A Pilot Study: Magic Tricks in the ELL Classroom Increasing Verbal Communication Initiative and Self-Efficacy

Instructional practices for English language Learners (ELLs) are multifaceted. They must address everything from communication skills to learner motivation. As a means of tapping student curiosity, learning to perform simple magic tricks is a creative task-based language teaching approach that promotes student self-confidence and engages them in interactionally authentic language. The learning of a magic trick becomes the means of helping students to use the linguistic knowledge they already have as well as a source for new linguistic knowledge. Teaching ELLs simple magic tricks is one approach that increases student communication and produces improvement in academics, self-confidence, resiliency, and social skills.


Introduction
The elementary and secondary school English Language Learner (ELL) population is the fastest growing student group in the United States and this trend is predicted to continue for decades (Sheng, Sheng, & Anderson, 2011). Effective academic and social integration of this population is not only dependent on increasing their oral language skills, but their confidence to speak English (Derwing & Munro, 2013). Finding effective, engaging, and motivating activities that meet the communication needs of ELLs can be difficult. Students may feel reticent to communicate in English in language development classes and their affective filter rises as their efficacy in English is challenged (Ash, Rice, & Redmond, 2014). English Language Development (ELD) teachers understand the importance of using a program that is rigorous while creating an environment where students feel safe, are willing to verbal encouragement from others (social persuasion), and has positive emotional and physical reactions to situations (psychological responses) (Bandura, 1992).
Self-Esteem. The term self-esteem is most often used to describe a person's overall sense of self-worth.
It involves a variety of beliefs about oneself (appearance, emotions, beliefs, and behaviors) and it plays an important role in personal motivation and success in life. It is also an influential factor impacting learning (Roman et al., 2008) and a student's academic motivation and performance (Aryana, 2010).
When students understand that their failures are the consequence of effort rather than ability, they may demonstrate greater persistence to achieve rather than develop an attitude of helplessness (Aryana, 2010). Developing a growth mindset teaches children to see their perseverance and effort as a key to success and failure as an opportunity to improve and grow.
Grit. The term "grit" was coined by Angela Duckworth, a psychology professor at the University of Pennsylvania. She defines grit as the tendency to sustain interest in and effort toward very long-term goals (Duckworth, Peterson, Matthew, & Kelly, 2007). Grit and resilience are often used interchangeably but they are two different constructs. Resilience is the optimism to sustain efforts despite difficulties and is a predictor of well-being (Stoffel & Cain, 2018). Grit is the motivational drive, perseverance, and passion toward a long-term goal, the sustained commitment to complete a specific activity despite occasions of failure or setbacks (Stoffel & Cain, 2018). Grit is directly correlated with academic success and a better predictor of achievement than IQ (Stoffel & Cain, 2018;Duckworth et al., 2007).

The Pilot Study
This pilot study was conducted to examine how an arts-based TBLT approach would impact the role these three constructs play in increasing verbal communication initiative of ELLs. The art of magic was chosen because of its ability to capture and hold the attention of people of all cultures, ages, and abilities. Teachers who take advantage of the multidimensional features of learning to perform a magic trick employ instructional practices that develop skills in storytelling and presentation making it an appropriate activity or task for ELLs.

Magic Tricks and TBLT
A small number of education researchers evaluated the effectiveness of using magic tricks as an arts-integrated, student-centered teaching approach. Each researcher concluded that future investigation should be done based on their positive results, which include: (a) magic tricks offer a creative means for stimulating the senses in special education students (Frith & Walker, 1983), (b) magic tricks enhance the learning experience and encourage creative problem-solving skills, observational techniques, and critical thinking (McCormack, 1985), (c) magic tricks provide a strategy for building teamwork and self-esteem in children with Emotional Behavior Disorders (Broome, 1989), and (d) magic tricks in an educational setting can help students with learning differences attain higher self-esteem and self-confidence (Ezell & Ezell, 2003).
Magic tricks as an arts-based approach embraces a holistic philosophy, which is essential to effective The use of magic tricks in the classroom has shown positive results to increase the self-efficacy and self-esteem of at-risk students and those with special education needs (Ezell & Klein-Ezell, 2003).
These findings have implications both psychologically and pedagogically. Psychologically, students experienced: (1) a greater sense of self-esteem and self-confidence in their abilities to perform the magic tricks, (2) a new self-determination to achieve their goals, (3) a self-actualization where the students realized their potential and (4) ametacognition of what they learned in each goal accomplishment. Ezell et al. reported an increase in student self-efficacy as their fear of failure was replaced by a willingness to take a risk.
The pedagogical implications reported in the literature impact both the students and the teachers. For students, there was a heightened sense of (1) engagement where they were energized to solve problems and use their creativity in performing, (2) collaboration where the students wanted to work together to better their skills, and (3) 21 st century skills that are needed in education and real-world settings, such as problem solving, collaboration and communication. Teachers experienced an increased level of efficacy as they helped students achieve their goals. They also developed a proficiency in teaching the magic curriculum. Finally, they felt satisfaction in having a creative, fun curriculum to teach, followed by celebrating the students' successes.
Throughout this pilot study, a systematic, magic trick-based approach was used as a supplement curriculum. Students continued to work in their Computer-Assisted Language Learning (CALL) programs and in the district-approved textbook for their ELD time.

Method
This paper reports on a pilot study conducted to develop pragmatic, social language and increase verbal communication initiative in English Language Learners (ELLs). The rationale of the study is in the use of an arts-themed, Task-Based Language Teaching (TBLT) approach based on the learning of simple magic tricks. The magic trick lessons were used as a supplemental program and student conversation initiation frequencies were observed as a measure of pragmatic language development. The use of an organized, systematic approach to using magic tricks for English Language Development has not been previously researched.

Participants
Participants were members of an ELL newcomer program known as the Accelerated Language Learning (ALL) program located at a junior high school in northern Utah. Students either lived within the school's boundaries or they were referred to the district and placed in the school on a guest status.
They were placed in the ALL program based on their length of time in the country and the results of the World-Class Instructional Design and Assessment (WIDA) Screener. The WIDA screener was administered within the first week of the students' school registration.
Participants in the pilot study were male and female students (n = 21; SD = 14) in grades seven through nine from various countries with diverse English language experience and proficiency (Table   1).

Research Design
Throughout the pilot study, students worked in their Computer-Assisted Language Learning (CALL) programs and in the district-approved textbook for their ELD time. The magic trick-based program was used as supplemental curriculum.
Assessing Perception of Language Acquisition Skills. A single subject design across participants was used to assess student perceptions in language acquisition and communication skills. Students provided answers in pre-and post-surveys on self-perceptions in three areas: (a) self-efficacy; (b) self-esteem; and (c) grit.
Assessing English Conversation Initiation. A single-subject A-B design across multiple participants was used to gather observations of the rate and frequency of English conversation initiation of the students. "Conversation initiation" was defined as any English word spoken in the presence of another person without a prompt, that is ELLs speak first. Participants in the single subject design included two students from each grade level for a total of six students. Although only six students were observed for initiation of conversation, the entire class of 21 students participated in the magic trick-based supplemental curriculum. Every student in the ELD program completed the magic program, performed in the magic presentations, and completed the pre-and post-surveys.
Assessing the Effects of Hocus Focus on ELLs. Students learning English as their second language are often reserved and reticent when speaking with native English speakers. To determine if the magic treatment would positively influence English conversation initiation, self-efficacy, self-esteem, and grit, a mixed-methodology was required. The researcher assessed the effects of the magic trick supplemental curriculum on student initiation of communication using a single subject A-B design with two replications as described by Kazdin (1982); quantitative data from analysis of the self-esteem surveys; and completion data to analyze the student performance at the end of the treatment.
To assess the Inter-Observer Agreement (IOA), a second observer (a staff assistant familiar with the students) took data on approximately 25% of the observations. Overall, the agreement was four observations for each of the six participants for a total of 24 observations. The observations were completed during the same time periods, but were tallied independently by the researcher and the second observer. They were also observed from different vantage points in the classroom. When comparing the data collected by the researcher to the second observer, the data tallies were found to be consistent.

Setting
The students attended the Accelerated Language Learning (ALL) program 85 minutes on an A/B schedule for the entire year while being mainstreamed for other content areas. The ALL program also serves as the students' language arts class.

Procedure
Parental Consent. The study was approved by the Institutional Review Board of Weber State University. Approval was sought and received from the participating school district and the school principal. Parental consent forms were sent home with each of the students (Appendix B). Upon receipt of consent forms, pre-intervention procedures were initiated.
Assessment Instruments. The rate of student conversation initiation was measured using a behavior rate tally sheet (Appendix C). Frequency tallies of students' conversation initiation were kept and the length of observations was recorded. The rate of conversation initiation was calculated using the formula: conversation initiation frequency divided by minutes observed.
To explore the self-esteem and self-efficacy, students completed two surveys: (1) the Self-Efficacy Scale Condensed Version, a 15-item Likert scale survey used in the magic trick curriculum (Appendix E) and (2) the Rosenberg Self-Esteem Scale, a 10-item Likert scale survey (Appendix F). To explore grit, students completed the Duckworth 8-Item Grit Scale (Appendix G).
Before starting the treatment, the three surveys were administered to obtain pre-intervention data. The surveys were administered as the teacher's "I want to get to know you?" and "How can I help you this year?" activity to obtain more accurate data of their self-perceptions. The surveys were translated verbally in the native languages (Spanish, Mandarin Chinese, Portuguese, and Amharic) of the participants to obtain data that would be as accurate as possible.
The practice sessions during the class and student participation in the magic show were also observed.
Upon completion of each weekly lesson, a Wizard's Academy Chart (Appendix D) was used to provide students with a visual representation of their successes and to reinforce their newly developing skills.
Following the successful completion of the magic trick curriculum, the three surveys were administered again as post-assessment in the same manner as the pre-assessment.

Treatment
The magic trick curriculum was introduced to the students (HocusFocusEducation.com). Lesson one introduces the history of magic and the contributions of magicians to modern society. It was presented to the entire class of 21 students. Lesson one also emphasizes the importance of the magicians' Code of Ethics, which includes not revealing the secret to a magic trick. As each student signed the Code of Ethics agreement, they were given a small steno notebook that served as their Wizard's Book of Secrets, a journal for their personal notes and an organized way to write the script for each magic trick for their performance.
Creating Introducing the Magic Tricks. Students were introduced to the magic lessons on a rotation basis determined by their grade level. Prior to beginning the study, four different lesson plans were selected for each group. This provided the opportunity for each group to learn tricks that were different from those of their peers (Table 2). Throughout the intervention, instructional sessions and student performances were captured on video and sent to the designer who reviewed and assessed them for treatment fidelity. Data were collected five times to determine baselines during each phase of the treatment.

Baseline One
Initial baseline data were established through observations on student conversational interactions with other students, teachers, and staff assistants following the teaching of a standard ELL introductory conversation activity. An example from this lesson was, "Hello, my name is… What is your name? It is nice to meet you too." The students were observed over a three-day period as they interacted with others without any further prompt from teachers or staff assistants.

Baseline Two.
Group One. Once baseline stability was determined for all participants, the second lesson was presented to the group of seventh graders. These steps were used for treatment procedures: (1) the teacher performed the two magic tricks included in the lesson while the students practiced being a good audience; (2) the teacher used a magic instructional DVD to reveal the secret and teach the tricks to the students; (3) the students used their steno notebooks (Wizard's Book of Secrets) to take notes on how to perform the trick in their native languages; (4) the students brainstormed ideas for a possible script (patter) to use while performing the trick in English using electronic translation devices and the Internet to accomplish this task; and (5) observations of English conversation initiation were tallied to determine a second baseline. In addition, data were kept on the students performing the tricks for one another.
Group Two. Group two was moved from baseline one into the treatment phase following the same procedure as group one using a different lesson plan and different magic trick. As with group one, observations of English conversation initiation were tallied to determine the second baseline and data were kept on the students performing the tricks for one another.
Group Three. Group three was moved from baseline one into the treatment phase following the same procedures that were used for groups one and two. Group three was also taught a different magic trick than both groups one and two. As with groups one and two, observations of English conversation initiation were tallied to determine the second baseline and data were kept on the students performing the tricks for one another.

Baselines Three and Four
Once baseline two was determined for all participants in each of the three groups, the treatment procedure was repeated with each group using the next selected magic lesson as previously determined for each group.
Observations of English conversation initiation were tallied to determine the third and fourth baselines and data were kept on the students performing the tricks for one another.

Baseline Five
The final baseline was established as the students performed for their assigned audiences: two preschool classes; a self-contained severe special education class; the school administration and secretaries; the cafeteria staff; the custodians; and the student government class. These audiences were specifically selected because the participants were typically hesitant to initiate conversation with these group of adults and students.

Results
Data that addressed participation in the magic trick curriculum to increase ELLs' English conversation initiations were evaluated using trend analysis across A-B phases and a calculation of Percent of Non-overlapping Data (PND). PND is one of the most frequently used effect size descriptors and takes into consideration baseline variability and slope changes. PND was determined by finding the highest data point in the baseline phase for each subject and calculating the proportion of data points in treatment that exceeded this point.
Overall PND ranged from 83% to 100% with the average of 88.72% (Table 3). Three students scored within the Highly Effective range of 90% and higher, while three participants scored within the Moderately Effective range of 70-89%, according to PND ranges provided by Louisiana State University. For the two female participants, pre-intervention performance showed a slight upward trend, whereas the four boys showed slight downward trend. The graphical and statistical outcome showed an overall upward trend following the intervention. Each student had increases in English conversation initiation following each selected magic lesson. The range of zero to two in the baseline increased to one to eight incidences following treatment. Some conversation initiations were simple greetings or requests, while others were more substantial sentences (Figures 1-6).     one ninth grade student per team).
The results of the pre-and post-surveys were entered into a spreadsheet where graphs illustrated any increases in students' perceptions of their self-efficacy, self-esteem, and grit. The results of the three pre-surveys indicated a wide range of responses among the 21 students. Each survey item was analyzed independently for student response. The mean responses on the post-surveys indicated that all 21 students generally had a higher sense of self-efficacy, self-esteem, and grit.
The Self-Efficacy survey results indicate there was an overall mean increase from 74 to 86 (12 points) as illustrated in Figure 7.  The statement, "I can say 'no' to someone who is asking me to do something wrong," produced the highest mean increase of 41 points, while the two statements, "I can live up to what my teachers expect of me," and "I ask teachers to help me when I get stuck on homework," produced the lowest mean increase of 0.5. Figure 8 illustrates the mean increases for each of the 15 questions in the self-efficacy survey.

Figure 8. Hocus Focus Self-Efficacy Scale Mean Scores
The results of the Rosenberg self-esteem survey indicated there was an overall mean increase from 14 to 19 points out of the possible 30. As Figures 9 and 10 demonstrate, all but one of the 21 students self-identified an increase in their self-esteem following the magic trick treatment. The student who did not show an increase maintained the score of 15. Three students revealed the highest increase in self-esteem with a mean score increase of 10, while the lowest increase as revealed by four students as a mean increase of one point.

Figure 10. Rosenberg Self-Esteem Scale Questions Mean Scores
The 8-Item Grit scale results reveal an increase from 2.9 to 3.4, with 5 being extremely "gritty" and a 1 being not gritty at all. Figure 11 indicates half of the students maintained the same score, while the other half had a one point increase. The highest score in the post-survey was a four, while the lowest score was a three. The statement, "I have difficulty maintaining (keeping) my focus on projects that take more than a few months to complete," garnered the greatest gain of 1.05 points increase, while the statement, "I have been obsessed with a certain idea or project for a short time but later lost interest," had the lowest at 0.5 points. Figure 12 illustrates the mean increase in grit by the eight items in the survey.

Figure 12. 8-Item Grit Scale Questions Mean Scores with 5 being very Gritty
The students had a profound increase in their English conversation initiations throughout the treatment and following the final performance of the tricks. The students also self-reported a substantial increase in their self-efficacy, self-esteem, and grit.

Further Observations and Discussion
This pilot study has strong implications for teachers and districts with students learning English as their second language as it is a viable supplemental curriculum for verbal English skill building. The participants in this study were diverse in language and culture and corroborated the research on ELLs' shyness when attempting to converse in English in the presence of native English speakers (Rice, Sell, & Hadley, 1991). The students entered the study with varying experiences with English, but all exhibited degrees of shyness when meeting the teachers and classmates. When prompted to speak in activities designed to introduce themselves, most students were extremely reluctant and much modeling, scaffolding, prompting, and encouragement was necessary.
The research data supported that this magic trick-based curriculum provides a platform to increase student self-esteem and self-efficacy. Prior research using a magic trick-based method focused on students with special needs or those considered at-risk; however, ELLs were included in this pilot study because they are considered at-risk populations in our schools.

Increasing ELLs' Initiating of Conversations in English
As the observation data indicates, the students' frequency of initiating English conversation with a native English speaker increased within each baseline. As the observations were made to establish the first baseline, the students were generally in a non-verbal or quiet stage. All conversation was initiated by teachers and staff members. However, an increase of genuine conversation initiation began to flourish once the magic lessons began. Students were observed greeting native English speakers, asking questions and making declarative statements such as, "My computer is not working," or "I have homework for math." As the teacher demonstrated the magic tricks, and sometimes failed to do them correctly, some of the students would make statements like, "I know, I know. I will do it." This was not only an initiation of conversation, but an opportunity for the students to demonstrate leadership skills and understanding by teaching the other students the trick.
Translator devices were used at times to create English conversation initiators. This provided opportunities for students to create unscripted, authentic dialogue of which they took ownership. The goal of all ELD teachers is for students to become more fluent English speakers (Fernando Gomez-Rodriguez, 2010).
In other content areas, teachers reported an increase in the students' willingness to approach them to make requests for help, requests to use the hall pass or to simply say hello. This was not included in the data collection, as it was not observed by the teacher or the inter-observer. However, it could provide anecdotal evidence of conversation initiation in English.

ELLs Performing the Magic Tricks
Students followed a process as they learned to perform their tricks: (1) completed their lesson, (2) practiced individually and as a group in the classroom, (3) practiced at home with their families (confirmed by their parents), and (4) presented their tricks in their final magic show. Research suggests that students require multiple opportunities to become proficient in the performance of a magic trick (Ezell & Klein-Ezell, 2003) and this was evidenced in this study as well. ELL students practiced individually but they also practiced as a group where ELL students experienced a low affective filter in their solo and group rehearsals which is purported to be necessary for optimal ELL learning environments (Schinke-Llano & Vicars, 1993).
Each student participated in their magic team's performance for an audience of native English speakers.
Students presented no hesitation to perform in front of their audience, successfully fulfilling the study's expectations.

ELLs Increase in Self-Reported Self-Esteem and Self-Efficacy
As the magic tricks were taught and the students had the opportunity to use their English in what they referred to as "fun lessons", speaking in English appeared to be less daunting for them. The students and teachers demonstrated an important component to learning (Bauman, 2012)-having fun! This study confirmed that implementing magic tricks as a part of the language learning process can provide that fun experience. Since the magic trick program used is formatted into eleven lessons, it was feasible to teach the lessons in small groups.
The participants and the teacher indicated this small group setting provided not only a bonding experience between the students, but with the teacher as well. This setting also helped lower the affective filter and increase the interpersonal relationships, which is critical for communication (Erokzan, 2013). Finally, the small groups allowed each group to perform a trick the other groups did not know. This, according to the students, further elevated their sense of self-esteem and self-efficacy by presenting something that could not be equaled by their peers.
Once the students worked through their lessons, practiced English language dialogue, and performed their magic shows, data demonstrated an increase in their self-efficacy, self-esteem, and grit. These traits were evident and observable in the students' body language as they smiled, celebrated, and wanted to watch the video of their magic show performances. The data obtained from self-reporting also indicated a significant change with improvement in the students' perceptions of themselves, with some much higher than others.
Research (Ezell et al., 2003) reports an increase in students' willingness to take a risk, which lead to higher levels of motivation and improvement in other areas. This proved to be consistent with ELL students as was evident in their attitudes and behaviors. Casual conversations between teachers resulted in celebrating the students' change in effort, work ethic, and production.
Students deserve to feel empowered, and when they feel that empowerment, it carries over into their personal lives and academic performance (Ezell et al., 2003). The students in this pilot study demonstrated a similar increase in empowerment in their survey responses and classroom performances.

Limitations and Recommendations
Given the time restraints of this study, the research method of single subject A-B design with two replications was used. While a multiple baseline across multiple subjects would be more effective in to eliminate possible threats to internal validity, the replication in this study speaks to the validity. Future research could also use the qualitative methodology to determine an increase in students' self-esteem, self-efficacy, and grit using journals and/or interviews in addition to the surveys.
While there was an overall increase in the students' self-reporting on their self-esteem, self-efficacy, and grit, the high increase in the statement, "All in all, I am inclined to feel that I am a failure," and the low increase in the statement, "On, the whole, I am satisfied with myself," proved to be disturbing to the researcher. Upon further discussion with the students, it was concluded that the timing of the survey may have influenced the students' responses. The post-surveys were given following the students' first-term grade reports. Teachers reported that, due to language barriers in the classroom, several students had received failing or near-failing grades. The students admitted they had internalized their low scores as being failures. Also, several of the students admitted to being homesick for their families as they were living in the United States with extended family. Teachers or future researchers would need to take this limitation into consideration when assessing the students' self-esteem, self-efficacy, and grit.
Allowing ELLs the opportunity to write their performances notes in their native language proved very effective. They did not need to focus on understanding their notes because they understood the language. Just as important was the opportunity for students to write their own scripts for the performance component. The use of translator devices allowed them to accomplish this task very effectively. The steno notebook (Wizard's Book of Secrets) was also used as a venue for the students to reflect on their understanding of the tricks and their feelings on their performances. It served as a journal and it was referred to often by the students. Future research might consider the impact these accommodations have on the success of the program when used in an ELD classroom.
With such a large group of students with varying English abilities, a management plan was necessary but was not initially taken into consideration. In this pilot study, breaking the students up into smaller groups or teams proved to be mutually beneficial to the students and the teacher. If the class size is small, breaking the students up into partner teams could provide that personal attention and allow the students to learn different tricks.
Videotaping the students' performances provided a great source of feedback for the students so they knew what areas of their performance they needed to work on. For teachers wanting to implement videotaping as a part of this type of TBLT activity, it could also provide a great opportunity for celebration as the students in this study stated they felt a sense of celebrity while viewing their magic shows. Finally, as a recommendation, video can serve as evidence of the students' growth in English, not only for data collection, but for parents who want to see their students' growth.
While the purpose of this pilot study was to track conversation initiation in English, further research could be conducted to observe any increase in English conversation, whether initiated by the ELL or not. In this pilot study, the teacher and inter-observer noticed an increase in the amount of conversations in which the ELLs were participants, but these were not tallied because they were not initiated by the ELL. Future research could be conducted on both initiation and participation to obtain a broader data set.
In working with newcomer ELLs on an A/B block schedule, the 15 minutes per day recommended by the curriculum designer proved to be insufficient time to spend with their lessons because of the language transfer. The students wrote notes for their tricks in their native language, but also brainstormed English phrases they could use to present the trick. This took more time than would be necessary for a native English speaker using the lessons. Not only were they practicing the trick, but also the English they were going to use. Additionally, with a day in between classes, thirty minutes was a more appropriate timeframe. Teachers wanting to use this treatment with newcomer ELLs may want to allot more time, and future researchers should also take this into consideration.
Finally, while the magic trick-based treatment was used as a supplemental curriculum, it is anticipated that the results are permanent. Further research needs to be conducted to determine the long-term effects of learning magic tricks on ELLs' self-esteem, self-efficacy, and grit. It would be interesting to determine if periodic refreshers of the tricks and magic performances throughout the year would further cement the students' gains. While each grade-level group performed three to four different tricks due to time restraints in this pilot study, it would be worth investigating if teaching more tricks would be beneficial. Given the positive gains attained by the participants, further research could determine if even higher increases could be attained.