Available via license: CC BY-NC 4.0
Content may be subject to copyright.
Does ABRACADABRA Help Improve the
English Reading Ability of Chinese
Elementary School Students? A Quasi-
Natural Experimental Study
Hui Gu,1 Jijun Yao,1 Ping Bai,2 Longjun Zhou,3
Alan C.K. Cheung,4 Philip C. Abrami5
1. Nanjing Normal University, Nanjing 210024, Jiangsu, China
2. Huajie Education Group, Lianyungang 222100, Jiangsu, China
3. Jiangsu Second Normal University, Nanjing 210013, Jiangsu, China
4. The Chinese University of Hong Kong, Hong Kong 999077, China
5. Concordia University, Montreal, Québec, H3H 2S2, Canada
Abstract: ABRACADABRA (abbreviated as ABRA) is a software
developed by Concordia University in Canada that focuses on natu-
ral spelling and aims to improve English reading ability. This re-
search is based on 129 first graders, 213 second graders, and 275
third graders in the elementary school of Lianyungang Ganyu
Huajie Bilingual School. We carried out a one-semester pre-and
post-test and quasi-natural experimental research design to explore
the effect of ABRA on students of different grades. The study results
showed that ABRA improved students’ abilities in all grades to var-
ying degrees, but the impact of the first graders got the most signif-
icant. The results of classroom observations and interviews with
teachers showed that teachers needed to apply systematic teaching
strategies and the control of class attention play a key role in it. To
improve students’ English ability, teachers need to effectively im-
prove their ability to apply information technology, especially in
English class. Particularly in low-grade classrooms, attention
should be paid to the management of class discipline to maintain its
efficiency.
Science Insights Education Frontiers 2021; 9(2):1221-1240.
Doi: 10.15354/sief.21.or041
How to Cite: Gu, H., Yao, J., Bai, P., Zhou, L., Cheung, A.C.K., & Abrami,
P.C. (2021). Does ABRACADABRA help improve the English reading ability of
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1222
Chinese elementary school students? A quasi-natural experimental study. Sci-
ence Insights Education Frontiers, 9(2):1221-1240.
Keywords: English Learning, E-Learning Software (ABRA), English Teaching, El-
ementary School, Experimental Research
About the Authors: Hui Gu, School of Educational Science, Nanjing Normal University, Nanjing 210024, Jiangsu,
China. E-mail: gu_nnu@163.com
Ping Bai, Huajie Education Group, Lianyungang 222100, Jiangsu, China. E-mail: 2691285308@qq.com
Longjun Zhou, School of Educational Science, Jiangsu Second Normal University, Nanjing 210013, Jiangsu, China.
E-mail: 294437034@qq.com
Alan C.K. Cheung, Department of Educational Administration and Policy, The Chinese University of Hong Kong,
Hong Kong 999077, China. E-mail: alancheung@cuhk.edu.hk
Philip C. Abrami, Centre for the Study of Learning and Performance, Concordia University, Montreal, Québec,
H3H 2S2, Canada. E-mail: philipc.abrami@concordia.ca
Correspondence to: Jijun Yao, School of Educational Science, Nanjing Normal University, Nanjing 210024, Jiang-
su, China. E-mail: yaojijun_njnu@163.com
Funding: This study was supported by Nanjing Normal University’s horizontal project “High-quality school con-
struction based on evidence-based education research” (S11060A41805). This article was also funded by the Pri-
ority Academic Program Development of Jiangsu Higher Education Institution (PAPD).
Conflict of Interests: None.
© 2021 Insights Publisher. All rights reserved.
Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Crea-
tive Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-
nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided
the original work is attributed by the Insights Publisher.
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1223
Introduction
EGARDING the formation and development of children’s second language,
there have been studies from the perspectives of linguistics and sociolinguistics
explaining the related factors and stages. The mother tongue transfer theory
believed that mastery of the mother tongue would affect the second language learners’
writing discourse organization, writing process and order, and language cognitive abil-
ity through the mother tongue’s cultural thinking, knowledge, or ability (Wang & Wen,
2004). However, the effect varies with the degree of grammatical similarity between the
mother tongue and the second language, especially the differences in pronunciation,
vocabulary, and grammar. Chinese has a negative transfer effect on English learners (Su,
2000). Investigation and research have also found that the native language ability of
Chinese college students plays a restrictive role in the transfer of English writing ability
(Wang & Wen, 2004). In response to this, researchers start from the level of cognitive
psychology and propose the steps of attention, memory, thinking, information pro-
cessing, mother tongue transfer, and interlanguage formation in the process of language
learning, taking the processing of interlanguage in the process of language development
as the key (Wu, 2000).
Paivio’s dual coding theory believes that a single sensation can only store a
limited amount of information. The combination of visual and auditory stimuli can be
an essential means for children to retain information (Mak et al., 2017). E-learning
software fits the feature precisely. In recent years, the impact of educational information
technology on the learning process and results and behavior of teachers and students,
has received increasing attention. Many studies have affirmed the role of e-learning
software in optimizing the educational process, increasing student engagement in learn-
ing, and enhancing teachers’ teaching capabilities (Chen & Zhao, 2017; Gu et al., 2016;
Li, 2016). It is its advantage that e-learning software can actively stimulate attention
and memory with dynamic images and sound interaction (Fan, 2014). Primarily based
on games’ natural integration and mobile characteristics, e-learning software can signif-
icantly improve students’ learning interest and engagement, which is particularly effec-
tive in English learning. Integrating the effects of e-learning software on students,
Chueng & Slavin’s meta-analysis results also found that information technology has
positive impact on students (d = 0.16), implying the great potential of e-learning soft-
ware in students’ language learning (Cheung & Slavin, 2012).
Many studies with electronic teaching software were carried out in class to find
out the effect on students’ abilities. Among them, ABRACADABRA (abbreviated as
“ABRA”) software has achieved remarkable results. Its full name is “A Balanced Read-
ing Approach for Children Always Designed to Achieve Best Results for All.” Wood
evaluated 20 types of English learning software in terms of shape, letter knowledge,
phoneme awareness, word-sound relationship, decoding ability, reading fluency, text
comprehension, etc., and believes that ABRA has a better overall performance (Wood
et al., 2017). So far, ABRA software has carried out fruitful experiments in Canada,
Australia, Hong Kong, China, Hunan and Nanjing1, and other regions and countries
R
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1224
(Comaskey et al., 2009; Deault et al., 2009; Di Stasio et al., 2012; Wolgemuth et al.,
2014; Mak et al., 2017; Wang & Wen, 2004). Abrami integrated experiments from dif-
ferent countries and territories. It did not conduct a country-based subgroup analysis of
the included literature based on its unity of experiment organization and implementation
and the consistency of comprehensive sample results. However, the results in the ability
variables show that there are potential factors that affect the differences in each experi-
ment (Abrami et al., 2015). At the same time, existing studies have found significant
heterogeneity in the experimental results of electronic information technology in differ-
ent countries and regions (Cheung & Slavin, 2012). This research was based on the co-
operation of Nanjing Normal University, Jiangsu Second Normal University, the Chi-
nese University of Hong Kong, and Concordia University in Canada. A survey was
conducted on the Lianyungang Huajie Bilingual School, a private school with relatively
abundant development space in China's more developed provinces. During the imple-
mentation, ABRA members of Concordia University in Canada introduced the design
concept of ABRA software and the experience of implementation and teacher training
in Canada, Australia and Kenya. Members of the Chinese University of Hong Kong
introduced the implementation of ABRA software in Hong Kong and the focus of ex-
perimental research. Nanjing Normal University and Jiangsu Second Normal University
mainly integrated the experiences of all parties to investigate the school and design ex-
perimental programs. This study aimed to explore another feasible path for the applica-
tion of ABRA software in the local Chinese context.
Literature Review and Questioning
ABRA software is web software developed by Centre for the Study of Learning and
Performance (CSLP) of Concordia University in Canada to improve student’s English
reading ability. It won the United Nations Best Learning Software Award in 2016.
ABRA software is based on the six core skills of English reading ability proposed by
the National Reading Panel (NRP): (1) Phoneme recognition skills: students can associ-
ate and correspond to the pronunciation of English letters and words; (2) Phoneme
awareness: the ability of students to distinguish and colloquially extract single syllables
from words; (3) Fluency: students can read English text quickly and expressively; (4)
Reading comprehension: understanding and retelling text content; (5) Listening com-
prehension: understanding and retelling spoken language content; and (6) Vocabulary
knowledge: recognize and understand the meaning of spoken words and text words.
According to the activity needs of teachers, students, and parents, it sets up three main
modules to guide the direction of activities of different groups (Abrami et al., 2015).
Student module activities are the main content, including 17 “learning in doing” games.
It takes natural phonics as the core, sets up games of different difficulty levels, and oth-
er activities related to reading fluency, reading comprehension, and writing: the parent
module and teacher module difficulty levels. Among them, the content of the parent
module mainly includes reading suggestions and reading materials for students at home;
the content of the teacher module primarily consists of the teacher’s teaching materials
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1225
and teaching guidance. Through the allocation of roles and tasks and the control of
game content, ABRA can help students choose their learning progress according to their
own situation and get timely feedback. Teachers can keep abreast of the situation of
students at any time, and parents can participate in the overall development of student
progress.
The experimental research of ABRA software first focused on the countries
where English is the native language. The subjects of the experiment were young chil-
dren who are just beginning to learn the English language. One of the experimental
studies with the largest sample size is Savage et al., based on 1,067 preschool and ele-
mentary school first- and second-year students in 74 classes in Alberta, Ontario, and
Quebec. The study results found that students in the experimental group generally per-
formed better after one semester of use than schools in the control group. It is particu-
larly prominent in item phoneme combination (p < 0.01), syllable-letter matching (p <
0.01), and fluency of phoneme decomposition (p < 0.01) (Savage et al., 2013). Studies
have also compared the characteristics of students and found that boys in the experi-
mental group are significantly higher in sentence comprehension, paragraph compre-
hension, and total scores than boys in the control group. At the same time, girls do not
show this difference (Abrami et al., 2016).
In some areas with official second language, researchers made some attempts
to help students with different language capabilities improve English language through
ABRA software. Wolgemuth et al. found that after a minimum of 120 minutes of inter-
vention per week, the English reading development of indigenous students gradually
kept pace with the non-indigenous peoples. In addition, indigenous students who have
also used ABRA software are more advanced in terms of word reading and phoneme
awareness. Researchers believe that this is due to the ceiling effect of the students of
indigenous peoples in phoneme learning, which fully demonstrates that ABRA software
can help students with weaker English foundation in the English language environment
(Wolgemuth, 2019). ABRA software has also conducted some research in non-English
speaking countries. The researcher took students with lower social and economic back-
grounds in Kenya as the research objects. They found that after 13 weeks of using
ABRA software. Students were improved in vocabulary and phonetics, as well as Eng-
lish, mathematics, science, and social subjects (Abrami, 2016). The Hong Kong exper-
iment is similar to the previous experiments in English-speaking, second-language par-
allel, and non-English-speaking regions, indicating the beneficial effects of ABRA
software. However, the research of Mak et al. discussed the variables of teacher atti-
tudes and teacher teaching behavior. They believe that teachers’ innovative application
of ABRA software, effective teaching content, and high attitude and interest will help
teachers play their role in software applications and promote students’ performance
(Mak, 2017).
To integrate the effects of ABRA software in different countries and periods,
Abrami et al. conducted a meta-analysis of 11 related experiments from Australia, Can-
ada, and China. They found that the overall effect size was 0.179. Among the six skills,
improving listening comprehension (g=0.381) and phoneme awareness (g=0.324) has
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1226
the most significant effect. At the same time, Abrami believes that the organization and
implementation of these experiments are homogeneous, and therefore did not sub-
analyze the reasons for the differences in performance of these studies (Abrami, 2015).
Based on the above research, it can be seen that ABRA software has a particular effect
on the improvement of students’ abilities, but the degree could change according to re-
lated factors.
Most existing research explores the software implementation process, methods,
and results combined with teachers’ teaching methods and strategies. Although the stu-
dents’ gender, economic background, and other factors are discussed, there is a lack of
comparison. At the same time, according to the critical period hypothesis of second lan-
guage learning, students have a certain critical period for second language learning. Still,
researchers do not have a consistent conclusion on the boundary of the critical period
(Snow & Hoefnagel-Höhle, 1978). Therefore, it is necessary to discuss the English
learning progress of students of different grades. At the same time, in the existing re-
search in China, there is a lack of experimental research on electronic teaching software
to improve student’s English ability (or second language learning ability). Based on
American positivism research tradition and methodology, evidence-based research has
gradually become the mainstream of its educational research paradigm. This paradigm
emphasizes the design of the quantitative analysis, experimental research, and discus-
sion of causality and related relationships. It aims to seek the best evidence and provide
specific practical approaches for school reform.
In the few existing experimental studies, due to the lack of standardization of
experimental design, insufficient scientific measurement tools, and improper data pro-
cessing methods, the scientificity and robustness of the conclusions are difficult to
guarantee (Lei, 2007).
In response to this, this study used ABRA software to carry out experiments. In
comparison, the software had undergone the scientific and rigorous software tools test-
ed by previous experiments. Its framework design, content arrangement, and testing
tools had been fully demonstrated and continuously improved in the conclusions. At
present, preliminary research has been carried out in many countries and regions around
the world. Therefore, the study will take Ganyu Huajie Bilingual School in Lianyun-
gang City, Jiangsu Province, as an example, focusing on students’ performance in grad-
ers of one, two, and three and different teachers’ teaching strategies. We used ABRA
software to explore the path and best plan for computer technology to assist students in
improving their English proficiency in the Chinese context.
Research Design and Methodology
To explore the improvement of the English ability of elementary school students by the
ABRA project, this study adopted a pre-test and post-test quasi-experimental research
design. The study conducted a mixed research design and didn’t change the existing
class framework to implement the experiment. Through qualitative research, random
classroom observations were conducted on the teachers of the experimental group and
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1227
the control group. After the experimental intervention, the three experimental group
teachers and the English group lesson preparation directors were interviewed to under-
stand the teacher’s feelings, attitudes, and other influencing factors to explore the exper-
imental results and the causes and mechanisms of differences between the groups. The
intervention period was the second semester of the 2018-2019 school year, from Febru-
ary 26, 2019, to June 14, 2019, which lasted 16 weeks.
Subjects
The research team included 956 students in 24 classes from Grade 1 to Grade 3 of
Ganyu Huajie Bilingual School in Lianyungang City, Jiangsu Province. In class group-
ing, to avoid the “observer effect” from affecting the students in the control group, none
of the teachers in the experimental class participated in the teaching process of the con-
trol class. The study included four classes in the first grade, three classes in the second
and third grades, with a total of 398 people as the experimental group. Conversely, four
classes were included in the first grade and four in the second and third grades, with 558
students as the control group. The conditions of the 23 classes participating in the ex-
periment were the same, and there were no significant differences in grades, family
background, and gender structure.
To ensure the accuracy of the calculation of experimental results, this study
first deleted the invalid pre-test and post-test questionnaires and obtained 726 valid pre-
test papers and 772 post-test questionnaires. The effective questionnaire rates were
81.06% and 75.94%. Considering the rationality of the arrangement of testers during the
experimental test and the adaptability of elementary school students to unfamiliar ques-
tions, the return rate of this questionnaire is still within a reasonable range. Secondly, to
focus on the changes in each student’s performance and avoid the resulting error caused
by the student transferring, dropping out of the experiment, or having only one test
score, student matching was performed on the before and after test scores after invalid
scores were deleted, and finally a total of 614 effective questionnaires were obtained
(Figure 1).
Selection and Training of Teachers in Experimental
Group
Teachers were randomly divided into experimental group or control group. Experi-
mental members follow the school’s system of dividing classes according to student
scores every year to ensure that each type is equally divided. Therefore, no additional
pairing was made between the participating teachers and the class to ensure that the
experiment was natural. The number of teachers in experimental groups and control
groups were shown in Table 1. There was no significant difference in teaching year and
lesson preparation between teachers in experimental group and control group.
In the intervention, teachers’ training was only for teachers in the experimental
group. Trainings were divided into online and offline methods. Online communication
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1228
Figure 1. The Graph of the Number of Students in the Experimental Group and
the Control Group.
focused on real-time answers to questions and sharing teaching materials and experi-
ence; offline discussions focused on possible problems and problem-solving strategies
after the ABRA game was applied in reality. Teacher training was mainly divided into
three stages. The first stage of teacher training started in February 2019. The research
team formally established an online communication channel and entered the school to
meet face-to-face with teachers to communicate about the background of the game and
the design of the game itself. The second training phase began in April 2019, focusing
on the teacher’s solution to operational problems. After observing the classroom, the
teachers of the experimental group discussed and made suggestions on the time alloca-
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1229
Table 1. Comparison of the Class Allocation and the Number of Students be-
tween the Experimental Group and the Control Group.
Experimental Group
Control Group
Teacher
Class #
Total Student #
Teacher
Class #
Total Student #
First
Grade
A
4
151
D
4
149
Second
Grade
B
3
122
E
3
121
F
2
86
Third
Grade
C
3
125
G
3
120
H
2
82
tion, operation, student attention, and systemic aspects of the teaching process. The
third stage of the school training started in May 2019. This stage focused on motivating
teachers and summarizing the situation, avoiding the slack mentality of teachers at the
end of the experimental stage and the pressure of the final exam.
Teacher Intervention
Teachers in the experimental and control groups had the following differences in teach-
ing content, including class organization, and teaching methods. In terms of the content
and teaching method, the control group teachers were required that the class continue to
follow the Elementary School English Curriculum Standard (2011 Edition) and com-
pleted the textbook content per the related courses. The experimental group teachers
were required to teach the Phonics method through the ABRA game, requiring no less
than 80 minutes of intervention time per month. The specific intervention time depend-
ed on the teacher’s teaching rhythm. The experimental group teachers also adopted the
class teaching system in class organization, mainly set up as the teaching link of “teach-
er question-student answer-teacher summary”. In terms of specific operations, the
teachers in the three grades fully considered the different learning stages of the students
in the three grades, and their teaching focuses were different. Based on the English
foundation of the students in each grade, the first-grade teachers focused on using the
natural phonemes in the game to consolidate the writing and pronunciation of the letters.
The second-grade teachers attached great importance to expanding words in games and
improving students’ interest in English learning. The third-grade teachers paid more
attention to the classification, regularity, and way of summarizing comments.
Research Tools
This study used the Group Reading and Assessment and Diagnostic Evaluation
(GRADE) standardized questionnaire developed with Canadian language and education
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1230
measurement experts. It was primarily conducted in the United States, France, India,
Canada, Denmark, and other countries. It has been recognized by international counter-
parts and has a scientific and authoritative standardization test of English proficiency.
Its standardized scope can cover the mastery of English reading skills from preschool
children to adults and aims to help the students improve their English ability in the as-
sessment. The test paper is divided into eight parts, which test the ability of students to
master the same (different) words required for natural spelling, the ability to recognize
rhymes, matching syllables and letters of the first and last letters in a word, letter recog-
nition, and first and last syllable recognition. The questionnaire has a high degree of
internal consistency (0.95-0.99), replica reliability (0.81-0.94), and test-retest reliability
(0.80) (Williams, 2001).
The research team also used unstructured interviews. For example, interviews
were conducted on teachers’ teaching methods, self-rating, ABRA game application
attitude, ABRA software suggestions, etc., to better analyze the results of each grade.
Analysis Method and Power Estimation
To determine the difference between the experimental and control groups after the in-
tervention, this study calculated the effect size to reflect the effect of the ABRA game
on students’ English ability improvement. The effect size calculation formula is:
Among them, d represents the experimental effect, that is, the effect size of the
experimental group minus the effect size of the control group. The effect size of the
experimental group is the difference between the mean of the post-test and the pre-test,
divided by the combined standard deviation. mpost-t and mpre-t represent the mean value of
the experimental group’s post-test and the experimental group’s pre-test, respectively. st
represents the combined standard deviation of the experimental group. The effect size
calculation process of the control group is the same as that of the experimental group.
In order to verify whether the sample size reaches the necessary statistical test
power, we selected the effect size d = +0.20 and statistical test power r2 = 0.80 that
showed that the effect has policy influence. The power test result showed that 80% of
the test power could be guaranteed with a sample size of more than 200 (Wu & Wen,
2011).
Results
In What Works Clearhouse (WWC), a comprehensive website of educational evidence
designed by the Institute of Education Sciences (IES) in the United States, to ensure that
the effect size accurately reflects the impact of experimental measures on students and
to reduce the great difference in students’ cognitive models for the experimental error,
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1231
the effect size of the vast difference between the experimental group and the control
group was set to be less than 0.25 (Clearinghouse, 2020).
The results of the pre-post test difference between the experimental group and
the control group in this study showed that in the overall difference between control
group and experimental group was less than 0.25, which is a slight difference. Thus, it
showed that the difference of pre- test was effectively controlled (Table 2).
English Test Results
The effect on grade one, two and three was different The performance improved a lot in
the first and third-grade students, and the first-grade students benefited the most. The
ABRA game did not significantly affect the second-grade students. The following part
will be based on the specific analysis of the performance of each grade on each test item.
First grade. The first-grade students have made significant progress in all items
in the GRADE test (except for recognizing the same words and recognizing syllables in
the last letter). Among them, the matching of notes at the end of words and recognition
of letters was a significant effect (d=0.83, d=0.80). In addition, considerable progress
has been made in word rhyme (d=0.58), word-initial matching (d=0.73), as well as dif-
ferent word recognition and initial listening recognition (Table 3).
Second grade. After the ABRA game, the second-year students did not achieve
significant progress in each GRADE test item. There was only a tiny effect on the
recognition of different words. Because the statistics were not substantial, it was impos-
sible to reject the null hypothesis; there was no significant difference between the two
groups in different word recognition abilities after the intervention (Table 4).
Third grade. In general, the third-grade students progressed after receiving
GRADE game intervention, but word-initial matching showed insignificant effect size
results. However, it has improved recognizing the same word, different dishes, and let-
ter matching at the end of a word, letter recognition, and syllable recognition of the first
(final) letter. Especially in the word rhyme (d = 0.43), the improvement effect of stu-
dents’ performance was the greatest (Table 5).
Classroom Observations Results
There were significant differences between the introduction of ABRA games and the
traditional classroom teaching mode on teachers and students in the classroom observa-
tion results.
In the course of teachers’ teaching, the experimental group teachers showed at-
tention to pronunciation. They were more active in letter pronunciation, syllables, and
word memory links, and they had fewer professional errors such as misspellings and
mispronunciations in presentations. In the form of teaching, the activities of ABRA
could be used flexibly to stimulate students’ interest in English learning. They could be
combined with books to expand and consolidate words. The teachers in the control
group paid more attention to the interpretation of the meaning and the repeated reading
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1232
Table 2. The Pre-Test Differences between Experimental Group and Control
Group.
Experimental Group
(N=250)
Control Group
(N=364)
Cohen’s d
Test Item
Mean
SD
Mean
SD
Total Score
39.26
7.85
40.74
6.90
0.20
Same Word Recognition
4.84
0.51
4.84
0.56
0.00
Different Word Recognition
3.80
0.62
3.89
0.38
0.19
Word Rhyme
4.85
1.80
4.88
2.02
0.01
Match the First Letter of a Word
3.57
1.65
3.59
1.38
0.01
Match the Last Letter of a Word
3.89
1.69
3.88
1.65
-0.01
Letter Recognition
7.92
2.76
8.63
2.40
0.28
Initial Syllable Identification
5.61
1.85
5.94
1.48
0.20
Last Letter Syllable Discrimination
4.79
1.63
5.11
1.49
0.21
SD: Standard Deviation.
Table 3. Changes in First Grade English Scores.
Test Item
Experimental
Group (N=67)
Control
Group (N=59)
Cohen’s d
Pre-test
Post-test
Pre-test
Post-test
Total Score
30.52
(5.77)
38.15
(7.77)
33.68
(5.37)
36.05
(8.05)
0.77***
Same Word Recognition
4.78
(0.67)
4.55
(1.05)
4.8
(0.66)
4.66
(0.91)
-0.08
Different Word Recognition
3.61
(0.81)
3.45
(1.03)
3.9
(0.3)
3.64
(0.86)
0.22
Word Rhyme
3.99
(1.66)
5.55
(2.05)
4.53
(1.67)
5.05
(2.27)
0.58**
Match the First Letter of a Word
2.66
(1.37)
2.96
(1.21)
2.81
(0.98)
2.31
(1.05)
0.73***
Match the Last Letter of a Word
2.79 (
1.54)
3.31
(1.41)
2.93
(1.42)
2.32
(1.28)
0.80***
Letter Recognition
5.39
1.95)
7.3
(2.46)
6.2
(2.33)
6.27
(2.67)
0.83***
Initial Syllable Identification
3.67
(1.68)
6.01
(1.71)
4.39
(1.5)
6.19
(1.95)
0.35
Last Letter Syllable Discrimination
3.64
(1.44)
4.82
(1.65)
4.12
(1.39)
5.42
(1.82)
-0.05
Note: The asterisk * indicates the result in the independent sample t test, *p < 0.1,**p < 0.05, ***p < 0.01.
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1233
Table 4. Changes in Second Grade English Scores.
Test Item
Experimental
Group (N=77)
Control
Group (N=136)
Cohen’s d
Pre-test
Post-test
Pre-test
Post-test
Total Score
40.94
(5.87)
46.96
(8.96)
38.04
(5.26)
49.53
(5.16)
-1.41***
Same Word Recognition
4.83
(0.47)
4.78
(0.75)
4.84
(0.53)
4.89
(0.34)
-0.2
Different Word Recognition
3.86
(0.53)
3.91
(0.4)
3.84
(0.49)
3.83
(0.54)
0.13***
Word Rhyme
5.09
(1.56)
8.97
(3.01)
4.2
(1.89)
8.31
(2.01)
-0.49***
Match the First Letter of a Word
4.58
(1.6)
3.68
(1.2)
3.45
(1.3)
3.81
(1.15)
-0.94***
Match the Last Letter of a Word
4.32
(1.52)
4.62
(1.34)
3.34
(1.53)
5.18
(1.02)
-1.22***
Letter Recognition
6.73
(1.92)
7.88
(2.53)
7.45
(1.83)
8.68
(1.94)
-0.14***
Initial Syllable Identification
6.22
(1.15)
6.94
(1.13)
5.92
(1.45)
7.43
(0.7)
-0.7
Last Letter Syllable Discrimination
5.3
(1.49)
5.61
(1.5)
5.01
(1.49)
6.63
(1.07)
-1.03***
Note: The asterisk * indicates the result in the independent sample t test, *p < 0.1,**p < 0.05, ***p < 0.01.
Table 5. Changes in Third Grade English Scores.
Test Item
Experimental
Group (N=106)
Control
Group (N=169)
Cohen’s d
Pre-test
Post-test
Pre-test
Post-test
Total Score
43.58
(5.5)
48.53
(6.0)
45.37
(5.07)
48.6
(5.4)
0.24**
Same Word Recognition
4.88
(0.43)
4.93
(0.25)
4.85
(0.54)
4.83
(0.59)
0.19
Different Word Recognition
3.88
(0.51)
3.84
(0.58)
3.93
(0.29)
3.83
(0.63)
0.15
Word Rhyme
5.23
(1.87)
8.18
(2.36)
5.54
(2.02)
7.64
(2.32)
0.43***
Match the First Letter of a Word
3.42
(1.45)
3.43
(1.03)
3.96
(1.43)
3.7
(1.14)
0.22
Match the Last Letter of a Word
4.26
(1.59)
4.08
(1.41)
4.64
(1.46)
4.34
(1.31)
0.09
Letter Recognition
10.38
(1.34)
10
(1.28)
10.42
(1.22)
10.09
(1.08)
0.00
Initial Syllable Identification
6.4
(1.44)
7.36
(1.06)
6.49
(1.06)
7.22
(1.01)
0.05
Last Letter Syllable Discrimination
5.14
(1.48)
6.08
(1.4)
5.53
(1.35)
6.41
(1.07)
-0.07
Note: The asterisk * indicates the result in the independent sample t test, *p
＜
0.1,**p
＜
0.05, ***p
＜
0.01.
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1234
of words and texts. They did not pay much attention to the role of activities in English
learning.
In terms of student performance, the students in the experimental group showed
more enthusiasm for English learning, were willing to improve their English ability in
activities, and showed extremely high sensitivity in word pronunciation and syllable
reflection. Compared with the students in the control group, the students in the experi-
mental group were more fluent in reading texts and words; they mastered more quickly
in performance and sentence comprehension; they did not show too much fear in the
face of unfamiliar words and sentences patterns. On the other hand, the experimental
group had some difficulties in class organization. Students were particularly excited
about the activities they were interested in. Therefore, teachers in the experimental
group needed to maintain classroom discipline and focus students’ attention on teaching
knowledge.
Teacher Interview Results
To understand the role of teachers in the use of ABRA games, this study further de-
signed semi-structured interviews. We interviewed the experimental group teacher A,
teacher B, teacher C, and director D of the English teaching and research section in the
first, second, and third grades. To further analyze the teacher’s application of the soft-
ware in preparation, class, after class, and the teacher’s psychological state.
The changes in the psychological state of the teachers were relatively con-
sistent, which was a change from semi-acceptance to acceptance to active integration.
The second-grade teacher B said frankly, “I didn’t teach very carefully at the beginning.
It was not until your second visit that I prepared more seriously and used it more.” The
underlying reason was that teachers were more or less burnout due to heavy tasks. The
first-grade teacher A also said that this was an important reason he could not fully ac-
cept the ABRA game initially. She took half of the first-grade class (4 classes), a class
of about 40 students, and “If a child’s learning foundation is weak and sometimes feels
that it is too late to complete the task of the textbook.” Teacher C worried, “I am afraid
that the grades of our third-grade students will fall in the city’s unified exam.”
After the first stage of use and the second stage of training, teachers generally
responded well. Because I found that “some syllables can also be memorized together
with words in the book” (Teacher A); “Students are particularly interested and feel that
they can teach students interesting English” (Teacher B); and “I am also familiar with
ABRA, unlike It was so difficult at the beginning” (Teacher C). In the third stage,
teachers had their innovative understanding of games. The first-grade teacher A felt that
“using the game of Same Words, sometimes we could review the words we have
learned”, and found that this was not just an English tool that can only be played.
In the interaction with students and student feedback, teachers said that stu-
dents like this software very much. “Sometimes I don’t need to say, when I was about
to open this software, they would shout ‘ABRA time’ together.” (Teacher B) “Because
they liked it so much, they suddenly became noisy and I found it hard to keep the disci-
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1235
pline.” (Teacher B) Moreover, Teacher B also said that sometimes “Students pay all
their attention to the game, and we had no time to explain the knowledge in the games.”
Teacher A also thought that some words are still too difficult for first-grade students
and can only be skipped or introduced roughly. It is difficult for students to remember
and apply after class. Many of them were a bit like left-behind children, because their
parents were too busy at work, and grandparents could take them to play at
home”(Director D) and “the family background is not good, most of the homes do not
own computers” (Teacher A, Teacher B, Teacher C)
In this study, the experimental results of the second grade were not ideal, and
the teacher interviews also deliberately explored the reasons. Teacher B of the second-
year experimental group also expressed his thoughts on the experimental results of the
first semester. “When teachers in the control group were teaching words, they would
consciously use natural phonics to their students.” This showed that the students in the
second-grade control group were also exposed to natural phonics. Although there were
two random classroom observations, teachers who could know in advance usually
wanted to “have something to show” (Director D), so they avoided vocabulary teaching
in the classroom observation class and focused on the reading and performance of the
text. Therefore, the two classroom observations found no signs of using natural spelling
by the teachers in the control group.
Teachers also had some opinions on the improvement of the software, especial-
ly in picture books. In terms of the display form of picture books, “Students are not very
interested in picture books; they just look at the pictures and read. If there are vivid an-
imations, and be shorter, students could learn easier.” (Teacher B) Regarding the diffi-
culty of the picture book, Teacher C believed that the content of the picture book was
too straightforward sometimes but too tricky for the third-grade students, so it was too
difficult to make a correct choice. I hope that the school could build the computer room
as soon as possible, because “if students play by themselves, it will be better than I
teach them how to play. At that time, it will be difficult to manage discipline and teach
one by one” (Teacher B).
Conclusion and Discussion
The study results found that ABRA software teaching positively affects students’ Eng-
lish, especially phoneme awareness. On the one hand, the results confirmed that natural
spelling, as an intermediary for second language acquisition, can decode words to deep-
en students’ vocabulary memory, improving spelling, writing, and reading comprehen-
sion abilities (Peng & Tao, 2009). At the same time, it is consistent with the research
conclusions of teaching natural phonics in the Chinese context, which proves the uni-
versality of natural phonics to improve students’ pronunciation (Jiang, 2016). On the
other hand, the research results support the positive effects of e-learning software on
students after being used in the classroom. Based on the double coding theory, the in-
formation organization form of the combination of audio-visual can retain complete
information so that students can better full reading in animation than static images. At
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1236
the same time, the storage space of each channel is limited. As each type of storage
space has a mutually independent memory system, vision and hearing can become “as-
sistants” to each other’s memory (Mak, 2017).
However, the performance of each grade is not the same. From the point of
view of the progress value of the experimental group itself from the post-test to the pre-
test alone, the progress of grade one to grade three tends to decrease gradually. Studies
have shown that 5-7 years old is a crucial period for children’s word order development,
and students’ native language ability will continue to grow at this stage. The work to be
done in the stage of mother tongue transfer becomes more complicated, and the difficul-
ty of second language learning will increase (Qi, 2009). From the point of view of the
software effect size of the control group and the experimental group, the research re-
sults show that the experimental group of the first grade has made the most progress;
followed by the experimental group of the third grade. The reason may be the first and
second-grade teachers’ excessively active phoneme intervention. In teacher interviews
and reflections on practical training, the research team found that the second-grade con-
trol group teachers understand natural spelling. She has a mentality of “not falling be-
hind” after the teachers of the experimental group started the natural spelling experi-
ment. Therefore, the teaching of “natural spelling” was also increased in the class so
that the experimental group of the second grade did not show more advantages as a
whole. Second, student attention was closely related to second language acquisition
based on electronic software. Compared with students in the first and third grades, se-
cond-year students can quickly learn the games in the classroom after a preliminary
understanding of English. However, since there is no suggestion of complete rule
awareness in the class, it is often difficult for students to maintain teaching discipline in
ABRA games. At the same time, students will not concentrate on learning phoneme
knowledge points but only pay attention to the game links, which affects the second-
grade students’ acquisition of ABRA learning effects (Ding, 2012).
The overall results of students in each item were also consistent with the re-
search results of Jiang’s application of natural phonics to third-grade vocabulary teach-
ing. The students in the experimental group had superior performance in vocabulary
spelling and spelling, phonetic awareness, and vocabulary selection ability (Jiang, 2016).
More specifically, students have improved their letter matching ability and phoneme
recognition ability to a greater extent, which was also consistent with the performance
of Mak’s various topics (Mak, 2017). The difference in the speed of progress in multi-
ple fields is that, on the one hand, the teacher focuses on the listening discrimination of
phonemes according to the level and stage of the students’ natural phonics during the
teaching process. On the other hand, due to identifying initial and ending letters and
matching the beginning and end of words, students need to come up with the word’s
correct spelling in their minds and then look for the correct answer on the test paper.
Students will fail to choose the correct answer because they cannot recall the right
words or recall them promptly. Mak et al. also mentioned that students of second lan-
guage learning need to construct brand-new words in their brains in some questions,
which will increase the difficulty of the questions (Mak, 2017).
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1237
Study Limitations and Perspectives
Based on natural spelling, this research is aimed at Ganyu Huajie Bilingual School in
Lianyungang City, Jiangsu Province, on the ABRACADABRA software. It is found
that teaching based on electronic software can significantly improve students’ English
ability. This proves the applicability of the “quoted” e-teaching software and the natural
spelling teaching method in the local English teaching process and also shows the posi-
tive role of promoting school digital teaching. It is also a reasonable attempt to build a
“college-school partnership”. However, the implementation effect of e-teaching will
vary due to the school’s software/hardware conditions, teachers’ teaching style, and
teachers’ mastery and understanding of e-teaching. Moreover, in this project, due to the
limitations of funds, time and personnel, the experiment failed to fully guarantee the
“non-intervention” behavior of the teachers in the control group. Therefore, although
the research conclusion proves the practical function of natural spelling and ABRA
software, this deficiency affects the robustness of the local findings to a certain extent.
In the future English teaching process, we should first teach natural spelling in
the lower grades. The study results found that the progress of lower-grade students is
more significant than that of upper-grade students. Educators should abandon the preju-
dice of low-grade students’ low learning efficiency and weak adaptability, and they
need to face up to the plasticity of low-grade students. Teachers need to pay attention to
the transfer of knowledge in the game and establish the awareness of students’ rules in
the interaction between teachers and students to enhance students’ interest. The process
of e-learning should be promoted. The research results support the positive impact of
audio-visual multi-dimensional electronic teaching tools on student performance.
Schools should actively arrange funds to upgrade electronic teaching software,
strengthen the training of teachers in the use of electronic teaching, and strive to
strengthen teachers’ teaching levels and promote the diversification of teaching meth-
ods. Teachers should also avoid "playing games for the sake of games" in the teaching
process. It is necessary to clarify teaching objectives, apply teaching methods rationally,
and strengthen systematic and efficient instruction.
The importance of experimental research lies in the ability to provide detailed
experimental procedures and results for evaluation. The accumulation of experimental
research can provide educators with effective and comprehensive teaching strategies,
reducing new programs’ “trial and error” costs. On this basis, comprehensive and sys-
tematic teaching measures can provide principals, teachers, and students with effective
experimental programs in many aspects and help education policymakers’ reference.
This research presents a relatively standard experimental process to promote the devel-
opment of experimental research in the direction of education. Based on the continuous
improvement and advancement of experiments in English learning, more and more veri-
fiable application models of English e-teaching software will be developed in the future.
In the case of meeting the development needs of students, we should pay attention to
the applicability and comprehensiveness of software and electronic applications in local
contexts while adapting to the differences between schools and teachers.
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1238
Notes
1. The experimental reports of Hunan and Nanjing have not yet been published.
References
Abrami, P.C., Wade, C.A., Lysenko, L., Marsh,
J., & Gioko, A. (2016). Using educational
technology to develop early literacy skills in
Sub-Saharan Africa. Education and Infor-
mation Technologies, 21(4):945-964. DOI:
https://doi.org/10.1007/s10639-014-9362-4
Abrami, P., Borohkovski, E., & Lysenko, L.
(2015). The effects of ABRACADABRA on
reading outcomes: A meta-analysis of ap-
plied field research. Journal of Interactive
Learning Research, 26(4):337-367. DOI:
https://www.learntechlib.org/primary/p/1473
96/
Chen, Y., & Zhao, K. (2017). The effect of in-
formation technology on the scientific self-
efficacy of middle school students. Open
Education Research, 23(6):60-70. [Chinese]
DOI:
https://doi.org/10.13966/j.cnki.kfjyyj.2017.0
6.007
Cheung, A.C.K., & Slavin, R.E. (2012). How
features of educational technology applica-
tions affect student reading outcomes: A me-
ta-analysis. Educational Research Review,
7(3):198-215. DOI:
https://doi.org/10.1016/j.edurev.2012.05.002
Clearinghouse, W.W. (January, 2020). What
Works Clearinghouse Standards Handbook,
Version 4.1. Washington, DC: U.S. Depart-
ment of Education, Institute of Education
Sciences, National Center for Education
Evaluation and Regional Assistance.
https://ies.ed.gov/ncee/wwc/Docs/referencer
esources/WWC-Standards-Handbook-v4-1-
508.pdf
Comaskey, E.M., Savage, R.S., & Abrami, P.
(2009). A randomised efficacy study of Web
‐based synthetic and analytic programmes
among disadvantaged urban Kindergarten
children. Journal of Research in Reading,
32(1): 92-108. DOI:
https://doi.org/10.1111/j.1467-
9817.2008.01383.x
Deault, L., Savage, R., & Abrami, P. (2009).
Inattention and response to the ABRACA-
DABRA web-based literacy intervention.
Journal of Research on Educational Effec-
tiveness, 2(3):250-286. DOI:
https://doi.org/10.1080/19345740902979371
Di Stasio, M.R., Savage, R., & Abrami, P.C.
(2012). A follow-up study of the ABRA-
CADABRA web-based literacy intervention
in Grade 1. Journal of Research in Reading,
35(1):69-86. DOI:
https://doi.org/10.1111/j.1467-
9817.2010.01469.x
Ding, J., Pan, F., Wang, Y., & Chen, Y. (2012).
The effect of 9 to 13-year-old pupils’ atten-
tion on academic performance. Medical
Journal of Communications, 26(6):569-
572+579. [Chinese] DOI:
https://doi.org/10.3969/j.issn.1006-
2440.2012.06.016
Fan, L. (2014). Research on the influence of
dynamic geometry software on students’ at-
tention. Dissertation; Liaoning Normal Uni-
versity. [Chinese] DOI:
https://doi.org/10.7666/d.Y2612733
Gu, X., Wang, C., & Wang, F. (2016). Has the
role of information technology happened? A
study on the influence of education infor-
mation. e-Education Research, 37(10):5-13.
[Chinese] DOI:
https://doi.org/10.13811/j.cnki.eer.2016.10.0
01
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1239
Jiang, N. (2016). Research on the Application of
Natural Phonics in Third Grade English Vo-
cabulary Teaching. Dissertation; Shenyang
Normal University. [Chinese]
http://cdmd.cnki.com.cn/article/cdmd-
10166-1016112890.htm
Lei, Y. Research on the effect of computer-
assisted teaching software on pupils' memo-
rizing English words under the network en-
vironment. Inner Mongolia Normal Univer-
sity, 2007. [Chinese] DOI:
https://doi.org/10.7666/d.y1173539
Li, L. (2016). Research on the influence of ICT
application ability of middle school chemis-
try teachers on the effectiveness of teaching
interaction. Dissertation; Central China
Normal University, 2016. [Chinese]
http://cdmd.cnki.com.cn/Article/CDMD-
10511-1016271076.htm
Mak, B.S.Y., Cheung, A.C.K., Guo, X., &
Abrami, P.C. (2017). Examining the impact
of the ABRACADABRA (ABRA) web-
based literacy program on primary school
students in Hong Kong. Education and In-
formation Technologies, 22(6):2671-2691.
DOI: https://doi.org/10.1007/s10639-017-
9620-3
Peng, P., & Tao, S. (2009). The role of word
decoding, English language comprehension
and general cognitive abilities in Chinese
children’s English reading learning. Foreign
Language Teaching and Research, 41(1):30-
37+80-81. [Chinese]
http://www.cnki.com.cn/Article/CJFDTotal-
WJYY200901007.htm
Qi, W. (2009). Retrospect of the study of chil-
dren’s language in my country. Journal of
Nanjing University of Posts and Telecom-
munications (Social Science Edition),
11(1):54-59. [Chinese] DOI:
https://doi.org/10.14132/j.cnki.nysk.2009.01.
002
Savage, R., Abrami, P.C., Piquette, N., Wood, E.,
Deleveaux, G., Sanghera-Sidhu, S., & Bur-
gos, G. (2013). A (Pan-Canadian) cluster
randomized control effectiveness trial of the
ABRACADABRA web-based literacy pro-
gram. Journal of Educational Psychology,
105(2):310-328. DOI:
https://psycnet.apa.org/doi/10.1037/a003102
5
Snow, C.E., & Hoefnagel-Höhle, M. (1978). The
critical period for language acquisition: Evi-
dence from second language learning. Child
Development, 49(4):1114-1128. DOI:
https://doi.org/10.2307/1128751
Su, L. (2000). The impact of mother tongue
transfer on second language learning. Jour-
nal of Beijing International Studies Universi-
ty, 7(4):44-52. [Chinese]
http://www.cnki.com.cn/Article/CJFDTotal-
JDEW200004009.htm
Wang, L., & Wen, Q. (2004). The transfer of
native language proficiency to second lan-
guage writing: cross-language rationale and
path. Foreign Language Teaching and Re-
search, 48(3):2 05-212+241. [Chinese] DOI:
https://doi.org/10.3969/j.issn.1000-
0429.2004.03.006
Williams, K.T. (2001). Group reading assess-
ment and diagnostic evaluation. American
Guidance Service.
Wolgemuth, J.R., Abrami, P.C., Helmer, J.,
Savage, R., Harper, H., & Lea, T. (2014).
Examining the impact of ABRACADABRA
on early literacy in Northern Australia: An
implementation fidelity analysis. The Jour-
nal of Educational Research, 107(4):299-311.
DOI:
https://doi.org/10.1080/00220671.2013.8233
69
Wolgemuth, J.R., Savage, R., Helmer, J., Harper,
H., Lea, T., Abrami, P.C., Kirby, A.,
Chalkiti, K., Morris, P., Carapetis, J., &
Louden, W. (2013). ABRACADABRA aids
Indigenous and non-Indigenous early litera-
cy in Australia: Evidence from a multisite
randomized controlled trial. Computers &
Education, 67:250-264. DOI:
https://doi.org/10.1016/j.compedu.2013.04.0
02
Wood, E., Grant, A.K., Gottardo, A., Savage, R.,
& Evans, M.A. (2017). Software to promote
young children’s growth in literacy: A com-
parison of online and offline formats. Early
Childhood Education Journal, 45(2):207-217.
DOI: https://doi.org/10.1007/s10643-016-
0779-9
Gu et al. ABRA Improves the English Reading of Chinese Elementary Students.
SIEF, Vol.9, No.2, 2021 1240
Wu, Q. (2000). Cognitive psychoanalysis on the
process of second language acquisition. For-
eign Language Teaching and Research,
44(4):290-295. [Chinese]
http://www.cnki.com.cn/Article/CJFDTotal-
WJYY200004009.htm
Wu, Y., & Wen, Z. (2011). Statistical analysis
process related to null hypothesis testing.
Psychological Science, 34(1):230-234. [Chi-
nese] DOI:
https://doi.org/10.16719/j.cnki.1671-
6981.2011.01.038
Received: 06 August 2021
Revised: 09 August 2021
Accepted: 17 August 2021
