Pre‐Service Teacher Attitudes Toward English Language Learners

Abstract Enrollment of English language learners (ELLs) in United States schools has increased more rapidly than general enrollment, resulting in an urgent need for teachers to be prepared and to meet the needs of ELLs in their classrooms. This study examined the language learning beliefs and attitudes toward ELLs among 286 pre-service teachers in a large public university. Results show that pre-service teachers demonstrated neutral attitudes toward ELLs. Participants’ language learning beliefs exhibited four components: “difficulty of learning,” “ease of learning,” “motivation” and “traditional view” of language learning. Controlling for language learning status, pre–service teachers tended to demonstrate less supportive attitudes toward ELLs when they held traditional views of language learning and when they viewed language learning (in general and English in particular) as relatively easy or not difficult. These findings suggest preservice teachers need evidence-based coursework in language development and language learning processes to overcome potential misconceptions and biases regarding ELLs.

when they begin believing and behaving in ways that are consistent with their teachers' expectations (Jussim, 1986). In addition, teachers who misunderstand the academic difficulties of their students from culturally and linguistically diverse backgrounds may over-identify students for special education programs (Chamberlain, 2005).
In the current study, we examine pre-service teacher beliefs about second language learning and relate them to their attitudes toward ELLs through MANOVA and hierarchical regression of survey data from a university preparation program that graduates the most teachers in the state. Specifically, we sought to examine the following research questions: 1. What are pre-service teachers' beliefs about second language learning and how supportive are they of ELLs?
2. Do pre-service teachers' demographic, educational or language background affect their beliefs about second language learning or attitudes toward ELLs?
3. How do pre-service teachers' beliefs about second language learning relate to their attitudes regarding ELLs, over and above demographic characteristics?

Literature Review
In a synthesis of the literature on teachers' beliefs, Pajares (1992) found that teachers' beliefs and attitudes affect their instructional decisions, especially in terms of defining tasks (including interpreting, planning and making general decisions), which in turn influences their classroom practices. For the purposes of the current study, teachers' language learning beliefs refer to the participants' views about the nature of learning and teaching a second language (Busch, 2010). The importance of language learning beliefs lies in their influence on "what and how students learn" (Wong, 2010, p.123). According to Karabenick and Noda (2004), teachers' beliefs "can impede attempting new instructional practices that are more conducive to ELL PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 5 students" (p. 56). In addition, teachers' lack of understanding of the second language learning process can lead to inappropriate referral of ELLs for special education services (Hardin, Mereoiu, Hung, & Roach-Scott, 2009).

Teachers' Language Learning Attitudes
Teachers' attitudes matter. Teachers who maintain a positive attitude generate a warm and supportive environment for students from culturally and linguistically diverse backgrounds (e.g., Walker, Shafer, & Iiams, 2004). In addition, in-service teachers' attitudes towards ELLs influence whether teachers welcome professional development on teaching ELLs (Karabenick & Noda, 2004).
Very few studies, however, have examined pre-service teachers' attitudes toward and beliefs about ELLs. Two studies with very small non-random samples found generally positive attitudes towards teaching students from diverse backgrounds and respecting them as individuals, but these results cannot be generalized to the larger population of pre-service teachers (Artiles & McClafferty, 1998;McDiarmid & Price, 1990). Only two studies examined pre-service general education teachers' attitudes toward ELLs specifically (Durgunoğlu & Hughes, 2010;Vélez Salas, Flores, & Smith, 2005). Durgunoğlu and Hughes (2010) found that 62 pre-service teachers in a medium-sized university in the US had slightly positive attitudes, on average, toward ELLs and neutral attitudes regarding their own preparedness and efficacy in teaching ELLs. Although these pre-service teachers had already completed two diversity courses and a minimum of 60 volunteer hours in a diverse classroom setting, Durgunoğlu and Hughes (2010) also found that pre-service teachers with lower knowledge about how to teach ELLs showed more negative attitudes. Similarly, Vélez Salas et al. (2005) found that 518 pre-service teachers in a large Hispanic-serving university in South Texas generally reported neutral attitudes toward PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 6 ELLs prior to taking any multicultural coursework. Vélez Salas et al. (2005) also found that younger, Mexican-American, bilingual and San Antonio-born pre-service teachers were more likely to hold more positive attitudes toward ELLs and language diversity.
Factors influencing teachers' attitudes toward ELLs. Byrnes and Kiger (1994) found that teachers' attitudes toward ELLs in their classrooms were comprised of three factors: language politics, ELL intolerance, and language support. Teachers' scores on these three factors were statistically significantly correlated with their reported willingness to include an ELL student in their classroom. Common societal attitudes about language, such as "English should be the sole language of government" and "immigrants must speak English to be considered American," make up the language politics factor. The ELL intolerance factor included items such as, "Having an ELL in the classroom is detrimental to the other students' learning" and "ELL students often used unjustified claims of discrimination as an excuse for not doing well in school." Language support was signaled by items such as "The government should spend additional money to provide better programs for ELLs" and "Regular-classroom teachers should be required to receive training to meet the needs of ELLs." Using the same instrument, Vélez Salas et al. (2005) confirmed Byrnes and Kiger's (1994) language politics and language support factors, but also found two others, language motivation and instructional expectations, among their sample of pre-service teachers in south Texas. Byrnes and Kiger (1994) found that the more a teacher favored English as the language of government and American identity, felt ELL students' inclusion in general education classes distracted from non-ELL students' learning, and felt additional money or teacher professional development was not needed for better ELL programs, the less the teacher wanted to teach an ELL in her or his classroom.  Researchers have identified many other predictors of teachers' attitudes toward ELL students among general education teachers. Teachers with the following experiences were found to demonstrate more positive attitudes toward ELL students than those who did not have or had less of the experience: earning a graduate degree (Byrnes, Kiger & Manning, 1997), coursework related to cultural and linguistic diversity or specifically to teaching ELLs (Byrnes et al., 1997;Flores & Smith, 2009;Youngs & Youngs, 2001), direct personal contact with diverse cultures (Youngs & Youngs, 2001), and more experience teaching ELLs (Byrnes et al., 1997;Flores & Smith, 2009;Youngs & Youngs, 2001).
Directly assessing teachers' knowledge, Durgunoğlu and Hughes (2010) found that teachers with strong knowledge about concepts and techniques in teaching ELLs reported more favorable attitudes toward ELL students than those with weaker knowledge. In addition, demographic characteristics such as gender (Youngs & Youngs, 2001), age (Vélez Salas et al., 2005) and ethnicity (Flores & Smith, 2009;Vélez Salas et al., 2005) influenced attitudes, with younger, female or Latino teachers generally having more positive attitudes toward ELLs than older, male or non-Latino teachers.
Preparing teachers for a diverse student population through coursework does not always seem to result in more positive attitudes toward learners from culturally and linguistically diverse backgrounds. Irvine (2003), in summarizing previous studies, found that teacher education programs, overall, do not change pre-service teachers' beliefs and that superficial coursework regarding diversity can actually result in more stereotypic beliefs or negative attitudes.
Similarly, Garcia and Guerra (2004) and McDiarmid and Price (1990) found that multicultural coursework did not change teachers' negative attitudes or beliefs. In addition, secondary social studies teachers in a large school district in Florida found the university preparation and school PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 8 district professional development provided to them to include ELLs in their social studies classes was inadequate (O'Brien, 2011).
Coursework, of course, can vary in effectiveness. Other researchers have found positive effects of coursework (Bernhard, Diaz, & Allgood, 2006;DeCastro-Ambrosetti & Cho, 2005;Friend, Most, & McCreary, 2009;Karathanos, 2009;Pappamihiel, 2007). DeCastro-Ambrosetti and Cho (2005), in a pilot study, found that pre-service teachers who took courses with diversity issues embedded in them improved their attitudes toward teaching learners from culturally diverse backgrounds. Other positive benefits of professional development included more use of ELLs' L1 in the classroom (Karathanos, 2009) and improved ELL achievement in classrooms of teachers who participated in a two-year professional development course focused on strategies to teach ELLs (Friend et al., 2009).
Teachers' own language learning background may also affect their attitudes toward ELLs, choice of teaching strategies and feelings of empathy toward their ELL students (Coady, Harper, & de Jong, 2011;Ellis, 2004;Vélez Salas et al., 2005). Vélez Salas et al. (2005) found that pre-service teachers who reported themselves as bilingual showed more supportive attitudes toward ELLs than those who considered themselves monolingual. Ellis (2004) found that English as a second language (ESL) teachers in Australia with some language learning experience showed empathy toward their ELL students, whereas teachers who were themselves highly competent bilinguals chose different teaching strategies and used their own language learning experience to help support and inspire their ELL students. Coady et al. (2011) found that teachers in Florida with at least an intermediate level of proficiency in another language rated themselves as better prepared to meet the needs of ELLs. Teachers' beliefs specifically concerning language learning may influence how they work with ELLs. The length of time required to learn a second language is one major area that has been examined. Reeves (2006) found that in-service secondary subject-area teachers in a lowincidence ELL district with no prior preparation in teaching ELLs believed two years to be sufficient for ELLs to acquire the English language. Furthermore, in a qualitative study of six pre-service teachers who were working one-on-one with middle-school ELLs, Bunten (2010) found that the pre-service teachers expressed surprise that ELLs who had been in US public schools since kindergarten were still struggling with writing in English. Initially, the majority of participants in two studies (Busch, 2010;Wong, 2010) believed that it takes from 3 to 5 years to master a second language. However, the stability of these beliefs over time seems to be contradictory between these two studies. The participants in Busch's (2010) study, after taking a course in second language acquisition, changed their belief by increasing the amount of time (5 to 10 years) it should take to learn a second language. In contrast, the pre-service teachers surveyed by Wong (2010), after a fourteen-month period of teacher preparation, decided that learning a language did not require that much time. Pre-service teachers in Busch's (2010) study also reported believing more strongly that some languages are more difficult to learn than others, and that English is a more difficult language, than they did before the class.

Teachers' Language Learning Beliefs
Another central belief related to language leaning is how students best learn language.
Pre-service teachers considered traditional repeating and practicing drills to be effective strategies in language learning (Busch, 2010;Wong, 2010), although some discovered over time that language is more complex than that, and that other strategies were also required (Busch, 2010). Pray & Marx (2010) explored pre-service teachers' attitudes toward the language in PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 10 which instruction is to be conducted. Most participants agreed that students learn English better if they are taught only in English.
Pre-service teachers may also differ in what they believe the role of correction is in language learning. Most pre-service teachers did not believe it to be right that language learners be corrected every time they made a mistake, and that there should be room allowed for making mistakes (Busch, 2010;Pray & Marx, 2010;Wong, 2010). Pre-service English teachers who were in their third or fourth year of a four-year teacher preparation course in Turkey were less likely to believe in the importance of immediate correction of student errors than students in their first or second years of the program (Inozu, 2011).
As schooling systems continue to see increasing enrollment of learners from linguistically diverse backgrounds, Walker et al. (2004) have raised an alarm that teacher attitudes toward ELLs "will significantly deteriorate" (p. 132). This deterioration in teacher attitudes may also be exacerbated by the lack of preparation of general education teachers to teach linguistically diverse students, and the accountability of teachers for the academic achievement of ELLs, among other factors. In response to this need and concern, the primary purpose of this study is to examine the relations between pre-service teachers' beliefs about second language learning and their views of ELLs. Teacher attitudes and expectations of linguistically and culturally diverse student populations have implications that may influence students' performance. Thus, it is essential that this issue be addressed.
In the current study, we hypothesize that pre-service teachers who believe in the importance of speaking correctly, correcting ELLs' errors, and that language learning is primarily a matter of memorizing vocabulary and grammar rules, will demonstrate less supportive attitudes toward ELLs as well. We also hypothesize that pre-service teachers who PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 11 endorse beliefs about language learning that are more traditional (e.g., repeating and practicing drills) are also likely to report less supportive views of ELLs.
Because previous studies found that pre-service teachers' language backgrounds (e.g., whether they learned a second language) is related to their beliefs about second language learning, we hypothesize that teachers with more language learning experience themselves will be more supportive of ELLs.

Method
A correlational research design was adopted to examine the relationship between preservice teachers' beliefs about second language learning and attitudes toward ELLs.

Participants
Participants were 286 undergraduate students (84% junior-level and 16% senior-level) who were enrolled in the teacher education program in one of the largest public universities in the US, located in a mid-size town dominated by the university. These pre-service teachers may expect to teach in urban, suburban or rural contexts, with growing numbers of ELLs in all settings. The majority of participants were female (98%), in the age groups of 18 to 21 years (83%), 22 to 25 years (15%), and 26-30 years (2%). Participants consisted of 87% White Non-Hispanic, 11% Hispanic, and 2% African-American (Non-Hispanic). At the time the study began, the program had over 1,000 undergraduate pre-service teachers, with 88% of them White Non-Hispanic, 8% Hispanic and 2% African-American (Non-Hispanic).

Procedures
Students enrolled in the teacher education program taking classes regarding language development or language assessment (N=561) were invited via electronic mail to participate in an anonymous, online survey that took about 15 minutes to complete. Participation was voluntary without incentives or compensation. Response rate was 51 percent.

Measures
Through the online survey, pre-service teachers reported on their demographic information and their views and beliefs about teaching and learning learners from non-native English-speaking backgrounds. Each section of the survey is described in more detail below.
Demographic or background information. Pre-service teachers provided information on their gender, age, grade or year in college, ethnicity, how many credit hours taken in teacher education coursework, how many credit hours taken in ESL coursework, and whether they ever learned a second language and whether they can speak a second language fluently.
Supportive views of ELLs. Pre-service teachers rated 12 items regarding their views of ELLs (see Appendix). These items were adapted from a survey published by the National Center for Research on Teacher Learning (Michigan State University) to assess pre-service teachers' beliefs about students from culturally diverse backgrounds (McDiarmid & Price, 1990). Items were rated using a 7-point scale (1=strongly disagree, 7 = strongly agree), and all items were reverse scored because original items reflected negative and unsupportive views of ELLs. Items were averaged (after reversing scores) to compute a composite for supportive views of ELLs, reflecting views that supported and acknowledged the needs of ELLs.
Beliefs about second language learning. Participants rated 38 items (α = .92) on a 7point scale (1=strongly disagree, 7 = strongly agree) adapted from the teacher version of the Beliefs about Language Learning Inventory or BALLI (Horwitz, 1985). The BALLI has been a principal tool for research on beliefs about (second) language learning, and has been used in a PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 13 variety of countries including the US, Canada, Hong Kong, Taiwan, Korea, and Malaysia (Nikitina & Furuoka, 2006;Peacock, 2001).
Prior research typically has found that the BALLI (Nikitina & Furuoka, 2006) consists of 4 factors (i.e., difficulty of learning, ease of learning, motivation, and traditional view), and we conducted principal components analysis (PCA) to reveal the internal structure of the data on the 38 items of the BALLI. An initial principal components analysis was conducted and any items with loadings less than.50 on the four components were eliminated from further analyses, using a conservative and robust rule-of-thumb that factor loadings over .50 are classified as "strong" item loadings (Zwick & Velicer, 1986). Principal components analysis was conducted again with the remaining 18 items on four components (see Table 1). Corresponding items for each of the four components were averaged to compute composites for "difficulty of learning" (α = .77), "ease of learning" (α = .75), "motivation" (α = .67), and "traditional view of language learning" (α = .69). Because results from our principal components analysis differed slightly from previous studies, we considered the items that belonged to each of the components in our analysis when labeling the components to most accurately reflect that construct. Importantly, there was empirical evidence showing that all items were "strong" items for their respective factors.

Plan of Analyses
Descriptive and preliminary analyses were first conducted, and differences in participants' demographic or background variables on the major variables of this study were examined. Recall that demographic or background variables included participants' gender, age, grade or year in college, ethnicity, how many credit hours taken in teacher education PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 14 coursework, how many credit hours taken in ESL coursework, whether they speak a second language, and whether they speak a second language fluently. Correlational analyses were then conducted to examine relations between specific components of pre-service teachers' beliefs about learning language and their views of ELLs. Finally, hierarchical multiple regression was conducted to examine the unique contributions of specific components of pre-service teachers' beliefs about learning language on their views of ELLs.

Preliminary Analyses
Descriptive statistics were conducted and the means and standard deviations for the major variables are presented in Table 2. Overall, the pre-service teachers' view toward ELLs was neutral to slightly positive. Of the four factors on beliefs of second language learning, "motivation" was highest (M= 6.27, SD = 1.58) and "traditional view" was lowest (M = 4.05, SD = 1.61). Major variables that were included in the correlational analysis were first screened for normality and outliers. None of the major variables were skewed according to the cutoff values of two for skewness and seven for kurtosis (West, Finch, & Curran, 1995), and no outliers were detected based on the frequencies and distribution of the major variables (Barnett & Lewis, 1994).

Relations of Participants' Demographic and Background Variables to Major Variables
Participants' demographic, educational, and language background variables were examined in relation to the major variables. Significant background variables were then included as a covariate when conducting correlational analyses to reduce the likelihood that omitted third variables would account for the observed relations between study variables.

Gender, age, and ethnicity.
To examine if there were statistically significant differences in major variables depending on participants' gender, age, (coded as between 18 to PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 15 21 years, 22 to 25 years, and 26 to 30 years), and ethnicity (coded as White Non-Hispanic and Hispanic/African American), three separate single-factor (gender, age, or ethnicity) multivariate analyses of variance were conducted on the 5 major variables (i.e., supportive views of ELLs, difficulty of learning, ease of learning, motivation, and traditional view of language learning).
No significant differences on the major variables were found for gender, Wilks's F (5,280) = 1.30, ns, or for age, Wilks's F (10,556) = 1.64, ns. However, differences on major variables were found for ethnicity, Wilks's F (5,280) = 3.48, p < .01. Univariate results indicate that White Non-Hispanic pre-service teachers perceived language learning as more difficult and also endorsed lower ratings on supportive views of ELLs than Hispanic/African American pre-service teachers, Fs (1,284) = 4.79 and 4.14, ps < .05, respectively.

Educational background variables.
To examine if there were statistically significant differences in major variables depending on participants' educational background variables, consisting of participants' year in college (coded as junior-or senior-level), number of credit hours taken in teacher education coursework (coded as none, a little, or a lot), and number of credit hours taken in ESL coursework (coded as none, a little or a lot). Three separate singlefactor (one for each of the educational background variables) multivariate analyses of variance (MANOVA) were conducted on the 5 major variables (i.e., supportive views of ELLs, difficulty of learning, ease of learning, motivation, and traditional view of language learning). No significant differences were found for the educational background variables of year in college, number of credit hours in teacher education coursework, and number of credit hours in ESL coursework, Wilks's Fs (5,279) (5,127) and (10,254) = .57, 1.31, and .48, ns, respectively. Language background. To examine if there were statistically significant differences in major variables depending on participants' language backgrounds (coded as never learned a PRE-SERVICE TEACHERS' ATTITUDES AND ELLS 16 second language, learned a second language, and fluent in a second language). A single-factor multivariate analyses of variance (MANOVA) was conducted on the 5 major variables (i.e., supportive views of ELLs, difficulty of learning, ease of learning, motivation, and traditional view of language learning). Significant differences were found for language background, Wilks's F (10,556) = 3.33, p < .01. Univariate effects indicated language background differences in difficulty of learning and motivation in language background, Fs (2,282) = 3.62 and 4.88, ps <.05 and .01, respectively. Post-hoc comparisons were conducted and results indicate that all three language background status groups differed from one another on perceptions of difficulty and on motivation in language learning. Specifically, those who were fluent in a second language perceived language learning as the least difficult and learning a second language required the most motivation (Ms = 5.84 and 7.00, respectively), whereas those who never learned a second language perceived language learning as the most difficult but also believed language learning required the least motivation to master (Ms = 6.15 and 5.82, respectively). Those who learned (but were not fluent in) a second language viewed second language learning as more difficult and requiring less motivation to master relative to those who were fluent, but as less difficult and requiring more motivation to master relative to those who never learned a second language (Ms = 6.04 and 6.31, respectively).

Correlational Analyses
Because ethnicity and language background variables were associated with the major variables, partial correlations were conducted controlling for participants' ethnic and language backgrounds (see Table 3). There was coherence among all four components of beliefs about second language learning, with all components positively related to one another. In regards to relations between beliefs about second language learning and supportive views of ELLs, all four components were negatively related to supportive views of ELLs. In other words, pre-service teachers who believed that language learning is overly difficult, extremely easy, requiring high degree motivation, or that correctness, vocabulary and grammar are key to language learning were also those who were less likely to hold supportive views of ELLs. These types of beliefs of language learning are exaggerated and rigid and are not student-centered.

Predicting Views of ELLs from Beliefs About Second Language Learning
Hierarchical multiple regression was conducted to predict participants' supportive views of ELLs from specific components of their beliefs about second language learning. Because participants' ethnic and language backgrounds were associated with their beliefs about second language learning, ethnic and language background variables were entered in the first step of the regression analysis (as covariates) with the four components of pre-service teachers' beliefs about second language learning (i.e., difficulty of learning, ease of learning, motivation, and traditional view) entered in the second step. Results indicated that "ease of learning" and "traditional view" were the components that have unique and additive contributions to views of ELLs, above and beyond contributions from participants' ethnic and language backgrounds (see Table 4).

Pre-service Teachers' Attitudes Toward ELLs
On average, the pre-service teachers we surveyed showed neutral to somewhat positive attitudes toward supporting ELLs. Our findings are similar to those of Durgunoğlu and Hughes (2010), who found slightly positive attitudes toward ELLs among their sample of pre-service teachers in the midwestern U.S. The overall attitudes also match those found by Vélez Salas et al. (2005) in their sample of pre-service teachers in south Texas. The attitudes found in our study, however, are less positive than those of the pre-service teachers surveyed by Artiles and McClafferty (1998) and McDiarmid and Price (1990). These differences may reflect real differences in attitudes among different populations of pre-service teachers, or they may simply be a product of biased samples, because these studies and ours employed non-random samples.

Beliefs About Second Language Learning and Attitudes Toward ELLs
Unlike Youngs and Youngs (2001) and Vélez Salas et al. (2005), we found no significant differences in teachers' attitudes toward ELLs based on gender or age. However, attitudes toward ELLs differed depending on pre-service teachers' ethnic backgrounds, with Hispanic and African American pre-service teachers endorsing more supportive views of ELLs than White Non-Hispanic pre-service teachers, which was similar to the finding of Vélez Salas and colleagues(2005) that Mexican-American pre-service teachers held more supportive views of ELLs than their non-Mexican-American (mostly White) classmates.
Further, language background mattered in how pre-service teachers viewed language learning. Pre-service teachers who studied a second language and mastered it fluently did not believe language learning to be an impossible task (i.e., lowest on difficulty of learning) but also believed that language learning required the most motivation, whereas those who never learned a second language perceived language learning as nearly impossible (i.e., highest on difficulty of learning) but also believed that language learning required the least motivation. Pre-service teachers who studied (but were not fluent in) a second language viewed second language learning as moderately difficult and requiring some degree of motivation to master. Our findings seem consistent with Vélez Salas and colleagues ' (2005) findings that bilingual pre-service teachers were more supportive of ELLs than monolingual pre-service teachers. In addition, our findings generally agree with Ellis's (2004) finding that even some language learning made teachers more sympathetic toward the challenges facing ELLs in terms of the effort and the motivation required to learn or master a second language. Combined with the finding of Coady et al. (2011) that teachers with at least an intermediate level of proficiency in another language were better prepared to teach ELLs, our finding suggests that having no experience with foreign language learning or taking a small amount of a foreign language may not help pre-service teachers gain sympathy for ELLs, whereas gaining intermediate or advanced proficiency in a language other than English may help.

The Relationship Between Pre-Service Teachers' Beliefs About Second Language Learning and Their Attitudes Regarding ELLs
No other studies to date have explored the relationship between pre-service teachers' language learning beliefs and their supportive attitudes toward ELLs. Our findings suggest that certain language learning beliefs do influence pre-service teachers' attitudes toward ELLs.
Specifically, pre-service teachers who hold a traditional view of language learning -that correctness is important and that vocabulary and grammar rules are key -also tend to hold less supportive attitudes toward ELLs, controlling for pre-service teachers' ethnic and language backgrounds and other components of language learning beliefs. These traditional views may indicate a lack of appreciation for the complexity of language learning and for the difficulty language learners encounter in formulating a "correct" product in every communicative attempt.
If teachers overemphasize correctness in language use, it may discourage ELLs from using the English they do know, which can in turn depress additional acquisition of English due to a lack of interaction in the language (Lightbown & Spada, 2006).  In addition, we found that pre-service teachers who believe languages in general, and English in particular, to be fairly easy, also tend to be less supportive of ELLs, controlling for their ethnic and language learning backgrounds and other components of language learning beliefs. Participants who scored high on the "ease of learning" component saw language learning in general, and English in particular, to be easy to moderately difficult. Pre-service teachers who scored high on "ease of learning" believed learning a new language would take less than 3 years or 3 to 5 years, despite research findings that many ELLs take much more time than that to achieve academically in English at grade level (Hakuta, 2011). Together with the findings regarding those holding a "traditional view" of language learning, these findings suggest that pre-service teachers may require more preparation regarding the difficulty and complexity of language learning, particularly the length of time it can take ELLs to master English at levels comparable to their native-speaking peers. The students in our study were enrolled in one of two second language acquisition courses, but had not yet completed both.

Limitations and Future Research
The participants in this study were recruited from the same university. Although this university produces a large percentage of the new teachers in its state, these pre-service teachers' attitudes and beliefs may not be generalizable to other institutions in that state or in other states.
With only around half of the invited students completing the questionnaire, there could be substantial differences in the attitudes and beliefs of the pre-service teachers who chose to complete the survey compared to those who did not. In addition, the measures of participants' attitudes toward ELLs and their language learning beliefs were all based on self-report. Future studies could use naturalistic designs to include observations of pre-service teachers during in field placements and student-teacher assignments to measure more directly their level of supportiveness or willingness to adapt their curriculum and/or pedagogy to meet the needs of the actual ELLs in their classrooms. Additionally, use of qualitative methods such as in-depth interviews could provide rich insight into pre-service teachers' attitudes and beliefs about ELLs and best practices for language learning. To better understand the direction of influence between teacher beliefs and attitudes and teacher practices, a longitudinal study would be able to illuminate whether beliefs and attitudes are subject to change over the course of teacher preparation, and whether these beliefs and attitudes translate into classroom practices and ultimately achievement (or lack thereof) of ELLs in participants' classrooms, once they are hired as new teachers. Following pre-service teachers over time as teacher candidates and then several years into their teaching careers could also indicate whether beliefs about second language learning and attitudes toward ELLs change with actual teaching experience.

Implications
Our study's results suggest that pre-service teachers must be presented with information on how long it takes ELLs to become proficient in academic English, as well as the difficulties of learning a second language and specific strategies for addressing these challenges. If additional preparation can change pre-service teachers' views on language learning, it may also help them become more supportive of ELLs in their classrooms. Ambe (2006) contends that the "transformation" of attitudes is not a simple process that only requires "recognizing or respecting diversity" (p. 694), but that teacher preparation programs must create inclusive environments and adopt new pedagogies that truly embrace multiple perspectives and experiences. Teacher education and preparation programs must necessarily engage the ideological structures that reproduce common attitudes in order to create this transformational shift for ELL students (Garcia & Guerra, 2004).  In conclusion, our study demonstrates that teachers' views of ELLs and their beliefs about second language learning have influences upon one another. The extent to which preservice teachers believe that language learning is relatively easy and the extent to which they hold traditional (and perhaps outdated) views of language learning predicted less supportive views of ELLs. Findings have implications for teacher education and professional development because pre-service teachers who spoke another language fluently were more supportive of ELLs. Thus, offering opportunities for pre-service teachers to learn more than one language may allow pre-service teachers to understand better the challenges that ELLs may experience and may therefore increase their empathy for ELLs. In addition, future research may explore the effects of having been an ELL and/or an immigrant oneself on teachers' attitudes and supportiveness toward ELLs. English is a very difficult or difficult language. .74 It is easier for someone who already speaks a second language to learn another one. .56 If someone spent one hour a day learning a second language, it would take him or her 5 years or more to become fluent. .64 It is important to repeat and practice a lot.
.50 I feel self-conscious speaking a second language in front of other people. .67 Learning a second language is different from learning other school subjects. .49 People who speak more than one language well are very intelligent. .55 English is a language of medium difficulty. .61 English is an easy or very easy language. .73 If someone spent one hour a day learning a second language, it would take him or her 2 years or less to become fluent. .72 If someone spent one hour a day learning a second language, it would take him or her 3 to 5 years to become fluent. .69 Females are better than males at learning second languages. .50 If I want, I believe that I will ultimately learn to speak a foreign language very well. .84 Everyone can learn to speak a second language. .74 You shouldn't say anything in a second language until you can say it correctly. .76 Learning a second language is mostly a matter of learning a lot of new vocabulary words. .72 If you are allowed to make mistakes in the beginning it will be hard to get rid of them later on. .61 Learning a second language is mostly a matter of learning a lot of grammar rules. .64