Comparing Experiential Approaches

Comparing Experiential Approaches: Structured Language Learning Experiences versus Conversation Partners for Changing Pre-Service Teacher Beliefs Shannon McCrocklin1a Abstract ARTICLE HISTORY: Received December 2019 Received in revised form February 2020 Accepted February 2020 Available online March 2020 KEYWORDS: Teacher education Experiential learning Second language acquisition Beliefs Classroom-based research Research has shown that language teachers’ beliefs are often difficult to change through education. Experiential learning may help, but more research is needed to understand how experiential approaches shape perceptions. This study compares two approaches, conversation partners (CONV) and structured language learning experiences (SLLE), integrated into a course in language acquisition. Participants (n = 32) completed a pre- and post-questionnaire that included: demographic questions, Likert scales on beliefs, ranking of second language acquisition (SLA) factors, and open-ended questions. Results showed differences from preto post-questionnaire for both groups for four Likert scale items (e.g., the importance of exact pronunciation) and six language learning factors (e.g., motivation). Further, both groups grew to recognize more factors. Slight differences emerged between groups on two items, the importance of intelligence and dominance in SLA. Qualitative analysis showed that, while CONV reported no changes in beliefs, SLLE reported finding language learning to be more difficult than believed. Further, SLLE wished to have more of the project, while CONV did not. © 2020 IJSCL. All rights reserved. 1 Assistant Professor, Email: shannon.mccrocklin@siu.edu Tel:+1-618-4533428 a Southern Illinois University, USA S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 1. Introduction I n recent years, there has been increasing interest in teacher cognition (Borg, 2015). Although there is a complex relationship between teacher beliefs and teacher actions, beliefs are foundational to actions and, as such, are important to explore (Borg, 2011). One strand of research in teacher cognition has focused on how teacher beliefs are affected by pre-service education. Although beliefs are often resistant to change in pre-service teacher education, experiential learning seems to be a promising approach to changing problematic beliefs. Despite research into the impact of experiential approaches as part of teacher education, only a few empirical studies have examined structured language learning experiences and even less have examined conversation partners. Further, previous research on conversation partners and structured language learning experiences focused on a single approach and did not make comparisons. This study seeks to examine the ways, if any, that these experiential approaches affect pre-service teacher beliefs and compares the impact of each. 2. Theoretical Framework 2.1. Language Teacher Education Education for effective language teachers must include not only information about the language of instruction, principles of L2 development, and approaches to teaching (Senior, 2006) but also must increase awareness of teaching practices and beliefs through reflection and critical evaluation (Farrell, 2018). Student-teacher beliefs, however, are often ingrained and difficult to change through education programs (Moodie, 2016; Tatto, 1998; Wideen, Mayer-Smith, & Moon, 1998; Wubbels, 1992). Research has shown that language teachers are often heavily influenced by their own language-learning experiences (Borg, 2003, 2011; Ellis, 2016; Kagan, 1992; Moodie, 2016), and student teachers often comb through received information to support and strengthen pre-existing notions, instead of challenging and re-evaluating beliefs (Kagan, 1992). Yet, work to examine and improve teacher education shows that programs can affect beliefs (Blume, Gerlach, Roters, & Schmidt, ISSN 2329-2210 2019; Busch, 2010; Kavanoz, Yüksel, & Varol, 2017; Mattheodakis, 2007; Mighani & Moghadam, 2019). Mattheodakis (2007) conducted a longitudinal study following students through a four-year teacher-education program for prospective English as a foreign language (EFL) instructors and found that several pre-service teacher beliefs did change. Mattheodakis noted that education influenced about half of the examined beliefs. Students’ beliefs about the importance of grammar and pronunciation in language learning tended to decrease as students encountered evidence and re-examined their original beliefs. Half of the beliefs were resistant to change, however, such as beliefs about the relative difficulty of language learning. Busch (2010) also conducted a longitudinal study over the course of three years, examining the beliefs of preservice teachers (n = 381) enrolled in a teachereducation program. She found that participants reported numerous changed beliefs. Further, while initial beliefs were often reported to be based on previous language learning experiences, changes of beliefs were attributed to experiential learning programs. 2.2. Experiential Learning Given the challenges in encouraging prospective teachers to evaluate their beliefs, researchers should continue to explore possible improvements to teacher-preparation programs. One path to improving language-teacher education may involve experiential learning, in which learning is viewed as something to be shaped (and re-shaped) through experience (Kolb & Kolb, 2005; Moon, 2004). An important element of the experiential learning design is reflection (Blume et al., 2019; Busch, 2010; Wright, 2010). “Teachers do not simply reproduce their own experience in the classroom. However, if they are to transcend its effects, it is reflection on the experience that is critical” (Ellis, 2006, p. 2). Matic (2011) noted that student teachers often lacked the ability to reflect critically on their own learning experiences, being unable, for example, to identify specific reasons why they enjoyed communicative activities more than grammartranslation, which prevented them from applying those realizations to their own teaching. Experiential learning allows the learner’s beliefs to be examined through new educational experiences, with reflection 71 72 Comparing Experiential Approaches leading to more refined ideas that may be better aligned to new theories or methods of teaching (Kolb, 1984). Reflection can, therefore, help resolve tensions or competitions between beliefs and new information (Farrell, 2006). The most common channel for introducing experiential learning and reflection in L2 teacher education is teaching practicums (Borg, 2009; Crandall, 2000). Additional methods to enhance experiential learning have included conversation partners and structured language learning experiences (SLLEs). Although conversation partners are recommended for L2 learners as a strategy for promoting acquisition (Oxford, 2002), there seems to be limited research into the use of conversational partners as part of ESL teacher education, specifically. One key study, Biondo Salomão (2011), explored the use of tandem conversation partners, in which people with two different native languages serve as conversation partners for the other, finding that student teachers found the experience useful for learning to describe their language in contrast with another, leading the author to recommend conversation partners be a part of teacher-education programs. Additionally, Keengwe (2010) explored the use of conversational partners as part of a multicultural education course for elementary education majors, finding that preservice teachers partnered with ESL students from a local language institute valued the experience and reported greater appreciation of linguistic and cultural diversity after the experience. Another approach to integrating experiential learning in language-teacher education is for pre-service teachers to re-enter the role of language learners through what Ellis (2006) terms a structured language learning experience (SLLE). Ellis notes that SLLEs may range from a single short lesson to a semester of study, but a key element is using the experience to encourage reflection on language teaching from the student perspective. Weed (1993) advocated for speaking in an unknown language to prospective teachers in short stretches and then encouraging the student teachers to reflect, both exploring emotional responses as well as critical reflections on the ways language modifications can affect comprehension. The ability of SLLEs to evoke emotional reactions, specifically frustration, was also noted in Washburn (2008) and WrightMaley and Green (2015). Further, SLLEs can help learners connect received information (content information presented in the classroom) with experiential knowledge and provide prospective teachers a chance to test out their ideas about second language learning (Flowerdew, 1998). While it is clear that teacher educators and researchers have explored various experiential approaches to ESL teacher education, no studies were identified that worked to compare two experiential projects. More research is needed to understand the ways that different experiential approaches may impact and shape perceptions of second language acquisition (SLA). Additionally, some of the findings have been based on anecdotal evidence or small case studies (e.g., Washburn, 2008; Weed, 1993). More work is needed that seeks to empirically examine changes to pre-service teachers’ beliefs as a result of particular interventions or experiences. Therefore, this study compares two experiential approaches, conversation partners, and structured language-learning experiences, as part of teacher education, exploring the ways each approach affects changes in beliefs. Specifically, this research study seeks to answer the following questions: 1. For students participating in a course in language acquisition, in what ways do structured language learning experiences (SLLE) and conversation partner experiences (CONV): a. affect beliefs about second language learning b. increase recognition of a number of factors that affect SLA? 2. What are the students’ reactions to the two experiential projects? 3. Methodology The present study was conducted at a university in the South of Texas in the United States of America. It was conducted in two sections of an advanced undergraduate course in language acquisition that covered first language acquisition, SLA, and bilingualism. The course counted for credit in degree plans, including majors and minors, in English as a Second Language, English, English Education, and Bilingual Education. All students enrolled in the course participated in the activities, but S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 students were provided informed consent and could decline to contribute their data to the study. 3.1. Participants After receiving information about the study, 32 students provided consent for their data to be included in the study. The participants included ISSN 2329-2210 66% females and 34% males with an average age of 24. The majority of participants (62.5%) were English (which includes English Education) or Education (Elementary or Bilingual) majors. An additional four students were Spanish majors, while a variety of other majors were also represented including Psychology, Biology, and Kinesiology. Table 1 shows the data for each group. Table 1 Participant Characteristics by Group Number of Participants Average Age % Bilingual in Spanish/English % Studied/Learned an L2 after age 12 Gender University Major Notably, the majority of participants, 69%, reported being bilingual in Spanish and English. This includes students who began learning both languages before the age of six and those who reported advanced proficiency in both languages. Bilingualism is common in South Texas and families are often able to maintain bilingualism into their fourth and fifth generations (Anderson-Mejías, 2005). Research suggests that bilinguals may have greater insight into second language (L2) learning than monolinguals in explaining how English functions to create meaning (Ellis, 2004), better understanding the challenges of identity formation across two cultures (Ellis, 2006), and experience trying to communicate in an L2 (Ellis, 2006). However, the researcher noted that in previous semesters of the course, bilinguals were also sometimes judgmental of monolinguals and adult L2 learners, thinking them lazy or unintelligent not to have acquired an L2. Such judgmental comments, uttered with concerning frequency, formed much of the impetus for re-envisioning and reformatting the language acquisition course and creating this study. 3.2. Experiential Treatments Each section, offered in different semesters, experienced one of the experiential learning CONV 11 23.5 64% 36% Male= 5 Female= 6 Education= 1 English= 7 Spanish= 1 Other= 2 SLLE 21 25 71% 29% Male= 6 Female= 15 Education = 10 English= 3 Spanish= 3 Other= 5 projects, either conversation partners (CONV) or the structured language learning experience (SLLE). The project was embedded into the language acquisition course as a major project relating to the SLA unit and goals. However, as SLA was only one portion of the course, usually comprising about 40% of the semester, it was important to keep the in-class time for both projects reasonable and limited. 3.2.1. Structured Language Learning Experience (SLLE) Participants in the structured language learning experience (SLLE) worked through the semester to learn French. Half of the work for the SLLE was completed through Duolingo, a web-based language-learning platform also available as an application for Android and iPhone. Although Duolingo has been criticized for inauthentic language and a strict curriculum (Cunningham, 2015), it has been shown to be effective in helping students improve their language abilities (Vesselinov & Grego, 2012) and is free and convenient. Students completed 11 units of French Lessons within Duolingo, estimated to total three to four hours of practice. After completing the basic units, students progressed through units such as food, animals, adjectives, plurals, colors, and possessives. 73 74 Comparing Experiential Approaches Further, students took part in four 50-minute lessons in class. The in-class lessons built upon material that students learned through Duolingo, such as greetings, which students were first introduced to during the basic units. Each in-class lesson focused on creating communicative experiences; for example, students introduced themselves to each other in the greetings lesson and completed a restaurant role-play to practice food vocabulary. In-class lessons also included activities from additional methods, such as Total Physical Response, which was used as a warm-up in the first lesson. Throughout the semester, students were asked to respond in short paragraphs (guidance given to submit four to five sentences) to six different reflective prompts, such as “After a few basic lessons in French, how would you feel trying the language with a native speaker? What do you feel like you have mastered at this point?” and “What was your favorite activity to learn French? Why?” Often, these were completed at the end of the in-class French lessons, but due to time constraints some had to be assigned for homework. 3.2.2. Conversation Partner Experience (CONV) Participants in the conversation partner experience (CONV) were matched with an ESL student at the campus-affiliated language institute at the beginning of the semester. Students were provided contact information for their partner match and were expected to arrange times for meeting that worked for both parties. Students were required to meet with their partner at least four times during the semester, each for 50-minute sessions. Students were required to write a report and reflective essay at the conclusion of the four meetings. The essay assignment provided guidance on length (1000-1500 words) along with reflective prompts, such as, “What do you think went well during your meetings? Is there anything that did not go well? Why?” and, “What do you feel that you learned about second language acquisition from this conversation partner experience?” Students submitted the reflective reports online through the course management website. As this project depended upon the responsiveness and reliability of their ESL partner matches, who voluntarily signed up to take part in the project to attain more English practice, some students struggled to arrange meetings and persistence was sometimes required to ensure partner follow-through. Students who continued to struggle to arrange a meeting with their initial partner were rematched in the second half of the semester. However, two students continued to face issues in arranging a meeting. As such, it was necessary to create a comparable alternative assignment that would still foster exposure to L2 learners. For these cases, students prepared a list of interview questions and identified two L2 learners (of any language) to interview. Their essay, then, included less reflection (although this was still a required component) to make space for the increased research/journalistic expectations for their findings. 3.3. Instruments To measure changes in beliefs, participants completed a pre-questionnaire during the first week of the semester and a post-questionnaire in the final week. The pre-questionnaire included demographic questions (including age, gender, major and information about languages learned/known), 10 Likert scale ratings on beliefs about language learning, and identification and ranking of 13 factors that affect SLA. Similar to Mattheoudakis (2007) and Horwitz (1985), students were presented in the pre- and post-questionnaire with 10 Likert scale items in which participants ranked their agreement with popular beliefs about language learning on a six-point scale, ranging from 1- Strongly Disagree to 6- Strongly Agree. Nine of the items were chosen from Lightbrown and Spada (2013). In Chapter 9 of How Languages are Learned, the authors work to introduce 18 popular ideas about L2 learning and to clarify the truth of the issue. The popular ideas had varying levels of truth and substance, ranging from mostly false to somewhat true. Although there was not a single true/false or right/wrong answer to these statements, they reflect ideas that are popular outside of the field and had the potential to illuminate beliefs of participants. The final item included in the Likert scales was chosen based on researcher experiences with bilinguals in the region, which hinted that another popular belief may be, “Learning a second language is easy”. S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 ISSN 2329-2210 within-subjects factor of TIME (pre-/post-) and a between-subjects factor of GROUP (CONV/SLLE) with effect sizes calculated as partial eta squared (ηp²). For survey items identified as having potential significant differences when examining TIME * GROUP, a post-hoc paired samples t-test was used to compare pre- and post-responses for each group, with effect sizes calculated as Cohen’s d. Participants were also provided with a list of thirteen factors that affect language learning. The factors were selected from textbooks introducing language acquisition: Brown (2014), Lightbrown and Spada (2013), and Ortega (2008). Participants were asked to circle any factors on the list that they believed affected SLA and then, for those circled, rank the factors in importance with one being the most important factor. For example, if a student selected three factors, such as self-esteem, motivation, and learning style, they were then asked to rank those three for relative importance. If a feature was not circled, it received a ranking of 14 (one step below 13, which would have been the lowest ranking possible, if a participant did circle all factors). For the open-ended responses, the researcher examined each written response identifying themes mentioned by two or more participants. An additional reviewer also examined the responses to identify themes and provided feedback that informed subsequent analyses. After a second round of review and analysis, responses were coded and tallied to provide counts and percentages for themes. The post-questionnaire included the same Likert scale ratings and ranking of factors tasks from the pre-questionnaire, along with openended questions about the project, such as, “In what ways did this project further your understanding of adult second language acquisition?” and, “If I do this project again in another semester, what are two suggestions you have to make the project more useful for other students?”. 4. Results 4.1. Changing Beliefs about L2 Learning In order to measure changes in beliefs about language learning, students ranked their strength of agreement with 10 statements. Table 2 shows each of the statements and, then, for each group, the average pre- and post-course ratings, and the difference calculated by subtracting pre from post. In the comparisons section, results from the mixed ANOVA, including the significance and effect sizes of TIME and TIME * GROUP are included. 3.4. Data Analysis To avoid running numerous t-tests, which can increase the chances of a Type 1 error, results were examined using a mixed ANOVA with the Table 2 Pre- and Post- Likert Scale Statement Rankings by Group with Comparison Results from Mixed ANOVA SLLE Comparisons Pre- Post- Difference Pre- Post- Difference Sign. TIME ηp² TIME Sign. TIME GROUP ηp² TIME * GROUP * CONV 1. People learn languages by imitating native speakers. 4.55 3.64 -0.91 3.9 3.67 -0.23 .103 .086 .262 .042 2. Highly intelligent people are good language learners. 3.36 3.36 0.00 3.48 2.14 -1.34 .014 .187 .067 .107 3. The best predictor of success in language learning is motivation. 4.54 5.45 0.91 4.57 5.0 0.43 .011 .197 .306 .035 4. The earlier a L2 is learned the greater the likelihood of success. 5.27 5.0 -0.27 5.62 5.67 -0.05 .690 .005 .335 .031 75 76 Comparing Experiential Approaches 5. Most of the mistakes that L2 learners make are due to interference from their first language. The best way to learn new vocabulary is through reading. 4.27 4.09 -0.18 4.52 4.28 -0.24 .429 .021 .923 .000 4.72 5.18 0.46 4.14 4.14 0.00 .493 .016 .346 .030 It is essential for learners to be able to pronounce all the individual sounds in the L2 8. L2 teachers should present grammatical rules one at a time and have learners practice before moving on to the next rule. 9. Learners' errors should be corrected as soon as they are made in order to prevent the formation of bad habits. 10. Learning a second language is easy. 4.36 4.18 -0.18 4.33 4.19 -0.14 .027 .017 .932 .000 4.36 3.91 -0.45 5.00 4.29 -0.71 .027 .153 .649 .007 4.09 4.27 0.18 5.19 4.76 -0.43 .376 .026 .243 .045 2.36 2.36 0.00 2.67 2.19 -0.48 .160 .065 .306 .035 6. 7. Results showed statistically significant differences from pre- to post-questionnaire for four Likert scale items: 2. highly intelligent people are good language learners (p = .014, ηp² = .187), 3. the best predictor of success in language learning is motivation (p = .011, ηp² = .197), 7. it is essential for learners to be able to pronounce all the individual sounds in the L2 (p = .027, ηp² = .017), and 8. L2 teachers should present grammatical rules one at a time and have learners practice before moving on to the next rule (p = .027, ηp² = .153). With the exception of the pronunciation item, which featured a small effect size, the other three items displayed a medium effect size when examining the effect of TIME. At the post-questionnaire, groups showed weakened agreement to the statements asserting the exact pronunciation of L2 sounds was essential and grammar rules should be presented one at a time. The waning emphasis on grammar and pronunciation instruction parallels Mattheoudakis (2007), which showed that strong beliefs regarding grammar and pronunciation instruction are susceptible to change as students encounter evidence to challenge preconceived notions. Both groups were more likely to agree, though, that motivation is critical in language learning. While motivation is complex, it is generally accepted as critical for successful adult SLA (Lightbrown & Spada, 2013) and both groups gained a greater understanding of its importance. The only difference to emerge between the groups within the Likert scale data was that SLLE reported declining agreement that highly intelligent people are more likely to be good language learners at the post-questionnaire (#2). CONV showed no difference on this item. This is also the only Likert item approaching significance among the interactions of TIME * GROUP (p = 0.067, ηp² = .107), although the effect size was medium. As a post-hoc test, a paired samples t-test was used to compare the pre- to post-questionnaire scores on this item. While CONV showed no difference (0.00), SLLE showed a lessening agreement (-1.34) which was statistically significant (p = .006) with a medium effect size (as calculated by Cohen’s d) of .65. This change of belief is likely connected to SLLE experiencing the challenges of learning an L2; students may have felt their ego threatened as they struggled with the task, perhaps realizing that their own intelligence only took them so far in learning an L2. 4.2. Recognizing Factors that Affect SLA To measure awareness of factors influencing SLA on the pre- and post-questionnaire, participants were provided with a list of thirteen factors known to affect language learning. Participants circled any factors they believed affected SLA and then, for those circled, ranked the factors in importance. Table 3 below shows each factor that was included along with the number of factors circled. For each group, the pre- and post-scores are included, along with the difference subtracting pre- from postquestionnaire. For the number of factors circled, a positive difference number indicates more factors recognized, while in remaining rows a positive difference indicates the factor became less important (was ranked lower in importance). In the comparisons section, results from the mixed ANOVA, including the significance and effect sizes of TIME and TIME * GROUP are included. S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 ISSN 2329-2210 Table 3 Pre- and Post- Recognition of SLA Factors by Group with Comparison Results from Mixed ANOVA Comparisons Pre- Post- Difference Pre- Post- Difference Sign. TIME ηp² TIME Sign. TIME GROUP ηp² TIME GROUP * SLLE * CONV Number of Factors Circled 7.64 8.64 1.00 5.52 7.86 2.34 .001 .322 .212 .051 Age Motivation Intelligence Learning Style Introversion/Extroversion Inhibition Anxiety Willingness to Communicate Self-Esteem Empathy Dominance Ethnic Group Affiliation Attitudes toward L2 Cultural Group 4.09 3.82 11.09 7.27 10.91 11.91 12.18 3.36 9.18 13.09 10.82 9.09 6.54 5.91 1.36 12.00 9.18 7.18 10.82 6.81 3.55 6.27 11.36 12.82 11.27 4.91 1.82 -1.46 0.91 1.91 -3.73 -1.09 -5.37 0.19 -2.91 -1.73 2.00 2.18 -1.63 3.86 3.05 9.24 5.67 10.38 13.19 11.19 7.52 11.43 14.00 13.29 11.86 9.86 3.29 2.42 12.57 6.42 8.52 12.48 8.14 5.81 6.38 12.67 12.52 11.10 6.57 -0.57 -0.63 3.33 0.75 -1.86 -0.71 -3.05 -1.71 -5.05 -1.33 -0.77 -0.76 -3.29 .762 .119 .017 .255 .007 .220 .000 .360 .000 .016 .798 .787 .006 .003 .079 .174 .043 .218 .050 .338 .028 .442 .179 .002 .002 .222 .175 .272 .256 .588 .312 .792 .271 .437 .260 .747 .080 .137 .405 .060 .040 .043 .010 .034 .002 .040 .020 .042 .004 .099 .072 .023 When examining the effect of TIME, the number of factors identified as impacting SLA increased, a statistically significant difference (p = .001), with a large effect size (ηp²=.322). Both groups became increasingly aware of the importance of five factors that affect SLA: Introversion/Extroversion (p=.007, ηp² = .218), Anxiety (p = .000, ηp² = .338), Self-Esteem (p = .000, ηp² = .442), Empathy (p = .016, ηp² = .179), and Attitudes toward the L2 Cultural Group (p = .006, ηp² = .222). Effect sizes ranged from medium for introversion/ extroversion, empathy, and attitudes towards L2 cultural group to large for anxiety and selfesteem. The increased recognition of factors suggests both groups did develop a deeper perspective on language learning. Additionally, despite differences on the Likert scale item on intelligence, both groups downgraded the importance of intelligence in the postquestionnaire (p = .017, ηp² = .174) with a medium effect size. Only one SLA factor, dominance, approached significance when exploring the interaction of TIME * GROUP (p = 0.080). While CONV found dominance to be less important on the post-test (+ 2.00) SLLE showed a greater recognition of the impact of dominance in L2 learning (-0.77). Although the result is somewhat surprising, given that dominance should be most noticeable in the context of interaction, with which CONV had greater experience, SLLE interacted with classmates through communicative activities which may have led to recognition of the role of dominance for learners in taking turns, holding the floor, and otherwise obtaining opportunities for practice. As a post-hoc test, a paired samples ttest was used to compare the pre- to postquestionnaire responses to this item. For CONV and SLLE the pre- to post-questionnaire differences were not statistically significant (p = .155 & .391, respectively), however, their opposite movement is appreciable. 4.3. Student Perceptions Despite limited differences between groups quantitatively, the qualitative analysis of the open-ended responses showed noticeable differences. When asked how the project helped further understanding of SLA, CONV focused on age as a factor, while SLLE regularly mentioned the importance of motivation. Over a third of CONV (36%) described learning the importance of age in language learning, while only 5% of SLLE noted age. In detailing how his understanding had grown, one CONV student wrote, “The older you get, the harder it becomes to develop a second language”. Age was the only factor mentioned by more than one 77 78 Comparing Experiential Approaches student in CONV. On the other hand, over a third of SLLE commented on the importance of motivation in language learning. One participant wrote, “Learning a second language helped me understand that it takes a lot of practice, motivation, and determination to succeed in learning a second language”. The only other factor mentioned by more than one SLLE participant was anxiety, commented on by two participants (10%). One participant wrote, “This project helped me understand the challenges faced by second language learners … The fear and anxiety of not being able to ‘ace’ second language learning hit me pretty hard. But, in that way, it helped me understand”. More importantly, when asked how the course changed student beliefs about L2 learning, the most common response (36%) from CONV was that it did not change beliefs. One participant wrote, “Honestly, my beliefs haven’t changed much. The course has simply solidified and refreshed me on materials regarding [second language learning].” This solidification of beliefs echoes findings of Kagan (1992) who found that students sift through received information to find support for, instead of challenges to, pre-existing notions. On the other hand, around a quarter of SLLE (24%) pointed out that, at the beginning of the semester, they thought language learning would not be particularly difficult, a belief that changed during the course. One participant said, “I honestly didn’t think that second language learning would be as hard as it really was”. Similarly, another student said, “I had read about potential challenges and how difficult it may be, but I really thought it would be a piece of cake until I actually did it myself”. Some students assumed that similarities between French and Spanish would help them. One participant wrote, “I thought it would be easy to learn French due to its similarities to Spanish. I was completely wrong. More work has to be put into it”. The SLLE was surprising to students, reproducing the goals of Weed (1993) and Washburn (2008) and did allow students to test out notions about language learning similar to Flowerdew (1998). These findings may also indicate that SLLEs have the potential to change a resistant idea identified by Mattheoudakis (2007), the relative difficulty of language learning, although no statistically significant changes were identified for the Likert item, “Learning a second language is easy”. Despite finding the SLLE project challenging, 48% of SLLE participants specifically mentioned wanting more of the project. When making recommendations for future semesters, students mentioned wanting more language days, songs, and practice. They wanted it to be “more of a main component of the course”. On the other hand, only one participant in CONV (9%) mentioned wanting to have more of the project, perhaps partially due to initial frustrations contacting and setting up meetings with their ESL partners. In describing potential changes to the project in future semesters, several participants in CONV voiced frustrations about arranging meetings. One participant recommended connecting students with their conversation partners through WhatsApp to get people in contact more easily, a strategy that had worked well for her. However, students also indicated that working to add other elements to the project may have made it more successful, such as using it to gather data about learner errors or including more reflection in class about the conversation partner sessions. The CONV responses did not align with previous studies that suggested students may find value in the chance to explain aspects of the language or may grow to value cultural diversity (Biondo Salomão, 2011; Keengwe, 2010). However, the lack of comments regarding diversity may be due to CONV partners being primarily native Spanish speakers from Mexico, partially sharing a cultural background with many of the CONV students. 5. Concluding Remarks Following theories of experiential learning, which suggest that learner growth is most likely to occur through experiences that allow students to re-examine their beliefs (Kolb, 1984), this study explored the use of experiential projects to enhance an upper-level course in language acquisition. Although previous studies have examined SLLEs as part of experiential learning for language teachers in education programs (Ellis, 2006; Flowerdew, 1998; Wright-Maley & Green, 2015), few studies have examined conversation partners (primarily Biondo Salomão, 2011 and Keengwe, 2010), and no studies were identified that worked to compare the two approaches. S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 ISSN 2329-2210 This study compared SLLE and conversation partner projects, each embedded into onecourse section, using pre- and postquestionnaires to track changes in beliefs. into a teaching methods course. Alternatively, the project could fit into a course devoted to specific language skills, such as a course in teaching L2 speaking and listening. When examining the quantitative data, CONV and SLLE responded similarly to their differing experiential projects. According to the Likert scale agreement ratings, both groups became more aware of the importance of motivation in language learning, while decreasing the emphasis put on ordered, careful grammar instruction and learners’ exact pronunciation of the L2 sound inventory. Both groups also became increasingly aware of the many factors that can affect SLA, circling on average one (CONV) to two (SLLE) more recognized factors in the post-questionnaire. Both increasingly recognized the importance of introversion/extroversion, anxiety, self-esteem, empathy, and attitudes toward the L2 cultural group. The numerous changes across groups, along with Mattheoudakis (2007) and Kavaonoz, Yüksel, and Varol (2017), further support that education can be effective in changing many pre-service teacher beliefs about language learning. Further, both projects were reasonably easy for the instructor to accomplish. For the SLLE, the researcher managed homework through an online dashboard and facilitated a limited number of in-class lessons. The researcher was able to utilize a language she had previously studied. For the conversation partners, the instructor arranged sign-up sheets to be disbursed at the intensive language institute and simply matched the volunteers to a student. However, given the challenges that arose in this project, instructors should explore additional formats for conversation partner implementation, addressing the logistics carefully to prevent frustration. Further, instructors may wish to examine other aspects of the project arrangements, such as the nature and timing of the reflections completed. However, differences emerged on two of the quantitative measures, primarily the importance of intelligence and dominance in SLA. When examining the open-ended responses, additional differences emerged. In addition to diverging foci (CONV: age, SLLE: motivation), students showed dramatic differences in perceptions of the value of the projects. Perhaps most importantly, students in CONV did not feel that the project substantially changed their views of SLA, were frustrated by the logistics of the project, and did not report wanting more of the project. On the other hand, students in SLLE found the project challenging, opening their eyes to the difficulty of learning a second language, and wanted more time and activities dedicated to the project. For teacher education programs interested in implementing more experiential learning, both projects may be useful considering both did change prospective teachers’ beliefs. However, these results suggest that SLLEs may be more suitable as part of language-teacher education as students reported valuing the SLLE project more. An SLLE project could reasonably fit into numerous courses; in addition to a language acquisition course, it may fit easily It is important to recognize the limitations of the study, however. First, it was challenging to control for time on task in the two different groups. Both groups completed the same number of readings on SLA and had roughly the same amount of time for interaction (four 50-minute sessions in which CONV met with their language partner or SLLE tried activities with classmates during course time). To try to match the time on task for SLLE, which also had significant work in Duolingo outside of class, CONV had more class time allotted to discussing the readings. In the end, the overall time on task for the SLA portion of the course was reasonably equal, but CONV had slightly less time for the experiential project. Additionally, the CONV group faced greater difficulties in getting started in their project because they depended on their conversation partner to set up meetings. This frustration affected students’ reactions to the project. Further, although rates of participation in the study across both sections were similar, the CONV group was about half the size of the SLLE due to lower enrollment in the CONV semester. Small numbers of participants in CONV may have made it more difficult to identify differing trends in the data. Finally, the composition of the groups differed in proportion of varying majors, which may have affected students’ learning goals for the course. 79 80 Comparing Experiential Approaches However, the attempts to compare two approaches identified some key differences. Future research is needed to continue to compare various forms of experiential learning. For example, researchers could compare oneon-one experiences (e.g., conversation partners and tutoring experiences) or compare various teaching experiences (e.g., tutoring versus whole class teaching practicums) to continue to find the ways that each experiential learning approach affects pre-service teacher beliefs. This information could help programs deploy pedagogical interventions to address particular student needs. Teacher education programs could, potentially, assess prospective teacher beliefs at the beginning of a program and utilize a specific experiential project (or a combination of experiential projects) to address problematic beliefs or specific needs. References Anderson-Mejias, P. L. (2005). Generation and Spanish language use in the lower Rio Grande Valley of Texas. Southwest Journal of Linguistics, 24(1), 1-12. Biondo Salomão, A. C. (2011). Collaborative language learning in Teletandem: A resource for pre-service teacher education. Profile Issues in Teachers Professional Development, 13(1), 139156. Blume, C., Gerlach, D., Roters, B., & Schmidt, T. (2019). The ABCs of inclusive English teacher education: A quantitative and qualitative study examining the attitudes, beliefs and (reflective) competence of pre-service foreign language teachers. TESL-EJ, 22(4), 1-18. Borg, S. (2015). Teacher cognition and language education (2nd ed.). New York, NY: Bloomsbury Academic. Borg, S. (2011). The impact of in-service teacher education on language teachers' beliefs. System, 39(3), 370-380. Borg, S. (2009). Language teacher cognition. In A. Burns, & J. C. Richards (Eds.), The Cambridge guide to second language teacher education (pp. 163-171). New York, NY: Cambridge University Press. Borg, S. (2003). Teacher cognition in language teaching: A review of research on what language teachers think, know, believe, do. Language Teaching, 36, 81-109. Brown, H. D. (2014). Principles of language learning and teaching: A course in L2 acquisition (6th ed.). White Plains, NY: Pearson. Busch, D. (2010). Pre-service teacher beliefs about language learning: The second language acquisition course as an agent for change. Language Teaching Research, 14(3), 318-337. Crandall, J. (2000). Language teacher education. Annual Review of Applied Linguistics, 20, 34-55. Cunningham, K. J. (2015). Duolingo [Review]. TESL-EJ, 19(1), 1-9. Ellis, E. M. (2016)."I may be a native speaker but I'm not monolingual": Reimagining all teachers' linguistic identities in TESOL. TESOL Quarterly, 50(3), 597630. Ellis, E. M. (2006). Language learning experiences as a contributor to ESOL teacher cognition. TESL-EJ, 10(1), 1-20. Ellis, E. M. (2004). The invisible multilingual teacher: The contribution of language background to Australian ESL teachers’ professional knowledge and beliefs. The International Journal of Multilingualism, 1(2), 90-108. Farrell, T. S. C. (2018). Reflective language teaching: Practical applications for TESOL teachers (2nd ed.). New York, NY: Bloomsbury Academic. Farrell, T. S. C. (2006). The teacher is an octopus: Uncovering pre-service language teachers’ beliefs through metaphor analysis. RELC Journal, 37(2), 326–248. Flowerdew, J. (1998). Language learning experience in L2 teacher education. TESOL Quarterly, 32(3), 529 - 535. Horwitz, E. K. (1985). Using student beliefs about language learning and teaching in the foreign language methods course. Foreign Language Annals, 18(4), 333340. Kagan, D. M. (1992). Implications of research on teacher belief. Educational Psychologist, 27(1), 65-90. Kavanoz, S., Yüksel, H. G., & Varol, B. (2017). Evolvement of pre-service language teachers’ beliefs through teacher education. International Journal of Progressive Education, 13(1), 119-135. Keengwe, J. (2010). Fostering cross cultural competence in preservice teachers through multicultural education S. McCrocklin/ International Journal of Society, Culture & Language, 8(1), 2020 experiences. Early Childhood Education Journal, 38(3), 197-204. Kolb, A.Y., & Kolb, D. A. (2005). Learning styles and learning spaces: enhancing experiential learning in higher education. Academy of Management Learning & Education, 4(2), 193-212. Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice-Hall. Lightbrown, P., & Spada, N. (2013). How languages are learned (4th ed.). Oxford, England: Oxford University Press. Matic, M. (2011). The English language learning experience and its impact on the future English language teachers’ career. Procedia Social and Behavioral Sciences, 11, 132-135. Mattheoudakis, M. (2007). Tracking changes in pre-service EFL teacher beliefs in Greece: A longitudinal study. Teaching and Teacher Education, 23, 1272-1288. Mighani, M. G., & Moghadam, M. Y. (2019). Building intercultural sensitivity in preservice EFL teachers through interactive culture-focused speaking tasks. International Journal of Society, Culture, & Language, 7(2), 27-39. Moodie, I. (2016). The anti-apprenticeship of observation: How negative prior language learning experience influences English language teachers’ beliefs and practices. System, 60, 29-41. Moon, J. (2004). A handbook of reflective and experiential learning: Theory and practice. London, England: Routledge Farmer. Ortega, L. (2008). Understanding L2 acquisition. London, England: Hodder Education. Oxford, R. L. (2002). Language learning strategies in a nutshell: Update and ESL suggestions. In J. C. Richards, & W. A. Renandyn (Eds.), Methodology in language teaching: An anthology of current practice (pp. 124-132). Cambridge, England: Cambridge University Press. ISSN 2329-2210 Senior, R. (2006). The experience of language teaching. Cambridge, England: Cambridge University Press. Tatto, M. T. (1998). The influence of teacher education on teachers’ beliefs about purposes of education, roles, and practice. Journal of Teacher Education, 49(1), 66-77. Vesselinov, R., & Grego, J. (2012). Duolingo effectiveness study: A final report. Retrieved from: https://s3.amazonaws. com/duolingo-papers/other/vesselinovgrego.duolingo12.pdf. Washburn, G. N. (2008). Alone, confused, and frustrated: Developing empathy and strategies for working with English language learners. The Clearing House, 81(6), 247-250. Weed, K.Z. (1993). Teacher awareness of the language learners’ situation. In D. Freeman & S. Cornwell (Eds.), New ways in teacher education (pp. 191-141). Alexandria, VA: TESOL Publications. Wideen, M., Mayer-Smith, J., & Moon, B. (1998). A critical analysis of the research on learning to teach: Making the case for an ecological perspective on inquiry. Review of Educational Research, 68(2), 130–178. Wright, T. (2010). L2 teacher education: Review of recent research on practice. Language Teaching, 43(3), 259-296. Wright-Maley, C., & Green, J.D. (2015). Experiencing the needs and challenges of ELLs: Improving knowledge and efficacy of pre-service teachers through the use of a language immersion simulation. Cogent Education, 2(1), 1-17. Wubbels, T. (1992). Taking account of student teachers’ preconceptions. Teaching and Teacher Education, 8(2), 137-149. 81