ESL,
special education, and at-risk were correlated to lower scores than
students not classified in these categories.
HLM Analyses of StudentsScores and Individual Factors
The longitudinal data including studentsmathematics scores and
individual factors were analyzed using HLM following the method
described by Hox (2002) and using HLM 7 software. Treating students
repeated scores for 3 years as nested within individual students allowed
for longitudinal analyses of the given data and four kinds of HLM
models, and permitted assess to whether student individual factors
affected mathematics test scores. The first model was the unconditional
model in which only outcome variable was modeled to determine the
variation within cases.
Unconditional Model: Model 1. The employed unconditional model
included only an outcome variable without any predictors and examined the
extent to which studentsscores statistically varied over time. The grand mean
was 29.23, and the estimated within-student variance (σ
) and between-
student variance (τ
) were 57.93 and 120.05, respectively. The unconditional
model did not include any predictors in level-1 and level-2 equations, and
allowed us to examine how much percentage of the total variance was
explained by STEM PBL for 3 years and how much was by student individual
factors. The ICC (Raudenbush & Bryk, 2002) from the first model was
calculated by the formula, ρ=τ
) and 0.675. In other words, 67.5 %
of the total variance in mathematics scores could be explained by individual
student factors and 33.5 % was attributable to the changes during STEM PBL.
The Final Model: Model 4. The fourth model contained predictors (i.e.
and performance
) and covariate variables (i.e. gender,
ethnicity, economic status, ESL, special education, and at-risk) in level
2, including the interaction effects of time controlling for any impact of
student individual factors. The HLM results for the fourth model were
summarized in Table 2showing two effects: main and interaction
effects. First, the estimates of main effects indicated how much each
predictor and covariate variable influenced studentsinitial score in
2008 (i.e. intercept). The main effects of time (β
=5.66,t= 5.949;
pG0.001), performance
= 16.046, t= 27.554; pG0.001),
= 25.403, t= 25.362; pG0.001), ESL, special
education, and at-risk were statistically significant. Moreover, the
nteraction effects of ethnicity, economic status, performance
the difference between middle and low level of performance
groups), and performance
(i.e. the difference between high and
low level of performance groups) with time variable were statisti-
cally significant.
To determine the different growth rate between middle and high
performance groups, another analysis was conducted using the middle
performance group as a reference group. The estimated values of two
interaction effects, time × performance
(i.e. the difference between high
and low levels of performance groups) and time × performance
, were
2.584 (t= 4.378; pG0.001) and 0.265 (t=0.293; p= 0.769). In other
words, the three estimates of all three performance groups were positive;
Estimates, variances, and effect sizes
Effect Estimate Std. Error t/Wald Z
Effect Size
Fixed effect
Intercept 20.646 0.839 24.608 4.890
Time 5.656* 0.951 5.494 1.340
Gender 0.128 0.488 0.262 0.030
Ethnicity 0.203 0.512 0.397 0.048
Economic Status 0.890 0.698 1.276 0.211
ESL 5.473* 1.054 5.190 1.296
Special Education 6.267* 1.671 3.751 1.484
At-Risk 2.081* 0.575 3.618 0.493
16.046* 0.582 27.554 3.800
25.403* 1.001 25.362 6.016
Time × Gender 0.458 0.505 0.907 0.108
Time × Ethnicity 1.159* 0.541 2.142 0.275
Time × Economic Status 2.165* 0.796 2.719 0.513
Time × ESL 1.729 1.115 1.551 0.410
Time × Special Education 1.839 2.231 0.824 0.436
Time × At-Risk 0.769 0.640 1.200 0.182
Time × Proficiency
2.585* 0.590 4.378 0.613
Time × Proficiency
2.850* 0.981 2.906 0.675
Random effect variance
33.938 2.763 12.285
τ00 17.826 3.576 4.985
12.772 2.827 4.518
ESL = English as a Second Language
however, the interaction effect between time and performance
not a statistically significant predictor of student mathematics scores
on TAKS. That is, the middle and high performance groups
demonstrated a statistically significant lower growth rate than the
low-level performance group during 3 years, whereas the growth rate
of the high performance group did not differ from the middle
performance group (see Fig. 1).
Time-invariant covariates presented varied estimates and significant p-
values (see Table 2). Among the main effects, the predictor variable (i.e.
performance level) and three covariates (i.e. ESL, special education, and
at-risk variables) were examined and determined to be statistically
significant, whereas, gender, ethnicity, economic status were not. That
is, students individual factors such as performance level, ESL, special
education, and at-risk, affected the initial scores in 2008 (i.e. intercepts of
the three trajectory lines in Fig. 1). Other than the interaction effects of
time with performance, interaction effects of time with gender, ethnicity,
economic disability, ESL, special education, gifted, and at-risk were
examined to determine whether they were significant predictors of student
achievement in mathematics. The interaction effects of time with ethnicity
= 1.159, t= 2.142; p= 0.033) and economic status (β
t=2.719; p= 0.007) were statistically significant. In other words, these
two interaction effects significantly impacted the slope of growth
trajectory lines in Fig. 1.
2008 2009 2010
Low performance
Middle performance
High performance
Fig. 1. Growth trajectory of diverse proficiency groups for 3 years
Additionally, standardized effect sizes were calculated by the following
equation (Hedges, 2007):
For example, a significant performance level of fixed effect between
low and middle groups was observed β
= 16.046, pG0.05) where the
mean score of the middle group (β
middle group
= 36.692) was higher than
the mean score of the low group (β
middle group
= 20.646). The standardized
effect size of performance levels between low and middle groups was 3.8.
Among the interaction effects, the growth rate interaction effect of
ethnicity was significant (β
= 1.159, pG0.05) and the growth rate of
Hispanic students (β
= 6.815) was higher than others (β
= 5.656). The standardized effect size of the ethnicity interaction
effect was 0.275. Similarly, other effect sizes were calculated and
interpreted (see Table 3).
In summary, the results showed that the studentsachievement in
mathematics was dependent on multiple factors as well as STEM PBL
instruction. Students who were high and middle level performers in
mathematics demonstrated almost no differences in terms of growth rate
of mathematics scores over 3 years. In addition, low performing groups of
students showed significantly higher growth rates than the high and
middle performing groups of students. That is, the enactment of STEM
PBLs in classrooms was more likely to demonstrate positive impacts on
Percent of variance explained at level 1 and level 2
Model Added Variables
explained at
level 1 (σ
explained at
level 2 (τ
2 Restricted Log
Likelihood (2LL)
Model 1 ––12,930.66
Model 2 Time 44.78 % 1.27 % 12,596.76
Model 3 Model 2 + Proficiency
42.13 % 82.16 % 11,684.81
Model 4 Model 3 + Gender,
Ethnicity, Economic
Status, ESL, Special
Education, At-Risk
41.41 % 85.15 % 11,600.91
Model 1 = unconditional model
students in low performance groups, rather than in the high and middle
performing groups.
Auxiliary Statistics. To obtain information on the longitudinal HLM models,
two auxiliary statistics, variance explained and 2LL were reported (Table 3).
The variance explainedwas computed for models 2 through 4 to estimate
how much within- and between-student variances (τ
and σ
were further explained as more predictors and covariate variables were added
(Raudenbush & Bryk, 2002). The proportions of variance explained of σ
and τ
were calculated at level 1 and level 2, respectively. Another auxiliary
statistics, 2LL, was calculated to select the best-fit model for the collected
data. The 2LL value of the fourth model was smallest, which indicated the
best-fit model.
Comparative HLM Analysis: STEM PBL vs. Non-STEM PBL Schools
The result of the fifth model with students in both STEM PBL and non-
STEM PBL schools verified the positive effect of STEM PBL on low
achievers. The difference of growth rates between low achievers in STEM
PBL schools and non-STEM PBL schools was 1.634, which was statistically
significant (pG0.01) with the higher growth rate of students in STEM PBL
schools. In addition, the result of the separate model (i.e. sixth model)
examined the positive impact of student ethnicity, economic disadvantaged
and at-risk factors within the STEM PBL learning environment. Hispanic
students in STEM PBL schools showed higher growth rate than non-Hispanic
students (β= 1.159, pG0.05), whereas the growth rates of Hispanic and non-
Hispanic students in non-STEM PBL schools were not statistically different
from each other. Economically disadvantaged students in STEM PBL
schools showed lower growth rate than those who were not economically
disadvantaged (β=2.165, pG0.05), whereas the economic disadvantage
factor was not statistically significant in non-STEM PBL schools. Moreover,
at-risk students in non-STEM PBL schools had statistically significantly
lower growth rate than non-at-risk students (β=2.573, pG0.05); however,
the difference of growth rates between at-risk students and non-at-risk
students were nonexistent in STEM PBL schools.
This study contributes to the scholarly significance of understanding the
effect of STEM PBL activities on student achievement. We found a
positive growth rate in studentsacademic achievement in mathematics
while STEM PBLs were implemented at the high school level, similar to
Baran and Maskan (2010) who reported positive effect sizes when
implementing STEM PBL activities at the university level. Results of the
present study supports the work others (cf. Capraro, Capraro, Morgan, 2013)
to provide varied learning environments for students who are at different
performance levels. Students in high, middle, and low performing groups in
this study demonstrated varied growth rates, which indicates that a learning
environment may influence different impacts on each performance group. In
other words, components of STEM PBLs such as group projects,
collaboration, ill-defined tasks, and student-centered environments inter-
relationally function with each other, and some components of STEM PBL
are more appropriate for specific performance levels of students (Abraham et
al., 2011; Cheng et al., 2008; Kajamies et al., 2010). Therefore,
implementing STEM PBLs in schools can have diverse impacts on student
achievement and attitude according to their performance levels.
Conversely, results of the present study differed from Yoon (2009)s
research concluding that high achievers received more positive impact with
student-directed and self-regulated learning environments. A student
centered learning environment is the main feature of a STEM PBL classroom
and we found that the low performing group of students improved at a higher
level than the high and middle performing groups when looking at student
achievement on mathematics under a STEM PBL learning environment.
The results of the present study support the findings that individual
student factors influence student academic achievement within the STEM
PBL learning environment. As Ma and Klinger (2000) insisted, SES was
a critical predictor of studentsmathematics scores in the regular classes.
According to the results of this study, a students economic status was
also found to be an important factor in improving mathematics test scores
through STEM PBL experiences. The estimate of the interaction effect of
time and economic status was negative indicating that students who were
of low economic status (i.e. students eligible for the free meal or reduced
meal) showed a negative growth achievement rate while engaging in
STEM PBL over the 3 years. The implication of the relationship between
students SES and academic achievement should be regarded as a serious
problem because a students economic status was a critical factor
influencing a students academic achievement in mathematics, even
though there were no statistically significant differences in the initial year.
In other words, low economic status was not a barrier for students in the
first year of this study; however, students in the low economic status
group ultimately received negative impacts from their engagement in
STEM PBLs. Educators will need to consider if any particular aspect of
STEM PBL was as barrier for economically disadvantaged students
learning processes.
When examining the factor of students ethnicity, there have been debates
regarding the impact of ethnicity on studentsacademic achievement with
the results varying by the research design and participantscharacteristics
(Capraro, 2001; Ma & Klinger, 2000). This study contained mostly Hispanic
student participants and showed a significant difference compared to the
other ethnic groups. Hispanic students had a higher growth rate on
mathematics tests for 3 years during the implementation of STEM PBL
activities. That is, results from this study imply that STEM PBL activities
benefitted Hispanic students to a greater extent than other student groups
within the contexts of this study. Hispanic students might have had
additional opportunities to practice mathematical academic terminologies
within their group, which might have resulted in a higher growth rate in
mathematics achievement (Capraro, Capraro, Yetkiner, Rangel-Chavez &
Lewis, 2010). STEM PBL provided more opportunities for Hispanic
students to communicate with peers and teachers than would traditional
lecture. The participantsdemographic feature may be a limitation of this
study, because it was hard to extend the results of this study to a comparison
of Hispanic students with African America, White, or Asian students on
mathematics performance, separately.
Lastly, this study likely represents one of the first studies utilizing
advanced research analysis. Whereas most of studies utilized ttest,
correlation, ANOVA, and ANCOVA (Baran & Maskan, 2010;
Dominguez & Jaime, 2010; Lou et al., 2011; Kaldi et al., 2011; van
Rooij, 2009), the present study employed longitudinal HLM, with diverse
student factors examining the effect of implementing STEM PBLs on
student achievement. By using longitudinal HLM, we investigated the
trajectory of improvement in studentsacademic achievement, not just a
simple comparison at a specific point in time. In addition, the estimates of
fixed and random effects in this study are more accurate than other
studiesresults because we controlled for more variables by using
longitudinal data.
Developing effective STEM education has been regarded as one of the
most significant challenges facing educators along with improvement in
student performance in the areas of science, mathematics, and engineer-
ing. However, the effectiveness of implementing STEM PBL in terms of
improving studentsscores in mathematics and science has not demon-
strated as much improvement as was previously expected. This study
provides an evaluation of implementing STEM PBL activities in schools
to determine if there would be an improvement in studentsacademic
achievement in mathematics. These findings should assist teachers and
educators to rethink how students of varied performance levels benefit
from engaging in STEM BPL activities, and guide them in restructuring
their instructional strategies for engaging diverse learners in their
classrooms. The results, that low achievers and Hispanic studentsgrowth
rates were statistically significantly higher through STEM PBLs, should
be considered by policy makers, educators, and teachers in designing
differentiated instruction.
For further study, we would like to suggest that researchers clarify the
reasons for the results obtained in this study. That is, they should
investigate why students of different performance levels showed different
growth rates and how student individual factors functioned with diverse
components of STEM PBL. For example, the low performing group in
this study showed more positive impacts from group collaborations while
engaging in STEM PBL classroom activities similar to Rivard (2004)s
study. However, it was impossible to determine why the positive impact
on low achievers resulted from the heterogeneous grouping in STEM
PBLs (Cheng et al., 2008). The data in this study were limited to disclose
the effectiveness of STEM PBL, thus not enough to investigate how and
why STEM PBLs positively influenced student achievement.
Abraham, R. R., Fisher, M., Kamath, A., Izzati, T. A., Habila, S. & Atikah, N. N. (2011).
Exploring first-year undergraduate medical studentsself-directed learning readiness to
physiology. Advances in Physiology Education, 35(4), 393395.
Baran, M. & Maskan, A. (2010). The effect of project-based learning on pre-service
physics teachers electrostatic achievements. Cypriot Journal of Educational Sciences,
5(4), 243257.
Barron, B. J. S., Schwartz, D. L., Vye, N. J., Moore, A., Petrosino, A., Zech, L.,
Bransford, J. D. & The Cognition and Technology Group at Vanderbilt (1998). Doing
with understanding: Lessons from research on problem- and project-based learning. The
Journal of the Learning Sciences, 7(3&4), 271311.
Benbow, C. P. (2012). Identifying and nurturing future innovators in science, technology,
engineering, and mathematics: A review of findings from the study of mathematically
precocious youth. Peabody Journal of Education, 87(1), 1625.
Boaler, J. (1997). Experiencing school mathematics: Teaching styles, sex, and settings.
Buckingham: Open University Press.
Butzin, S. M. (2001). Using instructional technology in transformed learning environ-
ments: An evaluation of project CHILD. Journal of Research on Computing in
Education, 33(4), 367373.
Capraro, R. M. (2001). Exploring the influences of geometric spatial visualization,
gender, and ethnicity on the acquisition of geometry content knowledge. New Orleans:
Paper presented at the annual meeting Southwest Educational Research Association.
ERIC Document Reproduction Service No. ED451057.
Capraro, R. M. & Slough, S. W. (2008). Why PBL? Why STEM? Why now? An
introduction to project-based learning: An integrated science, technology, engineering,
and mathematics (STEM) approach. In R. M. Capraro & S. W. Slough (Eds.), Project
based learning: An integrated science technology engineering and mathematics (STEM)
approach (pp. 16). Rotterdam: Sense.
Capraro, M. M., Capraro, R. M., Yetkiner, Z. E., Rangel-Chavez, A. F. & Lewis, C. W.
(2010). Examining Hispanic student mathematics performance on high-stakes tests: An
examination of one urban school district in Colorado. Urban Review, 42, 193209.
Capraro, R. M., Capraro, M. M. & Morgan, J. (Eds.). (2013). Project-based learning: An
integrated science, technology, engineering, and mathematics (STEM) approach (2nd
ed.). Rotterdam: Sense.
Chen, R. W., Lam, S. F. & Chan, J. C. (2008). When high achievers and low achievers
work in the same group: The roles of group heterogeneity and processes in project-
based learning. British Journal of Educational Psychology, 78, 205221.
Clark, A. C. & Ernst, J. V. (2007). A model for the integration of science, technology,
engineering and mathematics. The Technology Teacher, 66(4), 2426.
Dehejia, R. H. & Wahba, S. (2002). Propensity score-matching methods for nonexper-
imental causal studies. The Review of Economics and Statistics, 84, 151161.
Dewey, J. (1938). Experience and education. New York: Macmillan.
Dolmans, D., Grave, W. D., Wolfhagen, I. & Vleuten, C. (2005). Problem based learning:
Future challenges for educational practice and research. Medical Education, 39,732741.
Domínguez, C. & Jaime, A. (2010). Database design learning: A project-based approach
organized through a course management system. Computers & Education, 55(3), 1312
Federal Inventory of STEM Education Fast-Track Action Committee & Committee on
STEM Education National Science and Technology Council. (2011). The federal
science, technology, engineering, and mathematics (STEM) education portfolio.
Retrieved from
Hansen, K. & Jones, E. M. (2011). Ethnicity and gender gaps in early childhood. British
Educational Research Journal, 37(6), 973991. doi:10.1080/01411926.2010.515018.
Hedges, L. V. (2007). Effect sizes in cluster-randomized designs. Journal of Educational
and Behavioral Statistics, 32, 341370.
Hooper, S. & Hannafin, M. J. (1988). Cooperative CBI: The effects of heterogeneous
versus homogeneous grouping on the learning of progressively complex concepts.
Journal of Education Computing Research, 4, 413424.
Horan, C., Lavaroni, C. & Beldon, P. (1996). Observation of the tinker tech program
students for critical thinking and social participation behaviors. Novato: Buck Institute
for Education.
Hox, J. (2002). Multilevel analysis: Techniques and applications. Mahwah: LEA.
Huck, S. W. (2008). Reading statistics and research. Boston: Pearson Education.
Johnson, D. W., Johnson, R. T. & Holubec, E. (1998). Cooperation in the classroom (6th
ed.). Edina: Interaction Book Company.
Kajamies, A., Vauras, M. & Kinnunen, R. (2010). Instructing low-achievers in
mathematical word problem solving. Scandinavian Journal of Educational Research,
54(4), 335355.
Kaldi, S., Filippatou, D. & Govaris, C. (2011). Project-based learning in primary schools:
Effects on pupilslearning and attitudes. Education 313, 39(1), 3547.
Kilpatrick, W. H. (1918). The project method. Teachers College Record, 19, 319335.
Konstantopoulos, S. (2009). Effects of teachers on minority and disadvantaged
studentsachievement in the early grades. The Elementary School Journal, 110(1),
Lou, S. J., Liu, Y. H., Shih, R. C., Chuang, S. Y., & Tseng, K. H. (2011). Effectiveness of
on-line STEM project-based learning for female senior high school students.
International Journal of Engineering Education, 27(2), 399410.
Lubienski, S. T. (2002). A closer look at black-white mathematics gaps: Intersections of
race and SES in NAEP achievement and instructional practice data. The Journal of
Negro Education, 71(4), 269287.
Ma, X. & Klinger, D. A. (2000). Hierarchical linear modeling of student and school
effects on academic achievement. Canadian Journal of Education, 25(1), 4155.
Matteucci, M. & Mignani, S. (2011). Gender differences in performance in mathematics at
the end of lower secondary school in Italy. Learning and Individual Differences, 21(5),
543548. doi:10.1016/j.lindif.2011.03.001.
McKown, C. & Weinstein, R. S. (2008). Teacher expectations, classroom context, and the
achievement gap. Journal of School Psychology, 46(3), 235261.
Navarro, J. I., Aguilar, M., Marchena, E., Ruiz, G., Menacho, I. & Van Luit, J. E. H.
(2012). Longitudinal study of low and high achievers in early mathematics. British
Journal of Education Psychology, 82,2841.
Post, T. R., Medhanie, A., Harwell, M., Norman, K. W., Dupuis, D. N., Muchlinski, T.,
Andersen, E. & Monson, D. (2010). The impact of prior mathematics achievement on
the relationship between high school mathematics curricula and postsecondary
mathematics performance, course-taking, and persistence. Journal for Research in
Mathematics Education, 42(3), 274308.
Rabe-Hemp, C., Woollen, S. & Humiston, G. (2009). A comparative analysis of student
engagement, learning, and satisfaction in lecture hall and online learning settings.
Quarterly Review of Distance Education, 10(2), 207218.
Raftery, A. E. (1996). Bayesian model selection in social research. In P. V. Marsden (Ed.),
Sociological methodology (pp. 111163). Oxford: Basil Blackwell.
Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical linear models: Applications and
data analysis methods (2nd ed.). Thousand Oaks: Sage.
Richardson, V. & Fallona, C. (2010). Classroom management as method and manner.
Journal of Curriculum Studies, 33(6), 705728.
Rivard, L. P. (2004). Are language-based activities in science effective for all students,
including low achievers? Science Education, 88(3), 420442.
Robinson, A. (1990). Cooperation and exploitation? The argument against cooperative
learning for talented students. Journal for the Education of the Gifted, 14,927.
Sahin, M. (2009). Correlations of studentsgrades, expectations, epistemological beliefs
and demographics in a problem-based introductory physics course. International
Journal of Environmental & Science Education, 4(2), 169184.
Shores, M. L., Shannon, D. M. & Smith, T. G. (2010). Individual learner variables and
their effect on mathematics achievement as students advance from fifth to sixth grade.
Journal of Research in Childhood Education, 24(3), 187194.
Stockdale, S. L. & Williams, R. L. (2004). Cooperative learning groups at the college
level: Different effects on high, average, low exam performers. Journal of Behavioral
Education, 13(1), 3750.
Tan, C. K. (2012). Effects of the application of graphing calculator on students
probability achievement. Computers & Education, 58(4), 11171126.
Tate, W. F. (1997). Race-ethnicity, SES, gender, and language proficiency trends in
mathematics achievement: An update. Journal for Research in Mathematics Education,
28, 652679.
Texas Education Agency. (2008). Chapter 16: Reliability. Retrieved from http://
Texas Education Agency. (2009). Texas assessment of knowledge and skills performance
level descriptors. Retrieved from
Texas Education Agency. (2011). Selected sections from Texas Education Code (TEC)
and Texas Administrative Code (TAC). [PDF document]. Retrieved from http://
Thomas, J. W. (2000). A review of research on project-based learning. San Rafael:
Autodesk Foundation.
Tiwari, A., Lai, R., So, M. & Yuen, K. (2006). A comparison of the effects of problem-
based learning and lecturing on the development of studentscritical thinking. Medical
Education, 40, 547554.
Van Rooij, S. W. (2009). Scaffolding project-based learning with the project management
body of knowledge (PMBOK®). Computers & Education, 52(1), 210219.
Veenman, S., Kenter, B. & Post, K. (2000). Cooperative learning in Dutch primary
classrooms. Educational Studies, 26(3), 281302. doi:10.1080/03055690050137114.
Yoon, C. H. (2009). Self-regulated learning and instructional factors in the scientific inquiry of
scientifically gifted Korean middle school students. Gifted Child, 53(3), 203216.
Zady, M. F., Portes, P. R. & Ochs, V. D. (2003). Examining classroom interactions related
to difference in studentsscience achievement. Science Education, 87(1), 4063.
Zucker, S. (2003). Fundamentals of standardized testing. San Antonio: Pearson.
Sunyoung Han
Texas Tech University
Lubbock, TX, USA
Robert Capraro and Mary Margaret Capraro
Texas A&M University
College Station, TX, USA
... Students are generally categorised into ability groups based on their intelligence quotient or academic achievements, which is likewise referred to as ability class (high, medium and low). Previous literature has reported achievement gaps amongst students at national and international levels in traditional learning environments (Gambari et al., 2013;Han, Capraro, & Capraro, 2014;Yu, She, & Lee, 2010). Teachers utilising traditional approaches believe that teaching low academic-ability students, complex and abstract learning content is not appropriate because they may not be able to cope with complex tasks (Yu et al., 2010). ...
... The control group consisted of 49 science students, with 15 in the high-ability group, 19 in the medium-ability group and 15 in the low-ability group. In this study, the average science achievement from the previous year was utilised to classify students as follows: ≥ 70% as high, ≥ 50 -69% as medium and ≤ 49 as low academic ability (Han et al., 2014). ...
... The results indicated that high, medium and low academic-ability students benefitted; in addition, the low academic-ability students had the highest mean gain. This finding agreed with Yu et al. (2010) and Han et al. (2014), who reported that low academic-ability students performed better than high academic-ability students using non-traditional approaches. This result may be attributed to low academic-ability students learning in a social context through cooperation and collaboration, which is in line with Gambari et al. (2013) who determined that low academic-ability students performed on a par with high academic-ability students, if not better, in a cooperative learning setting. ...
Purpose – The integrated science, technology, engineering and mathematics (STEM) education has been reported to improve students’ science achievement. Nevertheless, few studies have focused on how this approach affected different ability groups. Lack of equity or the presence of achievement gap can be detrimental because they can reduce medium and low-ability students’ interest in science, which in turn can affect national development. Thus, the purpose of this study is to determine the main and the interaction effects of integrated science, technology, engineering and mathematics (ISTEMA) on students’ science achievement and how this approach affects students with different academic abilities. Methodology – The research adopted a 2x3 factorial design. The sample size consisted of 100 Nigerian science students from Year 11. A total of 51 students with different academic abilities (low, medium and high) were assigned randomly to an experimental group. The experimental group was taught genetics using a fivephased iterative ISTEMA process. Pre-test and post-test data were collected using 40 multiple-choice questions adopted from a national high-stakes examination. Analysis of covariance, paired sample t-test and one-way analysis of variance were utilised in the data analysis. Findings – Findings for research question one revealed a main significant difference in science achievement between year 11 students who learned using ISTEMA and those using traditional methods. No significant interaction effect was observed between the instructional approach and students’ academic abilities, that is, students’ academic abilities and the instructional approach did not interact to enhance students’ achievement. The findings for research question two indicated that high, medium and low academic-ability students benefitted; however, students with low academic abilities had the highest mean gain. Significance – Findings in this study have revealed empirically that the ISTEMA, as an instructional approach, has the potential to close the academic achievement gap. The findings may also serve as a guide for policymakers to promote STEM education in schools.
... This has become a critical challenge to be implanted in schools. Han, Capraro, & Capraro, (2015)'s research results imply that learning science in schools is beneficial for the development of early childhood in absorbing a variety of knowledge that will support their lives in the future. science learning is important for children one of which is to teach children about their bodies, and how to respect and care for the body, should be a top priority for early childhood teachers. ...
The knowledge of the science of human body parts for early childhood is very important so that children have the ability to recognize and support the cleanliness and health of members of the body, as well as so that they recognize their identity. In addition, introducing environmentally friendly material for early childhood teachers to enrich learning media. This study aims to improve student learning outcomes in science using environmentally friendly media. The topic raised in this search was about recognizing body parts and their benefits and treatments. This type of research is action research. Respondents involved 19 early childhood students. The results showed that there was an increase in subjects' understanding of swallowing extremities and treatment 60% in the pre-cycle phase, 80% in the first cycle and 93% in the second cycle. The findings show that the use of happy body media has a positive effect on limb recognition. Further research is recommended on environmentally friendly media and ways of introducing limbs to early childhood through media or strategies suitable for the millennial era. Keywords: Media (Happy Body), Early Childhood Science, Human Body Parts References: Anagnou, E., & Fragoulis, I. (2014). The contribution of mentoring and action research to teachers’ professional development in the context of informal learning. Review of European Studies, 6(1), 133–142. Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal development, and reproductive strategy: An evolutionary theory of socialization. Child Development, 62(4), 647. Black, M. M., & Hurley, K. M. (2016). Early child development programmes: further evidence for action. The Lancet Global Health, 4(8), e505–e506. Blok, H., Fukkink, R., Gebhardt, E., & Leseman, P. (2005). The relevance of delivery mode and other programme characteristics for the effectiveness of early childhood intervention. International Journal of Behavioral Development, 29(1), 35–47. Borg, F., Winberg, M., & Vinterek, M. (2017). Children’s Learning for a Sustainable Society: Influences from Home and Preschool. Education Inquiry, 8(2), 151–172. Borg, F., Winberg, T. M., & Vinterek, M. (2019). Preschool children’s knowledge about the environmental impact of various modes of transport. Early Child Development and Care, 189(3), 376–391. Buchsbaum, D., Bridgers, S., Weisberg, D. S., &, & Gopnik, A. (2012). The power of possibility: Causal learning, counterfactual reasoning, and pretend play. Philosophical Transactions of the Royal Society. Biological Sciences, 367(1599), 2202–2212. Burdette, H. L., & Whitaker, R. C. (2005). Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Archives of Pediatrics & Adolescent Medicine, 159(1), 46–50. Bustamante, A. S., White, L. J., & Greenfield, D. B. (2018). Approaches to learning and science education in Head Start: Examining bidirectionality. Early Childhood Research Quarterly, 44, 34–42. Carr, W. (2006). Philosophy, methodology and action research. Journal of Philosophy of Education, 40(4), 421–435. Colker, L. J. (2008). Twelve characteristics of effective early childhood teachers. YC Young Children, 63(2). Cook, C., Goodman, N. D., & Schulz, L. E. (2011). Where science starts: Spontaneous experiments in preschoolers’ exploratory play. Cognition, 120(3), 341– 349. Dewi Kurnia, H. Z. (2017). Pentingnya Media Pembelajaran. Jurnal Pendidikan Anak Usia Dini, 1 No.1, 81–96. Gelman, R., & Brenneman, K. (2004). Science learning pathways for young children. Early Childhood Research Quarterly, 19(1), 150–158. Gersick, C. J. (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31(1), 9–41. Gopnik, A., Meltzoff, A. N., & Kuhl, P. K. (1999). The scientist in the crib: Mind, brains, and how children learn. New York, NY: William Morrow & Company. Guo, Y., Wang, S., Hall, A. H., Breit-Smith, A., & Busch, J. (2016). The Effects of Science Instruction on Young Children’s Vocabulary Learning: A Research Synthesis. Early Childhood Education Journal, 44(4), 359–367. Hadders-Algra, M. (2019). Interactive media use and early childhood development. Jornal de Pediatria, (xx), 1–3. Han, S., Capraro, R., & Capraro, M. M. (2015). How Science, Technology, Engineering, and Mathematics (Stem) Project-Based Learning (Pbl) Affects High, Middle, and Low Achievers Differently: the Impact of Student Factors on Achievement. International Journal of Science and Mathematics Education, 13(5), 1089–1113. Harris, P. L., & Kavanaugh, R. D. (1993). Young children’s understanding of pretense. Monographs of the Society for Research in Child Development, 58(1), 1–92. Hayati, H. S., Myrnawati, C. H., & Asmawi, M. (2017). Effect of Traditional Games, Learning Motivation And Learning Style On Childhoods Gross Motor Skills. International Journal of Education and Research, 5(7). Hedefalk, M., Almqvist, J., & Östman, L. (2015). Education for sustainable development in early childhood education: a review of the research literature. Environmental Education Research, 21(7), 975–990. Herakleioti, E., & Pantidos, P. (2016). The Contribution of the Human Body in Young Children’s Explanations About Shadow Formation. Research in Science Education, 46(1), 21–42. İlin, G., Kutlu, Ö., & Kutluay, A. (2013). An Action Research: Using Videos for Teaching Grammar in an ESP Class. Procedia - Social and Behavioral Sciences. Jennifer M. Zosh, Emily J. Hopkins, Hanne Jensen, Claire Liu, Dave Neale, Kathy Hirsh-Pasek, S. L. S. and D. W. (2017). Learning through play : a review of the evidence. Kagan, J., Reznick, J. S., & Snidman, N. (1987). The physiology and psychology of behavioral inhibition in children. Child Development, 1459–1473. Kemmis, S., & Taggart, M. (2002). The action research planner. Victoria: Dearcin University Press. Lebel, C., & Beaulieu, C. (2011). Longitudinal development of human brain wiring continues from childhood into adulthood. Journal of Neuroscience, 31(30), 10937–10947. Luna, B., Garver, K. E., Urban, T. A., Lazar, N. A., & Sweeney, J. A. (2004). Maturation of cognitive processes from late childhood to adulthood. Child Development, 75(5), 1357–1372. Nayfeld, I., Brenneman, K., & Gelman, R. (2011). Science in the classroom: Finding a balance between autonomous exploration and teacher-led instruction in preschool settings. Early Education & Development, 22(6), 970–988. Nitecki, E., & Chung, M.-H. (2016). Play as Place: A Safe Space for Young Children to Learn about the World. Nternational Journal of Early Childhood Environmental Education, 4(1), 26–32. Olgan, R. (2015). Influences on Turkish early childhood teachers’ science teaching practices and the science content covered in the early years. Early Child Development and Care, 185(6), 926-942. Ramani, G. B. (2012). Influence of a Playful, Child-Directed Context on Preschool Children’s Peer Cooperation. New York: Merrill-Palmer Quarterly. Ravanis, K. (2017). Early childhood science education: State of the art and perspectives. Journal of Baltic Science Education, 16(3), 284–288. Russo-Johnson C, Troseth G, Duncan C, M. A. (2017). All tapped out: touchscreen interactivity and young children’s word learning. Front Psychology, 8. Schulz, L. E., & Bonawitz, E. B. (2007). Serious fun: Preschoolers engage in more exploratory play when evidence is confounde. Developmental Psycholog, 43(4), 1045–1050. Serpell, R., & Marfo, K. (2014). Some growth points in African child development research. New Directions for Child and Adolescent Development, 146, 97–112. Vouloumanos, A., & Werker, J. F. (2007). Listening to language at birth: evidence for a bias for speech in neonates. Developmental Science, 10(2), 59–64. Weisberg, D. S., & Gopnik, A. (2013). Pretense, counterfactuals, and Bayesian causal models: Why what is not real really matters. Cognitive Science, 37(7), 1368–1381. Winthrop, R., & Mcgivney, E. (2016). Skills for a Changing World: Advancing Quality Learning for Vibrant Societies.Brookings: Center for Universal Education. Zaman, B., & Eliyawati, C. (2010). Media Pembelajaran Anak Usia Dini. Bandung: Universitas Pendidikan Indonesia.
... On an average across OECD countries, one in ten disadvantaged students was able to score in the top quarter of reading performance in their respective countries (known as academic resilience), indicating that disadvantage is not destiny" (OECD, 2019a). On the other hand, Han, Capraro, and Capraro (2015) found that economically disadvantaged learners, on average, performed lower than their counterparts. They also found a statistically significant negative correlation coefficient between being economically disadvantaged and reading proficiency, r = -.145. ...
Despite the widespread use of social educational tools during the last decade in K12 education, the growth curve of utilizing these tools to enhance reading activities has remained lower than expected in learning environments. However, the rapid shift from face-to-face education to online education brought about by the pandemic has attracted educators who utilize these technologies to enhance reading scores among students. Drawing on data from Programme for International Student Assessment 2018, prediction models were conducted to investigate the effects of social reading activity tools on reading proficiency scores while controlling for other factors. The results indicate that reading emails, involvement in online chat (e.g., WhatsApp), reading online news, searching for information online to learn about particular topics and searching for practical information online (e.g., schedules) are significant predictors of higher reading proficiency scores. On the other hand, taking part in online group discussions or forums is a negatively significant predictor of reading proficiency scores. Moreover, reading proficiency scores differ depending on the types of schools students attended, their gender and their social, cultural and economic status when interaction was introduced in the model. These results should help educators as well as researchers to strategically utilize social reading activities according to the nature of the tasks they assign to students.
... Inti dari pembelajaran berbasis proyek adalah untuk menghubungkan pengalaman siswa dengan kegiatan pembelajaran di sekolah dan untuk memfokuskan pemikiran yang lebih serius saat siswa memperoleh pengetahuan baru.Pembelajaran berbasis proyek memiliki beberapa keunggulan yang membuatnya menonjol di antara metode pembelajaran lainnya. Karena mendukung pembelajaran mandiri dan terstruktur (Han et al., 2015;Lee et al., 2019), mampu peningkatan prestasi akademik (Demircioǧlu et al., 2010) ...
Full-text available
Metode pembelajaran konvensional tidak memberikan kesempatan kepada siswa untuk berpartisipasi aktif pada proses pembelajaran. Hal ini mengakibatkan hasil belajar matematika siswa berada pada kategori rendah. Ketidakmampuan siswa dalam mengingat materi yang telah diajarkan oleh guru menjadi faktor utama rendahnya hasil belajar, khususnya di di SMP YP-PGRI 4 Makassar. Salah satu solusi untuk meminimalkan permasalahan siswa adalah melalui penerapan Project Based Learning (PjBL). Oleh karena itu, tujuan penelitian ini adalah untuk mengetahui apakah hasil belajar matematika yang diajar melalui Model Project based Learning PjBL lebih baik dibandingkan dengan siswa yang diajar melalui model pembelajaran Konvensional. Penelitian ini termasuk penelitian eksperimen dengan metode komparasi. Hasil yang ditemukan adalah rata-rata hasil belajar matematika siswa yang diajar dengan menggunakan model PjBL lebih tinggi dibandingkan dengan rata-rata hasil belajar matematika siswa yang diajar dengan menggunakan model pembelajaran Konvensional. Oleh karena itu, Model PjBL menjadi rekomendasi untuk pembelajaran saat ini
... Local and international researchers have highlighted several factors believed to contribute towards learners' low performance in mathematics (Howie, 2003;Mji & Makgato, 2006;Spaull, 2013;Pournara et al., 2015;Beilock, 2015;Han & Capraro, 2014). In the succeeding paragraphs, some of these factors are discussed, coupled with the government's response to these concerns as indicated in the Action Plan to 2019 (Department of Basic Education_a, 2015). ...
... Inti dari pembelajaran berbasis proyek adalah untuk menghubungkan pengalaman siswa dengan kegiatan pembelajaran di sekolah dan untuk memfokuskan pemikiran yang lebih serius saat siswa memperoleh pengetahuan baru.Pembelajaran berbasis proyek memiliki beberapa keunggulan yang membuatnya menonjol di antara metode pembelajaran lainnya. Karena mendukung pembelajaran mandiri dan terstruktur (Han et al., 2015;Lee et al., 2019), mampu peningkatan prestasi akademik (Demircioǧlu et al., 2010) ...
Full-text available
Metode pembelajaran konvensional tidak memberikan kesempatan kepada siswa untuk berpartisipasi aktif pada proses pembelajaran. Hal ini mengakibatkan hasil belajar matematika siswa berada pada kategori rendah. Ketidakmampuan siswa dalam mengingat materi yang telah diajarkan oleh guru menjadi faktor utama rendahnya hasil belajar, khususnya di di SMP YP-PGRI 4 Makassar. Salah satu solusi untuk meminimalkan permasalahan siswa adalah melalui penerapan Project Based Learning (PjBL). Oleh karena itu, tujuan penelitian ini adalah untuk mengetahui apakah hasil belajar matematika yang diajar melalui Model Project based Learning PjBL lebih baik dibandingkan dengan siswa yang diajar melalui model pembelajaran Konvensional. Penelitian ini termasuk penelitian eksperimen dengan metode komparasi. Hasil yang ditemukan adalah rata-rata hasil belajar matematika siswa yang diajar dengan menggunakan model PjBL lebih tinggi dibandingkan dengan rata-rata hasil belajar matematika siswa yang diajar dengan menggunakan model pembelajaran Konvensional. Oleh karena itu, Model PjBL menjadi rekomendasi untuk pembelajaran saat ini
... STEM integration also offers students the opportunity to experience real-world situations simultaneously rather than gradually, to be assimilated at a later time [28]. Several studies have found that STEM education can have a positive impact on students' careers [15], students' attitudes towards STEM subjects [27], students' interests [14] and students' outcomes [11] Hence, the goal of STEM education is to enable students to acquire and understand knowledge based on their experience for solving their problems in scientific contexts. To achieve this goal to benefit students, educators need to change their teaching style. ...
Conference Paper
This research investigated the relationship between science content knowledge and mathematics content knowledge in relation to the preservice primary school teachers’ conceptions of STEM education in Indonesia. The survey method was used to collect data about science content knowledge, mathematics content knowledge, and STEM conceptualizations. A total of 139 preservice primary school teachers participated in this study. The results indicate that science and mathematics content knowledge can influence preservice primary school teachers’ conceptions of STEM education. There is no significant relationship between gender and STEM education conceptualization was detected.
Science, technology, engineering, and mathematics (STEM) project refers to experiments conducted to address problems, make improvements, or discover new things in those fields that can be tested using the scientific method. This paper describes a long-term STEM project in which pre-service science teachers (PSSTs) have a chance to discover new application fields of microalgae and use them for bio-efficacy. One of the PSSTs’ projects is described in detail. This STEM project-based activity lasted for 8 weeks. In the results, a product developed by PSSTs is a cling film which has an antioxidant effect owing to microalgae and sodium alginate. Besides, this environmentally friendly product is equivalent to plastic foils used nowadays. The production processes of this cling film will be presented in detail. According to these processes, some implications for improving the product and facilitating the process will be presented at the end of the paper.
STEM education has taken on high importance in Hong Kong K-12 education landscape. Despite policy advocacy and curriculum endeavour, the quality of STEM education varies significantly between schools. Research literature indicates that high-quality STEM education requires teachers’ rigorous delivery of topics and appropriate pedagogies, and one approach to improve such practices is teacher professional development (TPD). However, because current research on TPD has not given explicit consideration to the complex nature of STEM education, there remains a lack of a clear blueprint of how TPD should be conducted to build teachers’ capacity for STEM education effectively. This paper presents a case study that explores the necessary attributes and identifies the missing links of STEM education TPD by understanding how various TPD models supported a Hong Kong K-12 school embracing STEM education. Qualitative data collection methods, including semi-structured interviews and classroom observations, were employed to draw a picture of TPD implementations in the selected school. The findings suggest that, at the macro-level, effective STEM TPD should not stop at employing mixed use of TPD models; the models have to be integrated organically with respect to a school-based STEM curriculum implementation approach. A collaborative culture between teachers must be cultivated for effective inter-disciplinary integration. Collaborative action research should also be promoted to develop collective wisdom of STEM pedagogies. At the micro-level, TPACK and cross-disciplinary integration skills need to be focusing areas of STEM TPD. With these guiding principles, some possible strategies for effective STEM education TPD are suggested.
This retrospective study examined the impact of prior mathematics achievement on the relationship between high school mathematics curricula and student postsecondary mathematics performance. The sample ( N = 4,144 from 266 high schools) was partitioned into 3 strata by ACT mathematics scores. Students completing 3 or more years of a commercially developed curriculum, the University of Chicago School Mathematics Project curriculum, or National Science Foundation-funded curriculum comprised the sample. Of interest were comparisons of the difficulty level and grade in their initial and subsequent college mathematics courses, and the number of mathematics courses completed over 8 semesters of college work. In general, high school curriculum was not differentially related to the pattern of mathematics grades that students earned over time or to the difficulty levels of the students' mathematics course-taking patterns. There also was no relationship between high school curricula and the number of college mathematics courses completed.
This second edition of Project-Based Learning (PBL) presents an original approach to Science, Technology, Engineering and Mathematics (STEM) centric PBL. We define PBL as an “ill-defined task with a well-defined outcome,” which is consistent with our engineering design philosophy and the accountability highlighted in a standards-based environment. This model emphasizes a backward design that is initiated by well-defined outcomes, tied to local, state, or national standard that provide teachers with a framework guiding students’ design, solving, or completion of ill-defined tasks. This book was designed for middle and secondary teachers who want to improve engagement and provide contextualized learning for their students. However, the nature and scope of the content covered in the 14 chapters are appropriate for preservice teachers as well as for advanced graduate method courses. New to this edition is revised and expanded coverage of STEM PBL, including implementing STEM PBL with English Language Learners and the use of technology in PBL. The book also includes many new teacher-friendly forms, such as advanced organizers, team contracts for STEM PBL, and rubrics for assessing PBL in a larger format.
This study aimed to develop an interdisciplinary on-line learning project for female senior high school students and to explore their participation process and its learning effectiveness. The topic for the project was 'The creative design of a cup speaker'. The five-stage model comprised preparation, implementation, presentation, evaluation and revision (PIPER).The model was used for the integrated learning of science, technology, engineering, and mathematics (STEM). Throughout the project, the students were able to discuss and share knowledge about their projects via the STEM online platform. The study involved 40 volunteers from a girl's senior high school in Taiwan, grouped into ten teams of six students. Textual analyses, survey questionnaires, and interviews were used to collect data. The findings of the study show that the female students were engaged in the project and were able to combine theory with practice effectively to create cupspeakers according to the five stages of PIPER. In addition, this project created a new opportunity for female Taiwanese senior high school students to experience the joy of engineering design as well as to enhance the effectiveness of the STEM knowledge application. Therefore, the design of interdisciplinary and hands-on projects is seen as an important issue for future curriculum design.
This study compares the achievement of high and low ability eighth-grade students working cooperatively during computer-based instruction. Students were grouped either homogeneously or heterogeneously on ability, and received identical instruction on a fictitious rule-based arithmetic number system. No significant differences in achievement were found between the two grouping methods. However, the mixed ability treatment substantially improved the achievement of the low ability students without an accompanying significant reduction in the achievement of the high ability students. The results indicate that designers and teachers may have little to risk in terms of achievement, but potentially much to gain in socialization and interaction, by cooperative heterogeneous grouping during computer based instruction.
A major hurdle in implementing project-based curricula is that they require simultaneous changes in curriculum, instruction, and assessment practices-changes that are often foreign to the students as well as the teachers. In this article, we share an approach to designing, implementing, and evaluating problem- and project-based curricula that has emerged from a long-term collaboration with teachers. Collectively, we have identified 4 design principles that appear to be especially important: (a) defining learning-appropriate goals that lead to deep understanding; (b) providing scaffolds such as "embedded teaching," "teaching tools," sets of "contrasting cases," and beginning with problem-based learning activities before initiating projects; (c) ensuring multiple opportunities for formative self-assessment and revision; and (d) developing social structures that promote participation and a sense of agency. We first discuss these principles individually and then describe how they have been incorporated into a single project. Finally, we discuss research findings that show positive effects on student learning and that show students' reflections on their year as 5th graders were strongly influenced by their experiences in problem- and project-based activities that followed the design principles.
The purpose of this study was to determine the predictors of student grades in introductory physics courses utilizing problem-based learning (PBL) approach and traditional lecturing. The study employed correlational/predictive methods to investigate and describe/explain relationships of students' physics grades with their expectations, attitudes, epistemological beliefs about physics and physics learning, and demographic variables. The subjects involved in this study were 264 freshmen engineering students (PBL, n = 100; traditional, n = 164) at Dokuz Eylül University (DEU) in Izmir, Turkey. All students were surveyed at the beginning and at the end of the spring 2007 semester using the Maryland Physics Expectations Survey (MPEX) to determine their expectations, attitudes, and epistemological beliefs about physics and physics learning. Students' physics learning was measured via their end of semester physics grades. Correlational analyses indicated significant relationships between variables of the study. Forward stepwise linear regression analyses revealed the effort cluster and selected background variables (e.g., gender) as significant predictors of physics grades. Results suggest that further study is needed to investigate predictors and correlates of students' physics learning using qualitative measures to support and more clearly interpret the numerical findings.
Drawing from the 1990, 1996, and 2000 National Assessment of Educational Progress, this study examines Black-White disparities in 4th-, 8th-, and 12th-grade mathematics achievement and instruction. Substantial Black-White achievement gaps were identified, such as 12th-grade Black students scoring below 8th-grade White students. Furthermore, an analysis of race and SES together in the 1996 data revealed that student SES failed to account for much of the Black-White achievement gaps. Several instruction-related factors were also found to differ by race even after controlling for students' SES. This study provides evidence that, despite current reforms promoting high-quality mathematics education for all, Black students of both low and high SES are being left behind.