View in EDS HTML Full Text PDF Full Text

Linguistic Performance of Dual Language Immersion Students

Saved in:
Bibliographic Details
Title: Linguistic Performance of Dual Language Immersion Students
Language: English
Authors: Watzinger-Tharp, Johanna, Rubio, Fernando, Tharp, Douglas S.
Source: Foreign Language Annals. Fall 2018 51(3):575-595.
Availability: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
Peer Reviewed: Y
Page Count: 21
Publication Date: 2018
Document Type: Journal Articles
Reports - Research
Education Level: Grade 6
Descriptors: Immersion Programs, Bilingual Education Programs, Grade 6, Chinese, Language Proficiency, Longitudinal Studies, French, Spanish, Language Tests, Program Evaluation, Case Studies, Interpersonal Communication, Linguistic Performance, Listening Skills, Reading Skills, Oral Language, Second Language Learning, Teaching Methods, Educational Policy, Language of Instruction, Second Language Instruction, Native Language
DOI: 10.1111/flan.12354
ISSN: 0015-718X
Abstract: The assessment of dual language immersion (DLI) students' proficiency in the partner language plays a key role in program evaluation. This article reports findings from a large-scale study, drawing on cross-sectional assessment data from 73 schools and longitudinal data from 25 schools. Students were tested in the interpretive (listening and reading) and interpersonal (listening/speaking) modes of communication, using the ACTFL Assessment of Proficiency Towards Performance in Languages. The cross-sectional analysis shows that French and Spanish students performed at Intermediate Low across skills in fourth or fifth grade and Intermediate Mid in sixth grade. Sixth-grade Chinese DLI students attained near Intermediate Mid in listening and Novice High to Intermediate Low in reading and speaking. Across languages, students achieved the highest ratings in listening and, overall, performed more strongly in listening and reading than speaking. A regression analysis of students' change scores over time revealed significant progress in nearly all tested skills and languages. Baseline scores and partner language (Chinese, French, or Spanish) accounted for significant variation in growth rates; between-school differences were found not to be significant. The article discusses implications for assessment instruments and DLI policy and also addresses some instructional implications.
Abstractor: As Provided
Entry Date: 2018
Accession Number: EJ1189992
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGYhhu7DMcGhbOHeK2CcLHTAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDHNMRzJKZ57FKk7kjwIBEICBm9y6SA_2sUbXrYIzmn-gN3dnMFS3ugP1GOSRxY-NFgCCHDIDc-OpuR_a8xklmv5eVpAVGalGTSCN5DIMRvqWFNDGGi7JGhq-z2duDHZMyfDHNvBJA3wSLtouBssWd2wYhrxjBCRpIpX5OEXcLZzf_mimX1t2yIEq9aLUry1NpD_Ny3XQPwQEbam9F2Kik7U1Qdn2VXO53c-OVee_
Text:
  Availability: 1
  Value: <anid>AN0131600304;fla01sep.18;2018Sep06.10:47;v2.2.500</anid> <title id="AN0131600304-1">Linguistic performance of dual language immersion students </title> <p>The assessment of dual language immersion (DLI) students’ proficiency in the partner language plays a key role in program evaluation. This article reports findings from a large‐scale study, drawing on cross‐sectional assessment data from 73 schools and longitudinal data from 25 schools. Students were tested in the interpretive (listening and reading) and interpersonal (listening/speaking) modes of communication, using the ACTFL Assessment of Proficiency Towards Performance in Languages. The cross‐sectional analysis shows that French and Spanish students performed at Intermediate Low across skills in fourth or fifth grade and Intermediate Mid in sixth grade. Sixth‐grade Chinese DLI students attained near Intermediate Mid in listening and Novice High to Intermediate Low in reading and speaking. Across languages, students achieved the highest ratings in listening and, overall, performed more strongly in listening and reading than speaking. A regression analysis of students’ change scores over time revealed significant progress in nearly all tested skills and languages. Baseline scores and partner language (Chinese, French, or Spanish) accounted for significant variation in growth rates; between‐school differences were found not to be significant. The article discusses implications for assessment instruments and DLI policy and also addresses some instructional implications.</p> <p>immersion; performance assessment; Chinese (Mandarin); French and Spanish</p> <hd id="AN0131600304-2">INTRODUCTION</hd> <p>On February 28, 2017, the American Academy of Arts & Sciences released a report on the status of language education in the United States. The first paragraph of the report's executive summary concludes that “the vast majority of American citizens remain monolingual” (p. viii). According to the U.S. Census Bureau, only a small fraction of the country's 65 million multilingual residents is able to use their non‐English home language proficiently. Moreover, the American Academy of Arts & Sciences report ([<reflink idref="bib2" id="ref1">2</reflink>] ) pointed out that opportunities to learn languages have decreased, particularly at the elementary and middle school levels; in most states in the United States, less than 20% of K–12 students were enrolled in language courses in 2015 (p. 11).</p> <p>The report highlights dual language immersion (DLI)[<reflink idref="bib9" id="ref2">9</reflink>] as an educational model that is gaining traction around the country. DLI refers to formal schooling in two languages, typically starting in kindergarten or first grade, with at least 50% instruction in the partner language.[<reflink idref="bib10" id="ref3">10</reflink>] Students learn core subjects in a majority language, such as English, and a non‐English classroom partner language (e.g., Chinese, French, German, or Spanish). In the United States, student population demographics determine whether a DLI program is classified as one or two way. In one‐way programs, students whose dominant home language is English acquire an additional language (e.g., Chinese, French, German, or Spanish). Two‐way programs, in contrast, explicitly serve two language groups (e.g., students with English and students with Spanish as their home language in an English/Spanish DLI program). Two‐way programs require a balanced distribution of students’ language backgrounds, usually operationalized as a one‐third to two‐third, and ideally a 50‐50 ratio.</p> <p>Early on, bilingual schooling raised concerns about students’ academic performance in core subjects, such as English language arts and math. However, during the past several decades, studies have consistently demonstrated that learning subject matter content in two languages does not put students at a disadvantage and might even confer benefits (Collier & Thomas, [<reflink idref="bib8" id="ref4">8</reflink>] ; Lazaruk, [<reflink idref="bib17" id="ref5">17</reflink>] ; Marian, Shook, & Schroeder, [<reflink idref="bib20" id="ref6">20</reflink>] ; Steele et al., [<reflink idref="bib27" id="ref7">27</reflink>] ; Watzinger‐Tharp, Swenson, & Mayne, [<reflink idref="bib33" id="ref8">33</reflink>] ). Steele et al. ([<reflink idref="bib27" id="ref9">27</reflink>] ), in a rigorously designed study that controlled for selection bias, found positive effects of dual immersion on students’ reading scores and on English learners’ (ELs’) reclassification rates (i.e., the point at which an EL has attained the required level of English proficiency to be reclassified as a fluent English speaker).</p> <p>With increasingly robust evidence of DLI students’ academic achievement, stakeholders and researchers have begun to pay greater attention to assessment of partner language proficiency. Several states explicitly articulate education of a bilingual workforce that is able to compete successfully in the global marketplace as a key goal of their DLI programs (Burkhauser, Steele, Slater, Bacon, & Miller, [<reflink idref="bib5" id="ref10">5</reflink>] ). Clearly, systematic proficiency assessment will play an increasingly important role as DLI continues to grow nationwide. The present article reports findings from a large‐scale study of DLI students’ performance in interpersonal listening/speaking (ILS), interpretive listening (IL), and interpretive reading (IR) as a measure of their progress toward proficiency in Chinese, French, and Spanish. The study builds on recent research on DLI students’ proficiency attainment and proficiency growth over time and expands the existing body of research by drawing on cross‐sectional data from a wide range of schools and districts and by measuring students’ actual proficiency gains over 2 years.</p> <hd id="AN0131600304-3">LITERATURE REVIEW</hd> <p>The question of what level of proficiency students can realistically achieve through academic exposure to a second language has received considerable attention over the past several decades. The review of selected studies below provides an overview of relevant research organized by learning context, beginning with a brief discussion of college and secondary level studies, before focusing on DLI.</p> <hd id="AN0131600304-4">Proficiency attainment at the postsecondary level</hd> <p>Research into proficiency attainment began with Carroll's (1967) seminal national study of foreign language majors. Carroll's study found that seniors in French, German, Italian, and Spanish generally scored at 2+ on the Interagency Language Roundtable (ILR) scale in speaking (roughly corresponding to an ACTFL rating of Advanced High) and achieved slightly lower scores in Russian (1+, or Intermediate Mid). The scores in reading and writing were higher, ranging from 3 to 3+ (ACTFL Superior, which also corresponds to ILR 4 and 5) (Carroll, [<reflink idref="bib6" id="ref11">6</reflink>] , p. 134).</p> <p>Although Carroll (1967) concluded that the “net showing of foreign language majors is not impressive” (p. 134), the ratings were “much higher than those obtained from subsequent studies that used the ACTFL Oral Proficiency Interviews” (Tschirner, [<reflink idref="bib31" id="ref12">31</reflink>] , p. 202). However, in Carroll's study, students were not directly tested for proficiency, but rather received ILR ratings that had been converted from their scores on the skill‐based Modern Language Association (MLA) test. The study also relied on a smaller sample of institutions and students than originally planned. Given these limitations, Carroll's findings may not accurately represent how foreign language majors actually fared at the time of the study. Research at the college level after Carroll (e.g., Brecht, Davidson, & Ginsberg, [<reflink idref="bib4" id="ref13">4</reflink>] ; Cramer & Terrio, [<reflink idref="bib9" id="ref14">9</reflink>] ; Kaplan, [<reflink idref="bib16" id="ref15">16</reflink>] ; Sieloff Magnan, [<reflink idref="bib25" id="ref16">25</reflink>] ; Swender, [<reflink idref="bib29" id="ref17">29</reflink>] ; Thompson, [<reflink idref="bib30" id="ref18">30</reflink>] ; Tschirner & Heilenman, [<reflink idref="bib32" id="ref19">32</reflink>] ; Weinstein, [<reflink idref="bib34" id="ref20">34</reflink>] ) has shown that most students attain ACTFL Intermediate‐High to Advanced‐Low proficiency in speaking after 4 years of language study. These studies also found that speaking scores tend to cover a wide range, from Novice High to Advanced High.</p> <p>Compared with research on college students’ speaking proficiency, the receptive skills (reading and listening) have received less attention. Since Carroll ([<reflink idref="bib6" id="ref21">6</reflink>] ), only one comprehensive investigation of U.S. college students’ reading and listening proficiency has been published. Tschirner ([<reflink idref="bib31" id="ref22">31</reflink>] ) analyzed proficiency ratings for 3,000 college students, learning seven different languages. In the cognate languages (French, German, Italian, Portuguese, and Spanish), students’ reading proficiency, on average, reached Intermediate High to Advanced Low by their fourth year of study, although many were still rated as Intermediate Mid (p. 210). Mean proficiency levels in listening for French and Spanish proved to be significantly lower than in reading at all levels tested: the mean for listening ratings was about one sublevel below the mean reading scores in students’ second and third years and two sublevels below their reading scores in their fourth year (p. 219).</p> <p>In sum, recent studies suggest that after 4 years of language study, college students only begin to cross the threshold from Intermediate to Advanced proficiency. In reading and listening in particular, students are more likely to remain at the Intermediate level, perhaps not even scoring at Intermediate High, which requires the ability to perform some tasks at the Advanced level. In other words, foreign language majors do not typically graduate with the level of proficiency that is necessary for using the language in a professional setting, a finding that has implications for articulation across levels; students’ proficiency attainment before entering college; and the goal of a growing number of states to produce a bilingual, biliterate, and bicultural workforce.</p> <hd id="AN0131600304-5">Proficiency attainment at the secondary level</hd> <p>Research on proficiency at the secondary level, conducted in various settings, suggests that students begin to reach the Intermediate level after 4 years of study. Several studies on oral proficiency of secondary‐level students, using a wide range of assessment tools (e.g., Glisan & Foltz, [<reflink idref="bib13" id="ref23">13</reflink>] ; Huebner & Jensen, [<reflink idref="bib15" id="ref24">15</reflink>] ; Moeller & Reschke, [<reflink idref="bib22" id="ref25">22</reflink>] ; Norris & Pfeiffer, [<reflink idref="bib23" id="ref26">23</reflink>] , Steinmeyer, [<reflink idref="bib28" id="ref27">28</reflink>] ; Moeller & Theiler, [<reflink idref="bib21" id="ref28">21</reflink>] ) produced consistent findings. Overall, learners typically attain Novice Mid to Novice High after 2 years of middle or high school study, and Novice High to Intermediate Low after 4 years.</p> <p>A recent study of Chinese, French, and Spanish students in a large urban district (Davin, Rampert, & Hammerand, [<reflink idref="bib10" id="ref29">10</reflink>] ) produced similar results. In all three languages, most students reached Novice High in speaking and also writing after 3 years of study and approached Intermediate Low after 4 years (p. 256). In addition, the data showed an average reading score of Intermediate Low in French, Novice High to Intermediate Low in Spanish, and Novice Mid in Chinese. According to the authors, the results fell short of the district‐wide goal to graduate students with an Intermediate‐Low level of proficiency in all four skills (p. 257).</p> <p>The findings for proficiency levels at the secondary level to some extent mirror those of postsecondary learners and might also help explain them. Like their college‐level counterparts, after 4 years of study, secondary students are unlikely to move from one major level into the next (Novice to Intermediate) in all four skills. Moreover, college students are possibly unlikely to achieve Advanced levels, and especially Advanced Mid to High, because they do not arrive in college with at least an Intermediate‐Low foundation.</p> <hd id="AN0131600304-6">Dual language immersion</hd> <p>Given these findings, it is not surprising that the progress of learners in DLI programs is often compared, typically favorably, with traditional middle and high school programs. For example, Xu, Padilla, and Silva ([<reflink idref="bib35" id="ref30">35</reflink>] ) compared STAMP[<reflink idref="bib11" id="ref31">11</reflink>] scores of fourth‐ and fifth‐grade elementary DLI students (including those from the 2013 study) with 119 students in their fourth or fifth year (advanced placement [AP]) of Mandarin language study at two high schools. All of the schools were located in the same suburban district. The study found that 56% of fourth‐grade Mandarin DLI students scored at the Intermediate level in reading, with the majority among these rated at Intermediate Mid. In contrast, only 30% of fourth‐year high school students were rated as Intermediate (p. 30). Most high‐school students attained Intermediate levels in their fifth (AP) year of study, but none reached Intermediate High. In contrast, 23% of fifth‐year DLI students reached Intermediate High (p. 30). In speaking and writing, however, high school students outperformed their comparison DLI cohorts. Most high school students scored at Intermediate Mid, whereas fifth‐grade DLI students attained Intermediate Low (p. 31). Xu and colleagues attributed the speaking and writing results to the nature of the tasks, which, in contrast to the reading tasks, involved open‐ended questions. They hypothesized that the nature of open‐ended tasks may benefit older students who are likely to have more experience in crafting organized responses and to be more advanced in their cognitive development.</p> <p>Other studies, summarized in Table , have focused exclusively on DLI students at elementary or secondary levels. A 2013 Center for Applied Second Language Studies (CASLS) study analyzed STAMP test scores for DLI students in Chinese, French, Japanese, and Spanish in grades 6 through 12. Results were reported by skill (reading, writing, and speaking), but not separated by language; listening was not included in the study. Most of the tested population attained level 4 (characterized as roughly equivalent to ACTFL Intermediate Low) in grade 6 in writing (56%) and in grade 8 in speaking (59%). In reading, 50% of the tested students attained this level in grade 10. In grade 12, across the three tested skills, the study reported that 50% to 63% of the students scored at level 5 (corresponding to Intermediate Mid) or higher. However, the sample had decreased to around 30 twelfth‐grade test takers, making it difficult to draw generalizable conclusions about proficiency development by the end of the cohort's’ secondary schooling. Moreover, because the report did not disaggregate data by language, these remaining students may or may not have represented all four languages that were originally included in the study.</p> <p>Summary of DLI studies</p> <p> <ephtml> <table border="1" cellpadding="7"><tr><th>Study</th><th>N</th><th>Partner language(s)</th><th>Grades</th><th>Assessment</th><th>Skills assessed</th><th>Main findings for proficiency levels attained by majority of students (50%+)</th></tr><tr><td>Avant Assessment ()</td><td>NA</td><td>Cantonese, Mandarin, French, German, Japanese, Spanish, Russian</td><td>2–6</td><td>STAMP</td><td>Reading, writing, listening, and speaking</td><td>Intermediate Low (all four skills) ‐ fourth or fifth grade in French, German, Spanish, and Russian ‐ sixth grade in Japanese</td></tr><tr><td>CASLS ()</td><td>1,477</td><td>Chinese, French, Japanese, Spanish</td><td>6–12</td><td>STAMP</td><td>Speaking, reading, writing</td><td>Intermediate Low (all four languages) ‐ sixth grade in writing ‐ eighth grade in speaking ‐ tenth grade in reading</td></tr><tr><td>Burkhauser et al. ()</td><td>1,284</td><td>Japanese, Mandarin, Spanish</td><td>3–8</td><td>STAMP</td><td>Speaking, listening, reading, writing</td><td>Intermediate Low or higher* ‐ By eighth grade in all skills and languages, except Japanese listening and Chinese reading ‐ Intermediate</td></tr><tr><td>Padilla et al. ()</td><td>13–14</td><td>Mandarin</td><td>5</td><td>STAMP</td><td>Listening, speaking, reading, writing</td><td>Novice High to Intermediate High (fifth grade) ‐ Intermediate High in listening ‐ Intermediate Low in speaking/writing ‐ Novice High to Intermediate High in reading</td></tr><tr><td>Fortune & Tedick ()</td><td>218</td><td>Spanish</td><td>2, 5, and 8</td><td>SOPA/COPE</td><td>Speaking, listening</td><td>Intermediate High to Advanced Low (both skills) ‐ second grade: Intermediate High ‐ fifth grade: Advanced Low ‐ eighth grade: Intermediate High</td></tr></table> </ephtml> </p> <p>4 CASLS, Center for Applied Second Language Studies; DLI, dual language immersion; NA, not available; STAMP, Standards‐based measurement of proficiency. *At least 75% of students.</p> <p>Another, more recent national[<reflink idref="bib12" id="ref32">12</reflink>] study (Avant Assessment, [<reflink idref="bib3" id="ref33">3</reflink>] ) reported averages of test results in reading, writing, listening, and speaking by language.[<reflink idref="bib13" id="ref34">13</reflink>] French, German, and Spanish immersion students were at or near Intermediate Low in fourth grade across skills; Chinese and Russian students attained this level by fifth grade; and Japanese students attained this level by sixth grade. Cantonese students remained in the Novice range (near Novice High), except in listening. Where they reached the Intermediate level in fourth grade, overall, within each language, students performed highest in listening, approaching Intermediate Mid or between Intermediate Mid and High by fifth grade. Writing and speaking means were found to be generally similar to each other, but also lower, sometimes significantly so, than reading or listening. For example, the national average for fifth‐grade Spanish speaking was Intermediate Mid compared with Intermediate High in reading and listening.</p> <p>The descriptive data in these national reports provide a general picture of K–12 proficiency outcomes and complement more in‐depth analyses of DLI proficiency levels at one or two schools and languages (e.g., Padilla, Fan, Xu, & Silva, [<reflink idref="bib24" id="ref35">24</reflink>] ; Xu et al., [<reflink idref="bib35" id="ref36">35</reflink>] ) or across multiple schools, program types, or languages (e.g., Burkhauser et al., [<reflink idref="bib5" id="ref37">5</reflink>] ; Fortune & Tedick, [<reflink idref="bib11" id="ref38">11</reflink>] ). Padilla et al. ([<reflink idref="bib24" id="ref39">24</reflink>] ) analyzed proficiency assessments of K–5 students in a two‐way Mandarin Chinese/English program, located in an affluent community in Northern California. To measure students’ performance in listening/oral, reading, and writing, the cross‐sectional study relied on the Mandarin Proficiency Assessment (MPA), developed by Mandarin Chinese immersion teachers in the program together with Stanford University researchers. MPA results for listening/oral assessments showed that by grade 3, students were able to respond to teachers’ questions at the sentence level. A small subset of students who took STAMP tests in fifth grade in all four skills performed highest in listening, with 60% (8 of 14 students) attaining Intermediate High. In speaking and writing, the same percentage attained Intermediate Low (level 4). Reading results were distributed evenly, and more widely, than the other skills, from Novice High to Intermediate High.</p> <p>For their cross‐sectional study of English‐proficient Spanish DLI students’ oral proficiency in grades K–8, Fortune and Tedick ([<reflink idref="bib11" id="ref40">11</reflink>] ) analyzed results for 218 students in grades 2, 5, and 8 on two performance measures. The study found that students’ median scores for three speaking measures (oral fluency, grammar, and vocabulary) were Intermediate High in grade 2, three sublevels higher than kindergarten students who attained Novice High. The median score for students in grade 5 was one sublevel higher, at Advanced Low, but eighth‐grade students achieved below fifth graders, and more comparable to grade 2, with median scores at Intermediate High. Students achieved the highest levels in listening; roughly 70% to 80% attained the Advanced range in grades 2, 3, and 8 (p. 645). The authors of this study attributed the high oral proficiency levels of students in kindergarten and grade 2 to early total immersion,[<reflink idref="bib14" id="ref41">14</reflink>] because the results compared favorably to those obtained in previous studies of students in partial immersion (p. 649).</p> <p>The most comprehensive study to date was conducted by Burkhauser et al. ([<reflink idref="bib5" id="ref42">5</reflink>] ) in the Portland Public Schools. Burkhauser and her colleagues used random‐effect models to estimate proficiency gains over time for students in Japanese, Mandarin Chinese, and Spanish DLI based on data from three school years (2011–2014). By eighth grade, at least 75% of the students scored at level 4 (equivalent to Intermediate Low) or higher in all four skills, except for Japanese listening (51% at or above level 4) and Chinese reading (30% at level 4). Spanish DLI students attained, on average, near Intermediate High in reading, and Chinese DLI students approximated this level in speaking (p. 425). Japanese DLI students performed at Intermediate Mid in the productive skills (speaking and writing) and Intermediate Low in listening and reading (p. 426).</p> <p>In addition to determining proficiency levels in all four skills, as well as differences between immersion partner languages, the study also measured the potential effect of students’ home language on Spanish proficiency. When students were first tested in fourth grade, any differences that might have existed between heritage and nonheritage learners had essentially disappeared. However, students whose home language (Spanish) matched the immersion partner language appeared to be at a disadvantage in reading in grades 7 and 8, and in listening in grade 8 (Burkhauser et al., [<reflink idref="bib5" id="ref43">5</reflink>] , p. 425).</p> <p>Although it is difficult to compare the results of studies that represent a wide range of sample sizes, assessment tools, languages, program models, and research methods, generally speaking, it is possible to conclude that students who start DLI in kindergarten or first grade are likely to reach Intermediate levels of proficiency in listening and reading in fifth or sixth grade. Rather consistently, studies have found that speaking scores tend to be lower than the receptive skills and, also, that they vary widely, a finding reported for secondary and postsecondary studies as well. Studies that were conducted on a smaller scale found that students achieved at the upper Intermediate range by fifth grade (e.g., Fortune & Tedick, [<reflink idref="bib11" id="ref44">11</reflink>] ; Xu et al., [<reflink idref="bib35" id="ref45">35</reflink>] ). In contrast, Burkhauser and her colleagues ([<reflink idref="bib5" id="ref46">5</reflink>] ), who conducted a district‐wide study, reported achievement of Intermediate High generally no earlier than eighth grade. Studies that tracked DLI students’ proficiency levels into middle or high school (e.g., Burkhauser et al., [<reflink idref="bib5" id="ref47">5</reflink>] ; Fortune & Tedick, [<reflink idref="bib11" id="ref48">11</reflink>] ) found that proficiency gains tended to slow down or even decline in grades 7 and 8. The results for middle school grades are somewhat sobering, but not surprising, considering that instructional time in the immersion partner language typically declines from at least half of a school day in elementary grades, to no more than about 30% in middle school.</p> <p>The present study contributes to the existing literature on DLI proficiency attainment in several ways. First, it examined DLI students’ performance in three different skills across two modes of communication—the interpretive mode (listening and reading) and the interpersonal mode (listening/speaking), based on a large sample of students in three DLI languages (Chinese, French, and Spanish). The study also measured proficiency gains over time for students who were tested twice at two different grade levels. And, finally, disaggregated results from students in two DLI program types, Spanish one way and Spanish two way, are presented.</p> <p>Drawing on cross‐sectional and longitudinal assessment data, the study addressed the following research questions:</p> <p>At what levels do DLI students at different grade levels perform in interpretive listening and reading and in interpersonal listening/speaking, and to what extent do their performance scores differ by language?</p> <p>To what extent do DLI students’ scores change from one tested grade to the next, and are there statistically significant differences among languages and skills?</p> <hd id="AN0131600304-7">METHODS</hd> <hd id="AN0131600304-8">Context</hd> <p>The state of Utah established its public school DLI program in 2009 with funding from the state legislature (Senate Bill 41) after two pilot years with four Spanish one‐ and two‐way DLI programs in two school districts. As of 2018, there were 198 programs in six languages (Chinese, French, German, Spanish, Portuguese, and Russian). All DLI programs used a two‐teacher 50/50 model, which stipulates 50% of daily instructional time in each language. For grades 1 through 6, the DLI curriculum specifies the language of instruction for school subjects. For example, math is taught exclusively in the immersion partner language in grades 1 through 3 and shifts to the English half of the day in grade 4. Students receive a significant amount of highly structured partner language literacy instruction across all grades, ranging from roughly a third of the partner language portion of the day in the lower grades and increasing to half in grades 4 and 5. At the middle school level, typically grades 6 and 7, DLI students continue with a minimum of two courses in the partner language.</p> <p>The state of Utah's DLI model subscribes to separation of languages, and teachers’ 100% partner language use in all of their interactions with students. Administrative and curricular DLI guidelines as well as evidence sources are codified in a set of nine fidelity assurances (a codified set of requirements that schools need to meet to be part of the state DLI program) for elementary grades, and four such assurances for middle school. In addition, teachers are trained to adhere to a specific set of instructional strategies. As mandated by the assurances, all DLI students in grades 3 through 9 are assessed once a year with the ACTFL Assessment of Proficiency Towards Performance in Languages (AAPPL) test. At the time of this study, students were required to be tested in speaking in grades 3, 5, 7, and 9, and in listening and reading in grades 4, 6, and 8. However, some school districts opted to assess additional skills outside the state schedule.</p> <p>For each tested grade and skill, Utah DLI performance benchmarks establish a “proficient performance” and a “highly proficient performance” score (see Appendix). For example, the ILS benchmark for fifth‐grade Chinese is Novice Mid (N3) for proficient and Novice High (N4) for highly proficient. The benchmarks track students’ progress toward the ultimate goal of attaining Advanced proficiency by the end of their secondary schooling. These benchmarks were informed by assessment data from DLI programs in other states, research on DLI students’ proficiency attainment (e.g., CASLS, [<reflink idref="bib7" id="ref49">7</reflink>] ; Fortune & Tedick, [<reflink idref="bib11" id="ref50">11</reflink>] ), and proficiency data collected since AAPPL assessments were implemented in Utah in 2013.</p> <hd id="AN0131600304-9">Participants</hd> <p>Data are reported for two studies, one cross‐sectional and the other longitudinal.</p> <p>Cross‐sectional study: Participants for the cross‐sectional study included students in grades 3 to 6 who were enrolled in Chinese, French, and Spanish DLI in 73 schools, representing 17 districts, during the 2015 school year. Following Utah's AAPPL assessment schedule, all students in fourth and sixth grade were tested in listening and reading, whereas third‐ and fifth‐grade students were administered the speaking test. In addition, data were obtained from students who started Spanish DLI during the 2007 to 2008 and 2008 to 2009 DLI pilot years. (Chinese and French DLIs were implemented in 2009 to 2010, the inaugural year of the legislatively funded state model.)</p> <p>Longitudinal study: The longitudinal study drew on assessment data from two student cohorts who were tested in 2013 and again in 2015. Students who were in third grade in 2013 and in fifth grade in 2015 were tested in ILS; students who were in fourth grade in 2013 and in sixth grade in 2015 were tested in IR and IL. Collectively, these cohorts came from 25 schools that represented a total of 10 districts.</p> <hd id="AN0131600304-10">Assessment instrument</hd> <p>The AAPPL is a computer‐based Internet‐delivered test that assesses performance in the four skills across three modes of communication: ILS, IL, IR, and presentational writing. As a performance assessment, the AAPPL is curriculum driven and age appropriate, with tasks that are relevant to the elementary or secondary classroom. For example, at the Novice level, a typical task may ask students to name their favorite color and things they have that are that color. An ILS task at the Intermediate level may prompt students to talk about what they do when they get home after school. The AAPPL measures students’ performance, which ACTFL distinguishes from proficiency in terms of the context in which a specific language behavior is demonstrated. Performance is “the ability to use language that has been learned and practiced in an instructional setting” and is used “within familiar contexts and content areas” (ACTFL, [<reflink idref="bib1" id="ref51">1</reflink>] , p. 4).</p> <p>Proficiency, on the other hand, refers to an individual's ability to use language in real‐world situations and spontaneous unrehearsed contexts. However, ACTFL acknowledges that sustained performance over a variety of tasks at a certain level can point to, and serve as an indicator of, proficiency. In that sense, the AAPPL measures performance toward proficiency because it exposes students to a variety of tasks within a limited range of levels over several familiar contexts. Students who perform at a certain level in such tasks are likely to show a similar level of proficiency in real‐life situations.</p> <p>AAPPL scores are based on the ACTFL Performance Descriptors for Language Learners (ACTFL, [<reflink idref="bib1" id="ref52">1</reflink>] ) and are reported using the major ACTFL proficiency levels, although the Novice and Intermediate sublevels are more fine grained than the ACTFL scale so as to capture incremental progress of learners at these instructional levels. On the AAPPL scale, Novice includes four sublevels: Novice Low (N1), Novice Mid (N2 and N3), and Novice High (N4). Intermediate is divided into five sublevels, with Intermediate Low (I1), Intermediate Mid (I2‐I4), and Intermediate High (I5). One score (A) is given to students who perform at Advanced Low or above. The AAPPL test is available in two different forms: form A, recommended for students through grade 6, measures performance in the Novice to Intermediate Mid ranges; and form B, for students in grades 7 and above, measures Novice High to Advanced. Form A produces scores from Novice Low to the upper end of the Intermediate‐Mid (I4) range, whereas Form B returns scores from Novice High (N4) through Advanced (A). If a performance on form B does not reach the N4 level, it is recorded as below rating. As is the case for the ACTFL proficiency scale, the AAPPL sublevels do not represent equal intervals. Rather, the quantity and quality of language that is required to move through Novice and Intermediate levels, and into Advanced, are assumed to increase progressively.</p> <hd id="AN0131600304-11">Analyses</hd> <p>The cross‐sectional analysis relied on descriptive statistics (SAS 9.4) to report results of the AAPPL IL, IR, and ILS assessments by language and grade level. The growth study, drawing on 2013 and 2015 assessment data, used ordinary least squares regression to measure differences in students’ growth across languages, based on individual student growth data between two testing grades. This “change score” treatment, examining the average growth in scores across 2 years, investigates whether the change from baseline is meaningfully different from zero while avoiding any need to control for demographics. It avoids the difficulty of controlling for variables such as socioeconomic status or sex, because the students are compared with themselves—an individual student's sex, for example, does not change. A further complexity was examined, constructing a mixed model with students nested within schools to control for differences between schools. However, this approach yielded nonsignificant null model likelihood ratios indicating that an insignificant portion of the variation was attributable to between‐school differences.</p> <p>One‐ and two‐way Spanish results were disaggregated, because these two program types differ demographically by design: in two‐way programs, the non‐English partner language matches the home language for at least one third, but no more than two thirds, of the student population. However, ELs or students with a home language other than English were not disaggregated, because this information had not been collected rigorously as part of the AAPPL assessments.</p> <p>For the purpose of calculating test results, scores were coded on a 10‐point scale that matched the progression from Novice to Advanced. Scores from 1 through 4 represent the Novice range (with 2 and 3 for Novice Mid), and scores from 5 through 9 represent Intermediate, with three score points, 6, 7, and 8, representing Intermediate Mid, and 9 representing Intermediate High; 10 represents one score point for Advanced Low or higher. Scores “below N1” on the form A scale were coded as 0, and scores “below N4” in form B were averaged as 2.</p> <hd id="AN0131600304-12">RESULTS</hd> <hd id="AN0131600304-13">Students’ performance scores at different grade levels</hd> <p>Table shows the results of the cross‐sectional study, which was designed to determine DLI students’ performance in IL, IR, and ILS at different grade levels.</p> <p>Test scores by language, test, and grade level</p> <p> <ephtml> <table border="1" cellpadding="13"><tr><th /><th>Chinese</th><th>French</th><th>Spanish one way</th><th>Spanish two way</th></tr><tr><th>Test/grade</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th></tr><tr><td>IL/4</td><td>721</td><td>5.21</td><td>1.39</td><td>445</td><td>5.55</td><td>1.36</td><td>560</td><td>6.41</td><td>1.22</td><td>544</td><td>6.2</td><td>1.26</td></tr><tr><td>IL/6</td><td>260</td><td>5.79</td><td>1.46</td><td>215</td><td>6.31</td><td>1.26</td><td>355</td><td>7.12</td><td>1.09</td><td>225</td><td>6.77</td><td>1.2</td></tr><tr><td>ILS/3</td><td>1162</td><td>3.39</td><td>1.31</td><td>489</td><td>4.29</td><td>1.3</td><td>1121</td><td>4.67</td><td>1.45</td><td>783</td><td>4.87</td><td>1.64</td></tr><tr><td>ILS/5</td><td>677</td><td>4.33</td><td>1.36</td><td>247</td><td>5.45</td><td>1.6</td><td>484</td><td>5.67</td><td>1.58</td><td>450</td><td>5.53</td><td>1.69</td></tr><tr><td>ILS/6</td><td>255</td><td>4.41</td><td>1.3</td><td>210</td><td>5.82</td><td>1.64</td><td>311</td><td>6.33</td><td>1.65</td><td>221</td><td>5.92</td><td>1.79</td></tr><tr><td>IR/4</td><td>698</td><td>3.55</td><td>1.35</td><td>444</td><td>5.68</td><td>1.68</td><td>560</td><td>5.53</td><td>1.55</td><td>538</td><td>5.23</td><td>1.64</td></tr><tr><td>IR/6</td><td>249</td><td>4.37</td><td>1.61</td><td>215</td><td>6.74</td><td>1.34</td><td>352</td><td>6.52</td><td>1.47</td><td>224</td><td>6.33</td><td>1.45</td></tr></table> </ephtml> </p> <p>5 IL, interpretive listening; ILS, interpersonal speaking; IR, interpretive reading.</p> <p>Fourth graders were tested in listening and reading. In listening, Spanish students in fourth grade, on average, performed at Intermediate Mid; and Chinese and French students performed at Intermediate Low. Fourth‐grade Spanish reading scores averaged roughly one sublevel lower than listening, although Chinese reading scores were between Novice Mid and High, about two sublevels lower than Chinese listening. In contrast to Spanish and Chinese, French reading scores were slightly higher than listening.</p> <p>In ILS, students were tested in grades 3, 5, and 6. Third‐grade speaking scores were in the Novice Mid to Novice High range across all languages, although Spanish students approached Intermediate Low. Fifth‐grade students in French and Spanish attained Intermediate Low to Mid, whereas Chinese students scored between Novice High and Intermediate Low.</p> <p>Because sixth‐grade students took all three tests, it is possible to compare performance scores across skills and languages. Chinese and Spanish students performed strongest in listening compared with reading and speaking, whereas French students attained the highest scores in reading. Spanish and French sixth graders scored, on average, at Intermediate Mid (between I2 and I3) in listening and reading, and Intermediate Mid (I2) in speaking. The mean for Chinese sixth‐grade students was close to Intermediate Mid (I2) in listening, and between Novice High and Intermediate Low in speaking and reading.</p> <p>The differences between grade levels were greatest in reading, with the exception of Chinese. For example, sixth‐grade students in Spanish two‐way programs scored, on average, about one sublevel higher than their counterparts in fourth grade. French and Spanish one‐way reading results show a similar pattern, although they were slightly higher than Spanish two‐way results in grades 4 and 6. The mean for Chinese sixth‐grade reading was about one sublevel higher than fourth grade, and for both grades roughly two sublevels below French and Spanish.</p> <p>Table shows mean test scores for Spanish students in grades 7 and 8 and the percentage attainment at the top of Intermediate Mid (I4) and Intermediate High (I5). Because DLI shifts to a course‐based model in middle school, often referred to as DLI continuation, there is no longer a distinction between one‐ and two‐way programs.</p> <p>AAPPL results for Spanish: grades 7 and 8</p> <p> <ephtml> <table border="1" cellpadding="6"><tr><th>Test/grade</th><th>N</th><th>Mean</th><th>SD</th><th>% (N) at I4</th><th>% (N) at I5</th></tr><tr><td>IL/7</td><td>145</td><td>8.63</td><td>1.06</td><td>14.86 (22)</td><td>65.55 (97)</td></tr><tr><td>IL/8</td><td>109</td><td>8.76</td><td>0.88</td><td>14.68 (16)</td><td>65.14 (71)</td></tr><tr><td>ILS/7</td><td>164</td><td>7.46</td><td>1.77</td><td>43.29 (71)</td><td>15.85 (26)</td></tr><tr><td>ILS/8</td><td>108</td><td>7.34</td><td>1.65</td><td>44.44 (48)</td><td>11.11 (12)</td></tr><tr><td>IR/7</td><td>145</td><td>8.59</td><td>1.36</td><td>17.69 (26)</td><td>47.62 (70)</td></tr><tr><td>IR/8</td><td>109</td><td>8.50</td><td>1.42</td><td>13.76 (15)</td><td>52.29 (57)</td></tr></table> </ephtml> </p> <p>6 AAPPL, ACTFL Assessment of Proficiency Towards Performance in Languages; I4, top of Intermediate Mid; I5, Intermediate High; IL, interpretive listening; ILS, interpersonal speaking; IR, interpretive reading.</p> <p>Students tested in eighth grade received only minimally higher or sometimes lower scores than students in seventh grade, but students in both grades performed at higher levels than sixth graders. Spanish students performed, on average, at Intermediate Mid to Intermediate High (I4–I5) in listening and reading, and Intermediate Mid (I3–I4) in speaking, about one to two sublevels higher than students who were tested in sixth grade. About two thirds of the test takers in grades 7 or 8 attained Intermediate High. In reading, about half of the tested students also attained Intermediate High at these grade levels, and roughly 20% scored Advanced. In speaking, the most common score was Intermediate Mid (I4).</p> <p>Figures through 4 show the score ranges and median scores for Chinese, French, and Spanish AAPPL test scores by language, grade, and skill.</p> <p>There were marked differences in the distribution of scores among languages. A wide range in speaking scores was found in French and Spanish in all tested grades, whereas Chinese speaking scores were fairly consistent. Similarly, fourth‐grade reading scores were narrowly spread in Chinese, but they were spread widely in French and Spanish. Conversely, fourth‐grade Chinese listening scores were spread widely, but were tight in comparison with French and Spanish. One‐ and two‐way Spanish score distribution patterns were overall fairly similar to each other across grades and skills.</p> <p>The box plots illustrate that a considerable number of sixth‐grade French and Spanish students attained the highest possible rating of I4, equivalent to the top end of Intermediate Mid on the form A test scale. About 25% to 30% of Spanish students received I4 in listening, speaking, and reading; 40% of French students did so in reading, 25% in speaking, and about 20% in listening.</p> <hd id="AN0131600304-14">Proficiency growth between grade levels</hd> <p>In addition to cross‐sectional data, proficiency growth data were collected for a cohort of Chinese, French, Spanish one‐way, and Spanish two‐way students who took the AAPPL listening (IL) and reading (IR) tests in fourth grade and again in sixth grade,[<reflink idref="bib15" id="ref53">15</reflink>] and a cohort who took the speaking test (ILS) in third and fifth grade. Table shows mean proficiency scores for each cohort at the two tested grade levels; Figure illustrates mean growth, or change scores, for each skill across four language groups.</p> <p>Descriptive statistics for score changes between two grade levels</p> <p> <ephtml> <table border="1" cellpadding="13"><tr><th /><th>Chinese</th><th>French</th><th>Spanish one way</th><th>Spanish two way</th></tr><tr><th>Test/grade</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th><th>N</th><th>Mean</th><th>SD</th></tr><tr><td>IL/4</td><td>229</td><td>4.71</td><td>1.42</td><td>182</td><td>5.72</td><td>1.21</td><td>240</td><td>6.36</td><td>1.41</td><td>116</td><td>6.29</td><td>1.22</td></tr><tr><td>IL/6</td><td>229</td><td>5.77</td><td>1.44</td><td>182</td><td>6.28</td><td>1.28</td><td>240</td><td>7.17</td><td>1.09</td><td>116</td><td>6.81</td><td>1.32</td></tr><tr><td>IR/4</td><td>215</td><td>3.09</td><td>1.35</td><td>183</td><td>4.86</td><td>1.47</td><td>272</td><td>5.22</td><td>1.65</td><td>132</td><td>4.87</td><td>1.59</td></tr><tr><td>IR/6</td><td>215</td><td>4.34</td><td>1.58</td><td>183</td><td>6.83</td><td>1.29</td><td>272</td><td>6.60</td><td>1.48</td><td>132</td><td>6.38</td><td>1.38</td></tr><tr><td>ILS/3</td><td>560</td><td>2.56</td><td>1.05</td><td>193</td><td>5.25</td><td>1.69</td><td>396</td><td>4.13</td><td>1.75</td><td>242</td><td>5.09</td><td>2.01</td></tr><tr><td>ILS/5</td><td>560</td><td>4.51</td><td>1.2</td><td>193</td><td>5.56</td><td>1.66</td><td>396</td><td>5.53</td><td>1.56</td><td>242</td><td>5.36</td><td>1.69</td></tr></table> </ephtml> </p> <p>7 IL, interpretive listening; ILS, interpersonal speaking; IR, interpretive reading.</p> <p>In reading, all four language groups gained at least one sublevel. French and Spanish students scored at or near I1 (Intermediate Low) in grade 4 and moved to I2 or I3 (Intermediate Mid), with French students growing the most (1.97). Chinese students moved from Novice Mid (N3) to Novice High (N4). In listening, students in fourth grade scored higher than in reading, but then made smaller gains from grade 4 to grade 6. French and Spanish students attained at or near Intermediate Mid (I2) in fourth grade and scored at or near the next higher Intermediate‐Mid level in sixth grade. Chinese students progressed from near Intermediate Low (4.71) to near Intermediate Mid (5.77).</p> <p>Overall, students appeared to make greater gains at lower proficiency levels, as demonstrated, for example, by the speaking growth of Chinese and Spanish one‐way students compared with French and Spanish two‐way students. Similarly, in listening, Chinese students grew to a greater degree than French and Spanish students, who had attained at or near Intermediate Mid in fourth grade. Speaking growth varied widely across languages, ranging from .28 (Spanish two way) to 1.95 (Chinese). The .28 growth rate for Spanish two way represents modest progress within the Intermediate Low (I1) level, whereas the 1.95 growth rate for Chinese students represents improvement by two sublevels, from Novice Mid (N2–N3) to Novice High/Intermediate Low (N4–I1). Spanish one‐way students experienced a growth rate of 1.39, representing improvement by one sublevel (Novice High to Intermediate Low). In contrast, French and Spanish two‐way students achieved Intermediate Low (I1) in third grade, but they made smaller gains from third to fifth grade than their one‐way Spanish peers.</p> <p>The descriptive statistics provided in Table and Figure measure students’ mean proficiency gains, but they do not reveal whether they are statistically significant. That is, they do not show if the mean change scores that appear relatively small in comparison to others actually represent meaningful proficiency growth. Table shows the results of the ordinary least squares regression examining in which languages and skills students experienced statistically significant growth over time, based on their change scores. Chinese was used as the reference (or holdout) group because the relatively substantial growth across skills from their baselines is assumed to be meaningful. Testing French, Spanish one way, and Spanish two way as reference groups confirmed this assumption.</p> <p>OLS regression on change scores by language and skill</p> <p> <ephtml> <table border="1" cellpadding="6"><tr><th>Variable</th><th>Parameter estimate</th><th>SE</th><th>t value</th><th>p value</th><th>Adjusted R2</th></tr><tr><td>Listening/IL</td><td /><td /><td /><td /><td>0.36</td></tr><tr><td>Intercept</td><td>4.22</td><td>0.18</td><td>23.50</td><td><.0001</td><td /></tr><tr><td>French</td><td>0.13</td><td>0.13</td><td>1.03</td><td>.30</td><td /></tr><tr><td>Spanish one way</td><td>0.80</td><td>0.13</td><td>6.30</td><td><.0001</td><td /></tr><tr><td>Spanish two way</td><td>0.39</td><td>0.15</td><td>2.61</td><td>.01</td><td /></tr><tr><td>Baseline 2013</td><td>−0.66</td><td>0.03</td><td>−19.87</td><td><.0001</td><td /></tr><tr><td>Speaking/ILS</td><td /><td /><td /><td /><td>0.50</td></tr><tr><td>Intercept</td><td>3.97</td><td>0.09</td><td>44.98</td><td><.0001</td><td /></tr><tr><td>French</td><td>0.43</td><td>0.14</td><td>3.14</td><td>.001</td><td /></tr><tr><td>Spanish one way</td><td>0.63</td><td>0.10</td><td>6.05</td><td><.0001</td><td /></tr><tr><td>Spanish two way</td><td>0.31</td><td>0.12</td><td>2.49</td><td>.01</td><td /></tr><tr><td>Baseline 2013</td><td>−0.78</td><td>0.02</td><td>−31.44</td><td><.0001</td><td /></tr><tr><td>Reading/IR</td><td /><td /><td /><td /><td>0.35</td></tr><tr><td>Intercept</td><td>3.13</td><td>0.14</td><td>23.17</td><td><.0001</td><td /></tr><tr><td>French</td><td>1.77</td><td>0.14</td><td>12.27</td><td><.0001</td><td /></tr><tr><td>Spanish one way</td><td>1.42</td><td>0.14</td><td>10.24</td><td><.0001</td><td /></tr><tr><td>Spanish two way</td><td>1.31</td><td>0.16</td><td>8.34</td><td><.0001</td><td /></tr><tr><td>Baseline 2013</td><td>−0.60</td><td>0.03</td><td>−19.73</td><td><.0001</td><td /></tr></table> </ephtml> </p> <p>8 IL, interpretive listening; ILS, interpersonal speaking; IR, interpretive reading; OLS, ordinary least squares.</p> <p>In all languages and all tested skills, except French listening (p = .30), students progressed at statistically significant levels across 2 years relative even to the large gains in Chinese. Spanish one‐way students grew by twice the rate compared with Spanish two‐way students in listening and speaking, as shown by parameter estimates of .80 and .39, respectively (listening), and .63 and .31, respectively (speaking). Students’ relative proficiency growth is to a substantial degree determined by the language group (Chinese, French, Spanish one way, or Spanish two way) and by the skill that is measured. The adjusted‐R<sups>2</sups> statistics suggest that about 35% (listening and reading) to 50% (speaking) of the variation in growth is attributable to language group and the base score in 2013. As stated before, a mixed‐model approach, nesting students within schools, showed that with this relatively small sample of schools, no statistically meaningful between‐school differences were revealed.</p> <p>To summarize, the cross‐sectional data show that French and Spanish DLI students attained Intermediate Low proficiency (I1 on the AAPLL scale) in listening and reading in fourth grade, and in speaking in fifth grade. Students in sixth grade scored near I2 (Intermediate Mid) in speaking and I2 to I3 (Intermediate Mid) in listening and reading. Spanish middle school students in grades 7 and 8, on average, reached the upper end of Intermediate Mid (I4), with a significant proportion attaining Intermediate High (I5) in reading and listening. Sixth‐grade Chinese students attained near Intermediate Mid in listening and between Novice High and Intermediate Low in reading and speaking. Overall, students achieved higher scores in the interpretive modes (listening and reading) than in speaking, with the exception of Chinese reading scores, which were below both their speaking and listening scores. The growth analysis revealed that students progressed significantly over time, from one tested grade level to the next 2 years later, in all tested skills and across languages, except French listening. Reading growth rates were the most similar and uniformly the highest.</p> <hd id="AN0131600304-15">DISCUSSION</hd> <p>DLI was implemented as a state‐wide educational model in the state of Utah in 2009. Like their nonimmersion peers in the state of Utah school system, DLI students are assessed for their academic achievement in core subjects. Unlike academic achievement, partner language proficiency assessments are rarely administered as state‐wide accountability measures. Nevertheless, stakeholders in DLI expect evidence that students are attaining proficiency levels that will ultimately move them toward the goal of becoming bilingual and biliterate. National reports and several research studies have shown that DLI students achieve higher levels of proficiency than students in traditional world language programs. However, in comparison to the abundance of academic achievement studies, research into proficiency attainment, especially on a large scale, is still relatively rare. Only a few studies have systematically examined DLI students’ proficiency levels in different languages and skills and across grade levels, or their change scores as a measure of proficiency growth.</p> <p>The cross‐sectional data of the present study show that sixth‐grade French and Spanish students were able to perform interpretive (listening and reading) and interpersonal (speaking) tasks, on average, at the Intermediate Mid level of proficiency. Spanish DLI students, in one‐ and two‐way programs, attained higher levels in listening than in the other skills and also outperformed their French counterparts. Like the Spanish DLI students, Chinese students performed the highest in listening, in contrast to French students, who scored highest in reading. Chinese DLI students attained between Intermediate Low and Mid in listening, but they performed below Intermediate Low, at the major level threshold between Novice and Intermediate, in speaking and reading. The comparatively lower Chinese results can be attributed to students learning a nonalphabetic script, as other studies that compare proficiency assessment results across languages have concluded (e.g., Burkhauser et al., [<reflink idref="bib5" id="ref54">5</reflink>] ; CASLS, [<reflink idref="bib7" id="ref55">7</reflink>] ).</p> <p>A considerable number of French and Spanish students, with percentages ranging from 25% to 40%, scored at the top of the test scale (I4, Intermediate Mid) across all tested skills. These findings suggest that the AAPPL form A scale might not fully capture the proficiency levels of the tested population, because students might have scored above Intermediate Mid if they had been given the opportunity to do so. Indeed, the data for seventh and eighth grade, although limited to a relatively small Spanish pilot cohort, point in this direction. Given that half to two thirds of the seventh graders scored at Intermediate High and about 20% at Advanced in the interpretive modes (listening and reading), it is fairly safe to assume that at least some were able to perform at Intermediate High in sixth grade.</p> <p>The longitudinal growth analysis shows that students’ performance improved at statistically meaningful rates across languages and skills, with the exception of French listening. Tschirner ([<reflink idref="bib31" id="ref56">31</reflink>] ) attributed the challenge of developing listening skills in French to the language's deep orthography (i.e., the lack of correspondence between the writing system and pronunciation of words or syllables [p. 219]). His study of college students’ proficiency levels found that listening development in French stagnated after the second year of college instruction (p. 219). In the current study, French students experienced only small gains in listening, but improved by nearly two sublevels from Novice High to Intermediate Mid in reading. Chinese students gained a full sublevel in listening and improved by even greater rates in speaking and reading. The relatively large growth experienced by Chinese students is attributable to their baseline scores in the Novice range, lower than those of the other languages.</p> <p>Overall, the data for students’ proficiency growth over time presented in this study illustrate that gains in the Intermediate range (which might seem small compared with gains in the Novice range over the same time period) are nevertheless significant. As visualized by the inverted ACTFL pyramid scale, learners typically move relatively quickly through the Novice levels, but need more time and exposure to the language, including immersive experiences, to meet the linguistic demands of the Intermediate and Advanced levels (Sieloff Magnan & Back, [<reflink idref="bib26" id="ref57">26</reflink>] ; Tschirner, [<reflink idref="bib31" id="ref58">31</reflink>] ). The findings of this study also highlight the benefits of the AAPPL assessment measures, which are designed to capture incremental, but significant, changes in proficiency scores in the Intermediate proficiency range.</p> <p>The findings of this study should be viewed in the context of certain limitations, including inherent challenges related to real‐world data. The cross‐sectional study drew on large‐scale data from more than 70 schools, representing 17 school districts. During the years pertinent to the current study, the assessments were not administered centrally at the state level, but rather by districts and schools. Given this “local control,” the data required substantial hygiene efforts to identify and match students and test results correctly. Also, the growth study was limited to a small number of schools per language that had established DLI during the inaugural year of the state‐wide DLI program. Subsequent studies with greater statistical power (with a larger number of schools for each language) and a mixed‐model analysis might well reveal such differences. Future research will also determine if the growth findings hold when schools that adopted DLI in subsequent years are included.</p> <p>The data for seventh and eighth grade were limited to a small cohort of Spanish students in schools that self‐selected participation in the DLI pilot. Clearly, additional data from recent testing years and additional languages will need to be analyzed to determine the levels of proficiency students attain in middle school. More robust middle school data will also make it possible to compare proficiency trajectories in a truly dual immersion experience at the elementary level to course‐based, and less immersive, DLI continuation at the secondary level.</p> <p>The results from this study could have some important instructional implications. The findings for reading assessment show that, with a strong and consistent focus on partner language literacy, all students can make significant progress in the development of reading skills. In contrast, the development of interpersonal speaking might require more explicit attention than it currently receives, perhaps by intentionally capitalizing on students’ demonstrated strength in listening and emulating the highly structured instructional model for reading. Without a more concerted effort to develop speaking, students are less likely to move through and beyond Intermediate proficiency. In sum, the overarching goal of Advanced proficiency attainment in all skills requires a balanced and deliberate instructional approach.</p> <p>Finally, stakeholders should be aware of at least three important factors that affect language development in dual immersion settings. First, DLI students in U.S. schools may have few opportunities to use the non‐English immersion language outside the classroom. Moreover, at school, where the majority language is often also their first or their dominant language, DLI students are likely to opt for English when they interact informally or socially with their DLI peers. This is to be expected even more when the DLI program operates alongside a monolingual English track, as is the case in the state of Utah's DLI program. Along the same lines, studies have found that, without opportunities for using the language in various contexts and to receive feedback on form, DLI students’ attainment of high levels of proficiency may be hampered by limited grammatical accuracy (Genesee, [<reflink idref="bib12" id="ref59">12</reflink>] ; Harley, Cummins, Swain, & Allen, [<reflink idref="bib14" id="ref60">14</reflink>] ; Lindholm‐Leary, [<reflink idref="bib18" id="ref61">18</reflink>] ; Lyster, [<reflink idref="bib19" id="ref62">19</reflink>] ). Second, learning academic content in two languages places significant cognitive and linguistic demands on students. As these demands increase in higher grade levels, students’ proficiency development is likely to decelerate in comparison to their progress at lower levels. Similarly, public discourse reinforces the assumption that children in immersive settings, such as DLI, will acquire two languages effortlessly and rapidly, but DLI students, like all other language learners, need significant time and exposure to reach high levels of proficiency. Finally, middle and high school students will experience less classroom exposure to the language than students in elementary DLI. To moderate expectations of continued growth at the secondary level, it is advisable to make a clear distinction between “dual immersion” at the elementary level and a course‐based continuation of DLI in middle and high school. The value of DLI proficiency assessment extends beyond measuring students’ progress and guiding curriculum and instruction. It helps to educate those invested in DLI about the complexities of language development and the importance of establishing realistic proficiency goals.</p> <hd id="AN0131600304-16">CONCLUSION</hd> <p>Measuring DLI students’ ability to perform in the partner language plays a key role in evaluating DLI programs. First and foremost, frequent and systematic assessment demonstrates whether students meet grade‐level benchmarks and proficiency targets and will ultimately be able to attain Advanced proficiency by the time they graduate from high school. Partner language assessments also provide valuable information for curriculum design, classroom instruction, and professional development of teachers. However, there are some broader implications to be drawn from this study. Systematic and robust data collection and analysis are critical to mitigating unrealistic expectations, which may be inadvertently created when DLI program outcomes are compared with those of traditional world language classrooms. As DLI programs expand and mature across the country, evaluation based on well‐founded proficiency goals will make a stronger case for DLI as a mainstream educational model than “apples and oranges” comparisons that are likely to favor DLI.</p> <p>Practitioners, policy makers, and families are recognizing the potential of DLI to transform education and society by making schooling in two languages and the achievement of bilingualism, biliteracy, and biculturalism the norm rather than an exception. Robust proficiency assessment infrastructures that include iterative data collection, data analysis, and benchmarking, based on collaboration between practitioners and researchers, will be critical for DLI to realize its ambitious potential.</p> <ref id="AN0131600304-17"> <title>Notes</title> <blist> <bibl id="bib1" idref="ref51" type="bt">1</bibl> <bibtext>Johanna Watzinger‐Tharp (PhD, University of Texas at Austin) is Associate Professor of Linguistics, University of Utah, Salt Lake City, UT </bibtext> </blist> <blist> <bibl id="bib2" idref="ref1" type="bt">2</bibl> <bibtext>Fernando Rubio (PhD, State University of New York at Buffalo) is Professor of Spanish Linguistics, University of Utah, Salt Lake City, UT </bibtext> </blist> <blist> <bibl id="bib3" idref="ref33" type="bt">3</bibl> <bibtext>Douglas S. Tharp (MA, University of Texas at Austin) is Research Associate, Pedigree and Population Resource (Utah Population Database), Huntsman Cancer Institute, University of Utah </bibtext> </blist> <blist> <bibl id="bib9" idref="ref2" type="bt">9</bibl> <bibtext>The acronym DLI (dual language immersion) will be used to refer to schooling in two languages in one‐ or two‐way programs. </bibtext> </blist> <blist> <bibl id="bib10" idref="ref3" type="bt">10</bibl> <bibtext>For simplicity's sake, this article will use the term partner language to refer to the non‐English DLI language. </bibtext> </blist> <blist> <bibl id="bib11" idref="ref31" type="bt">11</bibl> <bibtext>STAMP is the Standards‐based Measurement of Proficiency. Throughout this article, we cite studies that use the STAMP measure. The STAMP literature states that STAMP levels are “related to” and “defined by” the proficiency levels and sublevels that are described in the ACTFL Proficiency Guidelines (e.g., Novice High, Intermediate Mid). Some STAMP results are reported using those designations; however, they are not equivalent to official ACTFL ratings. </bibtext> </blist> <blist> <bibl id="bib12" idref="ref32" type="bt">12</bibl> <bibtext>The report included test scores from students in U.S. schools, and international schools outside the United States, when information for years of study was also available. </bibtext> </blist> <blist> <bibl id="bib13" idref="ref23" type="bt">13</bibl> <bibtext>Immersion languages include Cantonese, Mandarin Chinese (simplified and traditional), French, German, Japanese, Spanish, and Russian. </bibtext> </blist> <blist> <bibl id="bib14" idref="ref41" type="bt">14</bibl> <bibtext>Total immersion refers to programs in which no English is used in the early grades. English is progressively introduced until there is an even distribution of the two languages in grade 5 or 6. </bibtext> </blist> <blist> <bibl id="bib15" idref="ref24" type="bt">15</bibl> <bibtext>For this study, the listening and reading cohorts were considered one and the same, although the number is slightly lower in reading than in listening. </bibtext> </blist> </ref> <ref id="AN0131600304-18"> <title>REFERENCES</title> <blist> <bibtext>ACTFL. (2012). ACTFL performance descriptors for language learners 2012 edition. Alexandria, VA: American Council on the Teaching of Foreign Languages. </bibtext> </blist> <blist> <bibtext>American Academy of Arts & Sciences. (2017). America's Languages: Investing in language education for the 21st century. Commission on Language Learning. Cambridge, MA: Author. Retrieved August 1, 2018, from https://<ulink href="http://www.amacad.org/multimedia/pdfs/publications/researchpapersmonographs/language/Commission-on-Language-Learning%5fAmericas-Languages.pdf">www.amacad.org/multimedia/pdfs/publications/researchpapersmonographs/language/Commission-on-Language-Learning%5fAmericas-Languages.pdf</ulink></bibtext> </blist> <blist> <bibtext>Avant Assessment. (2017). Avant STAMP results: National averages 2016–2017. Available at https://d3itqxtdxl1nz0.cloudfront.net/pdfs/STAMP‐National‐Averages‐Update‐2017.pdf </bibtext> </blist> <blist> <bibl id="bib4" idref="ref13" type="bt">4</bibl> <bibtext>Brecht, R. D., Davidson, D. E., & Ginsberg, R. B. (1993). Predictors of foreign language gain during study abroad. NFLC occasional papers. Washington, DC: National Foreign Language Center. </bibtext> </blist> <blist> <bibl id="bib5" idref="ref10" type="bt">5</bibl> <bibtext>Burkhauser, S., Steele, J. L., Slater, R., Bacon, M., & Miller, T. (2016). Partner‐language learning trajectories in dual‐language immersion: Evidence from an urban district. Foreign Language Annals 49, 415–433. </bibtext> </blist> <blist> <bibl id="bib6" idref="ref11" type="bt">6</bibl> <bibtext>Carroll, J. (1967). Foreign language proficiency levels attained by language majors near graduation from college. Foreign Language Annals 1, 131–151. </bibtext> </blist> <blist> <bibl id="bib7" idref="ref49" type="bt">7</bibl> <bibtext>CASLS. (2013). What levels of proficiency do immersion students achieve? Eugene: Center for Applied Second Language Studies (CASLS), University of Oregon. </bibtext> </blist> <blist> <bibl id="bib8" idref="ref4" type="bt">8</bibl> <bibtext>Collier, V.P., & Thomas, W.P. (2004). The astounding effectiveness of dual language education for all. Fairfax, VA: George Mason University. </bibtext> </blist> <blist> <bibtext>Cramer, H., & Terrio, S. (1985). Moving from vocabulary acquisition to functional proficiency: Techniques and strategies. French Review, 59, 198–209 </bibtext> </blist> <blist> <bibtext>Davin, D., Rampert, T., & Hammerand, A. (2014). Converting data to knowledge: One district's experience using large‐scale proficiency assessment. Foreign Language Annals, 47, 241–260. </bibtext> </blist> <blist> <bibtext>Fortune, T., & Tedick, D. (2015). Oral proficiency assessment of English‐proficient K‐8 Spanish immersion students. The Modern Language Journal, 99, 637–655. </bibtext> </blist> <blist> <bibtext>Genesee, F. (1994). Integrating language and content: Lessons from immersion. Santa Cruz, CA: National Center for Research on Cultural Diversity and Second Language Learning. </bibtext> </blist> <blist> <bibtext>Glisan, E. W., & Foltz, D. A. (1998). Assessing students' oral proficiency in an outcome‐based curriculum: Student performance and teacher intuition. Modern Language Journal, 82, 1–18. </bibtext> </blist> <blist> <bibtext>Harley, B., Cummins, J., Swain, M., & Allen, P. (1990). The nature of language proficiency. In B. Harley, P. Allen, J. Cummins, & M. Swain (Eds.), The development of second language proficiency (pp. 7–25). Cambridge: Cambridge University Press. </bibtext> </blist> <blist> <bibtext>Huebner, T., & Jensen, A. (1992). A Study of foreign language proficiency‐based testing in secondary schools. Foreign Language Annals, 25, 105–115. </bibtext> </blist> <blist> <bibl id="bib16" idref="ref15" type="bt">16</bibl> <bibtext>Kaplan, I. (1984). Oral proficiency testing and the language curriculum: Two experiments in curricular design for conversation courses. Foreign Language Annals, 15, 491–498. </bibtext> </blist> <blist> <bibl id="bib17" idref="ref5" type="bt">17</bibl> <bibtext>Lazaruk, W. (2007). Linguistic, academic, and cognitive benefits of French immersion. The Canadian Modern Language Review, 63, 605–627. </bibtext> </blist> <blist> <bibl id="bib18" idref="ref61" type="bt">18</bibl> <bibtext>Lindholm‐Leary, K. (2012). Success and challenges in dual language education. Theory Into Practice, 51, 256–262. </bibtext> </blist> <blist> <bibl id="bib19" idref="ref62" type="bt">19</bibl> <bibtext>Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced approach. Amsterdam: John Benjamins. </bibtext> </blist> <blist> <bibl id="bib20" idref="ref6" type="bt">20</bibl> <bibtext>Marian, V., Shook, A., & Schroeder S. R. (2013). Bilingual two‐way immersion programs benefit academic achievement. Bilingual Research Journal, 36, 167–186. </bibtext> </blist> <blist> <bibl id="bib21" idref="ref28" type="bt">21</bibl> <bibtext>Moeller, A. J., & Theiler, J. (2014). Spoken Spanish language development at the high school level: A mixed‐methods study. Foreign Language Annals, 47, 210–240. </bibtext> </blist> <blist> <bibl id="bib22" idref="ref25" type="bt">22</bibl> <bibtext>Moeller, A. J., & Reschke, C. (1993). A second look at grading and classroom performance: Report of a research study. Modern Language Journal, 77, 163–169. </bibtext> </blist> <blist> <bibl id="bib23" idref="ref26" type="bt">23</bibl> <bibtext>Norris, J. M., & Pfeiffer, P. (2003). Exploring the uses and usefulness of ACTFL Guidelines oral proficiency ratings and standards in college foreign language departments. Foreign Language Annals, 36, 572–581. </bibtext> </blist> <blist> <bibl id="bib24" idref="ref35" type="bt">24</bibl> <bibtext>Padilla, A.M., Fan, L., Xu, X., & Silva, D.M. (2013). A Mandarin/English two‐way immersion program: Language proficiency and academic achievement. Foreign Language Annals, 46, 661–679. </bibtext> </blist> <blist> <bibl id="bib25" idref="ref16" type="bt">25</bibl> <bibtext>Sieloff Magnan, S. (1986). Assessing speaking proficiency in the undergraduate curriculum: Data from French. Foreign Language Annals, 19, 429–437. </bibtext> </blist> <blist> <bibl id="bib26" idref="ref57" type="bt">26</bibl> <bibtext>Sieloff Magnan, S., & Back, M. (2007). Social interaction and linguistic gain during study abroad. Foreign Language Annals, 40, 43–59. </bibtext> </blist> <blist> <bibl id="bib27" idref="ref7" type="bt">27</bibl> <bibtext>Steele, J. L., Slater, R. O., Zamarro, G., Miller, T., Li, J., Burkhauser, S., & Bacon, M. (2017). The effects of dual‐language immersion programs on student achievement: Evidence from lottery data. American Educational Research Journal, 54, 282S–306S. </bibtext> </blist> <blist> <bibl id="bib28" idref="ref27" type="bt">28</bibl> <bibtext>Steinmeyer, G. (1984). Oral proficiency interviews at Amherst Regional High School: Report of a pilot project. Die Unterrichtspraxis / Teaching German, 17, 330–334. </bibtext> </blist> <blist> <bibl id="bib29" idref="ref17" type="bt">29</bibl> <bibtext>Swender, E. (2003). Oral proficiency testing in the real world: Answers to frequently asked questions. Foreign Language Annals, 36, 520–526. </bibtext> </blist> <blist> <bibl id="bib30" idref="ref18" type="bt">30</bibl> <bibtext>Thompson, I. (1996). Assessing foreign language skills: Data from Russian. The Modern Language Journal, 80, 47–65. https://doi.org/10.1111/j.1540‐4781.1996.tb01137.x </bibtext> </blist> <blist> <bibl id="bib31" idref="ref12" type="bt">31</bibl> <bibtext>Tschirner, E. (2016). Listening and reading proficiency levels of college students. Foreign Language Annals, 49, 201–223. </bibtext> </blist> <blist> <bibl id="bib32" idref="ref19" type="bt">32</bibl> <bibtext>Tschirner, E., & Heilenman, L. K. (1998). Reasonable expectations: Oral proficiency goals for intermediate‐level students of German. The Modern Language Journal, 82, 147–158. </bibtext> </blist> <blist> <bibl id="bib33" idref="ref8" type="bt">33</bibl> <bibtext>Watzinger‐Tharp, J., Swenson, K., & Mayne, Z. (2016). The academic achievement of Utah dual language immersion students. International Journal of Bilingual Education and Bilingualism. https://doi.org/10.1080/13670050.2016.1214675 </bibtext> </blist> <blist> <bibl id="bib34" idref="ref20" type="bt">34</bibl> <bibtext>Weinstein, A. (1975). Foreign language majors: The Washington perspective. ADFL Bulletin, 4, 18–27. </bibtext> </blist> <blist> <bibl id="bib35" idref="ref30" type="bt">35</bibl> <bibtext>Xu, X., Padilla, A.M., & Silva, D. M. (2015). Learner performance in Mandarin immersion and high school world language programs: A comparison. Foreign Language Annals, 48, 26–38. </bibtext> </blist> </ref> <p>PHOTO (COLOR): Chinese listening, speaking, and reading score distribution by grade</p> <p>PHOTO (COLOR): French listening, speaking, and reading score distribution by grade</p> <p>PHOTO (COLOR): Two‐way Spanish listening, speaking, and reading score distribution by grade</p> <p>PHOTO (COLOR): One‐way Spanish listening, speaking, and reading score distribution by grade</p> <p>PHOTO (COLOR): Listening, speaking, and reading score changes</p> <aug> <p>By Johanna Watzinger‐Tharp; Fernando Rubio and Douglas S. Tharp</p> </aug>
Header DbId: eric
DbLabel: ERIC
An: EJ1189992
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Linguistic Performance of Dual Language Immersion Students
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Watzinger-Tharp%2C+Johanna%22">Watzinger-Tharp, Johanna</searchLink><br /><searchLink fieldCode="AR" term="%22Rubio%2C+Fernando%22">Rubio, Fernando</searchLink><br /><searchLink fieldCode="AR" term="%22Tharp%2C+Douglas+S%2E%22">Tharp, Douglas S.</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Foreign+Language+Annals%22"><i>Foreign Language Annals</i></searchLink>. Fall 2018 51(3):575-595.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 21
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2018
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Grade+6%22">Grade 6</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Immersion+Programs%22">Immersion Programs</searchLink><br /><searchLink fieldCode="DE" term="%22Bilingual+Education+Programs%22">Bilingual Education Programs</searchLink><br /><searchLink fieldCode="DE" term="%22Grade+6%22">Grade 6</searchLink><br /><searchLink fieldCode="DE" term="%22Chinese%22">Chinese</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Proficiency%22">Language Proficiency</searchLink><br /><searchLink fieldCode="DE" term="%22Longitudinal+Studies%22">Longitudinal Studies</searchLink><br /><searchLink fieldCode="DE" term="%22French%22">French</searchLink><br /><searchLink fieldCode="DE" term="%22Spanish%22">Spanish</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Tests%22">Language Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Program+Evaluation%22">Program Evaluation</searchLink><br /><searchLink fieldCode="DE" term="%22Case+Studies%22">Case Studies</searchLink><br /><searchLink fieldCode="DE" term="%22Interpersonal+Communication%22">Interpersonal Communication</searchLink><br /><searchLink fieldCode="DE" term="%22Linguistic+Performance%22">Linguistic Performance</searchLink><br /><searchLink fieldCode="DE" term="%22Listening+Skills%22">Listening Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Skills%22">Reading Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Oral+Language%22">Oral Language</searchLink><br /><searchLink fieldCode="DE" term="%22Second+Language+Learning%22">Second Language Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Teaching+Methods%22">Teaching Methods</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Policy%22">Educational Policy</searchLink><br /><searchLink fieldCode="DE" term="%22Language+of+Instruction%22">Language of Instruction</searchLink><br /><searchLink fieldCode="DE" term="%22Second+Language+Instruction%22">Second Language Instruction</searchLink><br /><searchLink fieldCode="DE" term="%22Native+Language%22">Native Language</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1111/flan.12354
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0015-718X
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The assessment of dual language immersion (DLI) students' proficiency in the partner language plays a key role in program evaluation. This article reports findings from a large-scale study, drawing on cross-sectional assessment data from 73 schools and longitudinal data from 25 schools. Students were tested in the interpretive (listening and reading) and interpersonal (listening/speaking) modes of communication, using the ACTFL Assessment of Proficiency Towards Performance in Languages. The cross-sectional analysis shows that French and Spanish students performed at Intermediate Low across skills in fourth or fifth grade and Intermediate Mid in sixth grade. Sixth-grade Chinese DLI students attained near Intermediate Mid in listening and Novice High to Intermediate Low in reading and speaking. Across languages, students achieved the highest ratings in listening and, overall, performed more strongly in listening and reading than speaking. A regression analysis of students' change scores over time revealed significant progress in nearly all tested skills and languages. Baseline scores and partner language (Chinese, French, or Spanish) accounted for significant variation in growth rates; between-school differences were found not to be significant. The article discusses implications for assessment instruments and DLI policy and also addresses some instructional implications.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2018
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1189992
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1189992
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1111/flan.12354
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 21
        StartPage: 575
    Subjects:
      – SubjectFull: Immersion Programs
        Type: general
      – SubjectFull: Bilingual Education Programs
        Type: general
      – SubjectFull: Grade 6
        Type: general
      – SubjectFull: Chinese
        Type: general
      – SubjectFull: Language Proficiency
        Type: general
      – SubjectFull: Longitudinal Studies
        Type: general
      – SubjectFull: French
        Type: general
      – SubjectFull: Spanish
        Type: general
      – SubjectFull: Language Tests
        Type: general
      – SubjectFull: Program Evaluation
        Type: general
      – SubjectFull: Case Studies
        Type: general
      – SubjectFull: Interpersonal Communication
        Type: general
      – SubjectFull: Linguistic Performance
        Type: general
      – SubjectFull: Listening Skills
        Type: general
      – SubjectFull: Reading Skills
        Type: general
      – SubjectFull: Oral Language
        Type: general
      – SubjectFull: Second Language Learning
        Type: general
      – SubjectFull: Teaching Methods
        Type: general
      – SubjectFull: Educational Policy
        Type: general
      – SubjectFull: Language of Instruction
        Type: general
      – SubjectFull: Second Language Instruction
        Type: general
      – SubjectFull: Native Language
        Type: general
    Titles:
      – TitleFull: Linguistic Performance of Dual Language Immersion Students
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Watzinger-Tharp, Johanna
      – PersonEntity:
          Name:
            NameFull: Rubio, Fernando
      – PersonEntity:
          Name:
            NameFull: Tharp, Douglas S.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2018
          Identifiers:
            – Type: issn-print
              Value: 0015-718X
          Numbering:
            – Type: volume
              Value: 51
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Foreign Language Annals
              Type: main
ResultId 1