Journal of Linguistics and Language Teaching
Volume 17 (2026) Issue 1
Are We There Yet?
Factors Affecting Proficiency Development for Speaking
and Writing in Intensive English Study
K. James Hartshorn, Wendy Baker-Smemoe, Matthew Millar & Benjamin McMurry (Brigham Young University; Provo (UT) USA)
Abstract
Though millions of students study in intensive language programmes, expected gains and the effects of widely accessible factors such as linguistic distance, L2 proficiency, sex, age, and learning contexts remains unclear. Using the ACTFL proficiency scale, this study analysed the speaking and writing development of 2,325 Novice Mid to Advanced Mid English language learners (CEFR pre-A1 to C1) over a 15-week semester in an intensive English programme. Foreign Service Institute language categories and the Automated Similarity Judgment Program were used to determine linguistic distance. The results show that on average, students progressed approximately one third of a sublevel for speaking and writing and that only lower-proficiency learners (Novice Mid to Intermediate Low) advanced a full proficiency sublevel. Though linguistic distance and age had a slight impact on language development, initial proficiency exerted a substantially stronger effect, with large gains for lower-proficiency learners compared to higher-proficiency students, who slowed or regressed in their measured proficiency. Implications for interpreting and addressing advanced learner needs are discussed.
Keywords: L2 Proficiency, Intensive English Programmes, Linguistic Distance, Speaking, Writing
1 Introduction
The United States hosts more than one million international students annually, with enrollment rates exceeding pre-pandemic levels (Open Doors 2024). Students’ deep aspirations and substantial financial investments create an ethical imperative to provide accurate information about how long study may be required to achieve specified levels of second language (L2) proficiency. To provide such insight, researchers must understand general patterns of L2 development and the factors that influence language learning. Nevertheless, as Jaekel et al. (2023) observe, all students are often expected to learn at the same pace regardless of factors such as L1 background, gender, and age. Moreover, the literature remains unclear as to whether students in intensive programmes achieve greater language gains than those in extensive programmes (Goertler et al., 2016; Juffs, 2020; Liu & Yin, 2021; Serrano, 2022; Winke et al., 2020).
Therefore, the present study seeks to address this gap by examining the effects of demographic factors on L2 language speaking and writing development including linguistic distance (LD) between the learner’s L1 and L2, the initial L2 proficiency level, and the learners’ sex and age. The findings should benefit students, teachers, and programme administrators seeking to optimise language learning, and may also be of interest to researchers and theorists investigating L2 development.
2 Review of Literature
2.1 Measuring L2 Proficiency
Accurate measurement of L2 proficiency gains is essential for understanding student language development. When assessing language proficiency, scholars and educators often use frameworks widely accepted by academic institutions and government entities such as the Interagency Language Roundtable (ILR), the Common European Framework of Reference for Languages (CEFR), and the American Council on the Teaching of Foreign Languages (ACTFL). Accordingly, data in this study are based on slightly modified ACTFL standards for proficiency, which have been well documented as valid and reliable measures of language proficiency (e.g., Alpine Testing Solutions 2020a, 2020b).
The ACTFL scale presents proficiency levels for speaking, listening, reading, and writing (ACTFL, 2012). Trained raters assign proficiency levels to students on a scale that includes five levels: Novice, Intermediate, Advanced, Superior, and Distinguished. The Novice, Intermediate, and Advanced levels are further divided into three sublevels – Low, Mid, and High. Ratings are given based on students’ ability to perform specific language functions. For example, a Novice Mid learner is expected to be able to communicate primarily through memorised phrases, and an Intermediate Mid learner can communicate with uncomplicated language. However, an Advanced Mid learner is expected to narrate, describe, use past and future tenses, and handle complicated situations.
2.1.1 Programme Analyses of Expected Proficiency Gains
2.1.1.1 University Foreign Language Programmes
Though frameworks such as ACTFL assess language proficiency at a given time, they do not specify how quickly a learner should move from one level to the next, nor whether learners at various proficiency levels should all progress at the same rate. Studies have attempted to determine the language proficiency of students at various stages in university foreign language programmes for Russian (Thompson 1996), German (Goertler et al. 2016), Spanish (Viera & Arispe 2021), and multiple languages (Isbell et al. 2019; Winke et al. 2020). Though outcomes have varied somewhat across programmes, most have demonstrated that students were able to reach a Novice-High level after one year, Intermediate-Low after two, Intermediate-High after three, and Advanced-Low after four years (e.g., Goertler et al., 2016). Nevertheless, Winke et al. (2020) found that in studying programs for Spanish, French, German, Italian, and Russian at multiple universities, only Spanish learners achieved Advanced Low in speaking skills by the fourth year. Those studying the other languages achieved Intermediate Low to Intermediate High.
2.1.1.2 Intensive Language Programmes
Unlike university foreign language study where the target language may be just one part of a broader curriculum, intensive language programmes focus exclusively on advancing L2 language development as rapidly as possible. Though studies are limited, most have compared intensive programme outcomes with more traditional or extensive classroom experiences. Reviews by Liu and Yin (2021) and Serrano (2022) suggest that, given the same number of hours of language instruction, students in intensive programmes usually outperform those in other language learning contexts in vocabulary acquisition, oral and written skills, and overall proficiency. Since most of these studies were conducted with children (e.g., Collins et al., 1999), or involved small adult samples (e.g., Serrano, 2011), more studies are needed.
Such analyses of adult intensive language programmes are important for understanding how L2 skills develop. Juffs (2020) conducted an extensive examination of students enrolled in Intensive English Programmes (IEPs) across five semesters. Although the findings of this study were enlightening, standardised testing such as ACTFL, which could be used for comparison with other programmes, was not employed. While previous research on foreign language programmes suggests that students may achieve one sublevel gain on the ACTFL proficiency scale over a semester, we are unaware of any similar large-scale analyses in an IEP context.
2.2 Measuring Linguistic Distance
This study examines the effect of L1 and L2 differences on L2 learning. In describing language differences, some scholars have used the term language distance (e.g., Corder 1979, 1981) while others have used linguistic distance (e.g., Chiswick & Miller, 2005). Though technical differences between these terms could be argued, given the overlap in usage in the present context, we use the more general term linguistic distance, described as a measure of difference between languages in terms of vocabulary, pronunciation, grammar, and other features (Xinxin et al., 2022). Understandably, such measures may be difficult to calculate given the complexity of language. In this study, we consider two measures including the U.S. State Department’s Foreign Service Institute’s classification of languages (Hart-Gonzalez & Lindemann, 1993) and the Automated Similarly Judgment Program (ASJP; Wichmann et al., 2022). Each of these will be described briefly below.
2.2.1 U.S. Foreign Service Institute Language Classifications
After extensive research and experience, the U.S. State Department’s Foreign Service Institute (FSI) (n.d.) has established the number of classroom hours needed for native English speakers to achieve a particular proficiency level (Hart-Gonzalez & Lindemann, 1993). In this framework, the key to success is based on time on task and learning intensity (Jackson & Kaplan, 1999). For example, learners with an average to superior aptitude will need 240 hours of instruction to reach an Intermediate Mid proficiency level for languages such as Spanish, Swahili, or Swedish. By contrast, it may take 480 hours of instruction to gain an Intermediate Low proficiency level for native English-speaking students who are learning languages such as Arabic, Chinese, or Korean. All major languages taught by the FSI have been placed into one of four categories where Category I languages take the least time for native English speakers and Category IV languages take the most time. Though not without some criticism (Van der Slik 2010), this language categorisation functions as a type of empirical linguistic distance shown to be an effective and reliable quantitative measure (Chiswick & Miller, 2005).
2.2.2 Automatic Similarity Judgement as Linguistic Distance
Another approach to linguistic distance comes from the Automatic Similarity Judgement Program (ASJP; Wichmann et al., 2022). The ASJP is a collaborative work that uses lexicostatistical computational methods to analyse words from different languages phonetically. The process computes a Levenshtein distance based on the insertions, deletions, or substitutions by which the words in the two languages differ (Levenshtein, 1965). Differences are normalised to account for word length and chance similarities between words. There is growing interest in using ASJP methods for determining linguistic distance (e.g., Jaekel et al., 2023; Xinxin et al., 2022). The ASJP calculation results in a measure of linguistic distance where a smaller number means less distance and a larger number represents a greater distance. For example, using the ASJP method, the distance between English and Swedish is .62 while the distance between English and Chinese is 1.0.
2.2.3 Comparative Studies of Linguistic Distance
Some studies of linguistic distance suggest it may play a significant role in language development. For example, Bernhardt et al. (2015) found that native English speakers enrolled in Romance foreign language classes scored at roughly Intermediate Mid in writing while those enrolled in languages with greater linguistic distance to English such as Arabic, Chinese, and Japanese scored at Intermediate Low. Similarly, Juffs (2020) observed that students with a Romance language background were more likely to progress in some areas than were students of other language backgrounds. Xinxin et al. (2022) concluded that linguistic distance impacts English language learning, more so with the productive skills of speaking and writing.
Despite these findings, other researchers have found little or no effect of foreign language studied when examining the effect of linguistic distance for learners of French, Spanish, Russian, and Chinese (e.g., Isbell et al., 2019). Moreover, Elder & Davies (2020) concluded that though linguistic distance seemed to be a subtle factor in language development, its effects could not be isolated clearly enough to identify any clear pedagogical implications. Thus, the effect of linguistic distance on proficiency gains in the four skill areas remains unclear. Additional research is essential if we are to clarify the effects of linguistic distance on L2 development.
2.3 Initial Proficiency and Language Gains
Another factor that may affect the rate of language development is the learner’s proficiency level. Several studies have demonstrated greater proficiency gains for students with lower proficiency, including several discussed above (e.g., Bernhardt et al., 2015; Goertler et al., 2016; Winke et al., 2020; Juffs, 2020). While research continues, the current trend suggests that the rate of student progress slows as learners become more proficient (Danesh & Shahnazari 2020). Isbell et al. (2019) concluded that the most likely predictor of no gain or decreased proficiency for foreign language students at the posttest was a higher proficiency at the pretest. They also found that lower proficiency students were more likely to make progress over a semester than were the higher proficiency students. Thus, examining the effect of initial language proficiency at an IEP could yield important insights about how this factor influences the rate of language development.
2.3.1 The Effects of Learner’s Sex on Language Development
Though some evidence suggests that the biological sex of learners may play a role in L2 learning, the results appear to be complex and inconclusive (Chan, 2021). Some studies suggest that females outperform males in reading, writing, and listening (Burstall, 1975), in oral proficiency and pronunciation (Ehrlich, 1997), and in overall proficiency gains (Boyle, 1987). Główka (2014) noted that girls outperformed boys in various aspects of language development regardless of whether those differences were explicitly recognised by teachers or the students themselves. Scholars have observed that males and females tend to process language learning in different ways (Burman et al., 2008) and that even from birth, females may be better equipped to learn a second language (Dionne et al., 2003). Green & Oxford (1995) also noted that female students utilise language learning strategies more than male students.
Van der Silk et al. (2015) observed that female learners outperformed males in speaking and writing tasks, but that the male students performed better in listening and reading. Arabski & Wojtaszek (2011) have suggested that male learners benefit the most from learning that involves sensory information associated with processing activities such as reading, writing, video, or auditory tasks. They also suggest that while males are more likely to outperform females on speaking tasks when they have visual stimuli, females may be better equipped to succeed with grammar tasks, abstract speech, and integrated tasks.
While these studies suggest observable differences between male and female learners, findings across studies are not consistent. Other studies have not demonstrated meaningful differences in language development in men and women across a variety of contexts (Fattahi & Nushi, 2021; Green & Oxford, 1995; Zughoul & Taminian, 1984). A conspicuous lack of differences between males and females has even been noted while researchers have observed stark differences in strategy use and learning processes (Maghsudi et al., 2015). Van der Silk et al. (2015) have concluded that while there may be notable differences between male and female language acquisition in their earlier years, these differences may fade as students get older. More research is thus needed to help clarify the role of sex on L2 development.
2.3.2 The Effects of Age in Adulthood on Second Language Development
Research suggests that most young children learn a second language with great facility (e.g., Sierens et al., 2019) but that after puberty, language learning tends to become more difficult (e.g., Xue et al., 2021). It also has shown that while adults may struggle with implicit learning that children seem to manage easily, adults benefit from more advanced cognitive and metacognitive abilities that allow them to be more aware, strategic, and self-directed as they monitor their purposes, motivations, and the use of the language itself (Dekeyser, 2018). Research also suggests that while it may take older adults more time to learn, they tend to be successful in most aspects of second language learning (Van der Ploeg, 2023), though young adults may have some short-term advantages over older adults (Cox & Sanz, 2015). Though such findings are informative regarding age extremes, the available research says very little about the effect of age among young to middle-aged adults. Such insight would also be extremely valuable in understanding L2 development.
2.3.3 Proficiency Measures for Speaking and Writing
Most studies suggest that different language skills may not develop at the same rate. For example, Bernhardt et al. (2015) examined both speaking and writing for over 2000 language learners studying several foreign languages. They found that writing ability generally surpassed speaking ability after two years. Other studies have suggested that these two skills simply develop differently depending on individual factors such as experience and aptitude. Weissberg (2006) noted that students of English as a Second Language (ESL) develop either speaking before writing, writing before speaking, or the two skills develop simultaneously depending on factors such as previous experience with the L2 and personality. Hubert (2013) also observed that different students progressed in the three ways described by Weissberg across all proficiency levels.
Similarly, Thompson (1996) examined ACTFL gains in the speaking, listening, reading, and writing across five years of study. She found that while the average OPI proficiency level advanced each year for both speaking and writing, writing developed faster than speaking. For example, after one year in the programme, the median writing score was Novice High, while for speaking it was Novice Mid. However, for learners tested after five years, the median speaking score was Advanced to Advanced High, while the median writing score was Intermediate High to Advanced Low. Thus, though speaking developed more slowly than writing at first, in the end, students gained more proficiency in speaking than writing.
This review of literature has shown that the development of speaking and writing in programmes such as IEPs is still not well understood due to limited research and inconsistent results. Moreover, the specific effects of linguistic distance, language proficiency, sex, and age need to be better understood for IEP practitioners and administrators to optimise the learning experiences of their students. Even where the literature may suggest partial answers to some of the questions of interest in this study, the field must accrue a critical mass of consistent research results before we can draw meaningful conclusions about the language learning phenomena we investigate (Porte & McManus, 2018).
3 Research Questions
Given this review and the current gaps in the literature, this study seeks to examine the following research questions:
How well correlated is the construct of linguistic distance based on the FSI language categories and the ASJP score?
To what extent do English language learners gain proficiency in the skills of speaking and writing over the course of one semester and are gains equal for both speaking and writing?
Do students in an intensive programme progress in proficiency gains at the same rate as students in an extensive programme?
To what extent do the factors of relative linguistic distance, initial proficiency, sex, and age affect ESL proficiency gains in speaking and writing over one semester?
This section has described the need for researchers and language educators to better understand the factors that influence L2 development in the productive skills of speaking and writing across intensive and extensive L2 learning contexts. Literature was presented regarding the measure of L2 proficiency, intensive and extensive programmes, the notion of linguistic distance, and factors that may affect L2 development such as L1, initial L2 proficiency levels, and the sex and age of the learners. Based on this literature, the four preceding research questions were articulated and will be addressed in the subsequent sections.
4 Methods and Materials
4.1 Student Data
This study used preexisting pretest and posttest data from equivalent test forms from 2,325 students who studied in an Intensive English Programme (IEP) at a large university in the western United States. These anonymised data had no personal identifying information and met all Institutional Review Board (IRB) requirements.
Table 1 presents linguistic distance (LD) based on FSI language categories, along with the number and percentage of students in each category: Category 1 = 1,624 (69.85%), Category 2 = 27 (1.16%), Category 3 = 140 (6.02%), and Category 4 = 534 (22.97%). Student ages ranged from 17 to 64 (M = 25.55, SD = 7.10), with somewhat more female students (56%) compared to male students (44%):
L1 | Number | Percentages | FSI |
Spanish | 1,321 | 56.82% | 1 |
Portuguese | 272 | 11.70% | 1 |
Chinese | 203 | 8.73% | 4 |
Korean | 163 | 7.01% | 4 |
Japanese | 159 | 6.84% | 4 |
Russian | 40 | 1.72% | 3 |
Mongolian | 32 | 1.38% | 3 |
Thai | 27 | 1.16% | 3 |
Haitian Creole | 25 | 1.08% | 2 |
French | 25 | 1.08% | 1 |
Arabic | 9 | 0.39% | 4 |
Ukrainian | 6 | 0.26% | 3 |
Italian | 5 | 0.22% | 1 |
Albanian | 4 | 0.17% | 3 |
Chuvash | 4 | 0.17% | 3 |
Hungarian | 4 | 0.17% | 3 |
Farsi | 3 | 0.13% | 3 |
Malagasy | 3 | 0.13% | 3 |
Turkman | 3 | 0.13% | 3 |
Chechen | 2 | 0.09% | 3 |
Kazakh | 2 | 0.09% | 3 |
Slavic languages[1] | 2 | 0.09% | 3 |
Vietnamese | 2 | 0.09% | 3 |
Armenian | 1 | 0.04% | 3 |
Bengali | 1 | 0.04% | 3 |
Burmese | 1 | 0.04% | 3 |
Central Khmer | 1 | 0.04% | 3 |
German | 1 | 0.04% | 2 |
Indonesian | 1 | 0.04% | 2 |
Swedish | 1 | 0.04% | 1 |
Tagalog | 1 | 0.04% | 3 |
Tajik | 1 | 0.04% | 3 |
Total | 2325 | 100.00% |
|
Table 1: Participants by L1 and LD
Student English language proficiency ranged from Novice Low to Advanced Low. Although proficiency tests examined the skills of listening, reading, speaking, and writing, the analyses in this study are limited to the productive skills of speaking and writing. After eliminating the students that did not have both pretest and posttest scores, data from 2,325 students were analysed. Table 2 shows the number of students at each ACTFL proficiency level at the pretest along with CEFR approximates. For the purposes of this study, each proficiency level was assigned a number: Novice Low-Mid = 1, Novice High = 2, Intermediate Low = 3, Intermediate Mid = 4, Intermediate High = 5, Advanced Low = 6:
ACTFL Level | CEFR* | N | Percentages |
Novice Low-Mid | Pre-A1 | 40 | 1.72% |
Novice High | A2 | 366 | 15.74% |
Intermediate Low | B1 | 861 | 37.03% |
Intermediate Mid | B1 | 745 | 32.04% |
Intermediate High | B2 | 277 | 11.91% |
Advanced Low | B2 | 36 | 1.55% |
Total |
| 2,325 | 100.00% |
Table 2: Students According to Level
(*CEFR proficiency levels are approximate)
4.2 IEP Context
Participants in this study were enrolled full-time in an Intensive English Program (IEP). The target class size was 15 students though some classes were slightly larger or smaller depending on placement testing. The curriculum included four courses – Reading, Writing, Listening and Speaking, and Grammar. Each class met for 65 minutes Monday through Thursday with additional quizzes, exams, or other activities held on Fridays for a total of 18 to 22 hours per week. Though most students hope to study at an English-medium university in North America, a minority simply desired to improve their English well enough to use it in their employment in their home country.
While in the programme, these students lived in regular housing surrounding the university that hosted the IEP. Most had native English-speaking roommates and followed the encouragement of programme teachers and administrators to explore, be involved, and actively participate in opportunities in the community to interact with native English speakers. In the classroom, teachers not only worked to increase student knowledge of English, but they consistently helped students practice using the language consistent with principles associated with skill acquisition theory (DeKeyser, 2020).
4.3 Instruments
All data for this study were collected under the same conditions in the same computer laboratory within the IEP.
Students were given thirty minutes to respond to a specific writing task such as:
Some people say that physical exercise should be a required part of every school day. Other people believe that students should spend the whole school day on academic studies. Which opinion do you agree with? Use specific reasons and details to support your opinion and describe the potential immediate and long-term consequences of this opinion.
In-house software provided the writing task, simple directions, and space for students to type their responses. The software included basic editing tools such as copy, cut, and paste functions. Nevertheless, students had no access to the internet or other tools that would evaluate or correct spelling or grammar choices. The task window closed after thirty minutes, so responses that were not submitted by that time were then automatically recorded with no additional opportunities for editing.
For speaking, students used headsets to record twelve proficiency-specific prompts (ACTFL, 2012) as illustrated in Table 3 where the sequence, number of tasks, task proficiency level, and the time allowed are presented. Recording was started and stopped automatically by the software.
Order | Number of Tasks | Task Proficiency Level | Preparation Time | Speaking Time |
1 | 1 | Novice | 0:10 | 0:30 |
2 | 3 | Intermediate | 0:10 | 0:30 |
3 | 1 | Advanced | 0:15 | 0:45 |
4 | 1 | Superior | 0:20 | 1:20 |
5 | 1 | Advanced | 0:15 | 0:45 |
6 | 2 | Superior | 0:20 | 1:20 |
7 | 1 | Advanced | 0:15 | 0:45 |
8 | 2 | Intermediate | 0:10 | 0:30 |
Table 3: Task Level and Allowed Time for Speaking Tasks
4.4 Ratings
Reliability of ratings is essential for valid analysis and interpretation (e.g., Knoch & Chapelle, 2017). The IEP completed the rating prior to providing data to the researchers. The IEP reported several steps used to assign an appropriate score for speaking and writing for each student. To maximise the consistency of rater scores, well developed rubrics were used based on ACTFL proficiency guidelines. The rubrics show the requirements for each proficiency level, the program’s operationalisation of these requirements, and the ACTFL scale equivalents (Appendices A and B).
Raters were programme teachers and administrators who were well trained and had participated in extensive calibration practice (McNamara, 1996). Some raters were also ACTFL certified. Each composition was evaluated by a minimum of two human raters with additional raters added for unusual cases where scores were separated by more than one proficiency sub-level. To account for rater bias, Facets software (Linacre, 2018) utilised Many Facets Rasch Modeling (MFRM) to generate fair averages from the raw scores provided by the raters, thus compensating for rater[2] variability (e.g., Eckes 2011). These adjusted rubric scores for speaking and writing were used in subsequent analyses.
4.5 Numerical and Categorical Interpretations of Proficiency
As we intended to use statistical analyses to examine language development, proficiency was quantified from expert ratings based on the rubrics provided in Appendixes A and B and their subsequent fair averages generated by Rasch modelling. However, we recognised that it could also be reasonable to interpret proficiency differently from a categorical approach based on the rubric ranges such as Novice High (1-1.99), Intermediate Low (2-2.99), Intermediate Mid (3-3.99), and so forth. For example, consider the Intermediate Mid proficiency range. With a quantitative interpretation of proficiency used for statistical analysis, we might reasonably expect a score of 3.9 to suggest greater proficiency than a score of 3.1. However, with a categorical interpretation, scores of 3.1 and 3.9 might reasonably be considered as the same Intermediate Mid proficiency level. Since the numerical and categorical interpretations of language proficiency both have strengths and limitations, we account for both within this study.
5 Results
This section presents the results that help answer our four research questions beginning with general results and concluding with the potential impact of factors of interest that may influence proficiency gains.
5.1 General Results
Before examining the potential impact of our factors of interest on the proficiency gains of English language learners, this section presents results to our first three research questions associated with the correlation of FSI language categories and ASJP scores, the comparability of speaking and writing gains, and the comparability of proficiency gains in intensive and extensive programs.
The first research question addressed the correlation between the FSI language categories and the ASJP as a measure of linguistic distance. We calculated the Spearman’s rho correlation coefficient for all cases for which we had rankings for the four FSI language categories and the ASJP scores comparing English and other languages published by Xinxin et al. (2022), which included 14 categories for our data. Findings suggest that these two measures of linguistic distance are well correlated, rs(2245) = .876, p < .001. Given the high correlation, subsequent analyses were limited to the FSI categories since they are more well known and widely used.
The second research question examined English proficiency gains for speaking and writing over one 15-week semester, and whether gains were equal for both speaking and writing. Using the numerical interpretation of language proficiency described previously, we observed that on average, students substantially increased their proficiency for speaking, F(1,2324) = 474, p < .001, (ηp2 = .169), as well as for writing, F(1,2324) = 367, p < .001, (ηp2 = .136). For speaking, mean proficiency levels increased from 3.91 (SD = .971) to 4.26 (SD = .981). In other words, on average, students progressed from Intermediate Low (but near the threshold of Intermediate Mid) to well within the Intermediate Mid range. The results for writing mirrored those of speaking, with mean proficiency scores increasing from 3.92 (SD =.990) to 4.26 (SD = .970).
Since a gain score of 1.0 represents one full sublevel with the numerical interpretation of language proficiency, these results show that overall, students only progressed about one-third of a sublevel for speaking (M gain = .35) and for writing (M gain = .34). These results also show that gains were virtually the same for both speaking and writing. Only 19.43% of the students advanced one full proficiency sublevel or more for speaking, and 21.88% advanced one full proficiency sublevel or more for writing.
The third research question addressed whether students in an intensive programme progressed in proficiency gains at the same rate as students in extensive programs. Though our data sample was large, our conclusions remain tenuous since our data was historical rather than experimental. Results from our analyses show that most Novice Low-Mid proficiency students in this study progressed to the next sublevel within one semester (15 weeks) for speaking (80%, M gain = .93) and writing (70%, M gain = .77). In contrast, the available literature suggests that most students in foreign language programmes typically need approximately one year to reach the same gains at the outset of their study (Goertler et al., 2016; Thompson, 1995; Winke et al., 2022). Thus, these findings suggest that, at least at lower proficiency levels, students may be more likely to progress more rapidly in intensive programmes than in extensive programmes (e.g., Collins et al., 1999).
The fourth question examined the extent to which proficiency gains in speaking and writing over one semester are influenced by the factors of linguistic distance, proficiency level, sex, and age. Multiple linear regression was used for these analyses. Since not all students exhibited the same proficiency level for speaking and writing, separate analyses were conducted for speaking and writing gains.
5.2 Gains in Speaking
Results suggest that the four factors examined in this study account for just over 16% of the variability associated with speaking gains (adjusted R2 = .162), F(4,2319) = 113, p <.001. Table 4 breaks down the relative effect of each explanatory variable. These findings suggest that, though sex had no significant effect on speaking gains, the other remaining predictors were meaningful to varying degrees. The standardised betas (β) provide the relative contribution toward speaking gains from each variable. They suggest the weakest contribution was from linguistic distance, followed by the student’s age, with the strongest effect from initial student proficiency (more than 3.7 times the effect of age and about 4.6 times the effect of linguistic distance). Moreover, these negative β values indicate that greater speaking gains were observed for younger learners with lower-proficiency who encountered smaller linguistic distances:
Predictor | B | SE | t | p | β |
Intercept | 1.858 | .088 | 21.059 | < .001 | — |
Proficiency | -0.323 | .016 | -20.871 | < .001 | -0.410 |
Student Age | -0.012 | .002 | -5.690 | < .001 | -0.109 |
Ling. Distance | -0.054 | .012 | -4.503 | < .001 | -0.089 |
Sex (Male) | 0.011 | .030 | 0.347 | 0.729 | 0.013 |
Table 4: Linear Regression for Predictors Affecting Speaking Gains
5.3 Gains in Writing
The four predictors examined in this study account for just over 19% of the variability associated with writing gains (adjusted R2 = .192), F(4,2319) = 139, p <.001. Table 5 breaks down the relative effect of each explanatory variable. Neither student age, nor sex were statistically significant factors for writing gains at the traditional .05 alpha, though both linguistic distance and initial proficiency were more meaningful predictors. Standardised betas show that initial proficiency had nearly four times the impact on writing gains compared to linguistic distance where lower-proficiency learners facing less linguistic distance between the L1 and L2 experienced the greatest writing gains.
Predictor | B | SE | t | p | β |
Intercept | 1.923 | 0.092 | 20.950 | < .001 |
|
Proficiency | -0.378 | 0.016 | -23.370 | < .001 | -0.441 |
Ling. Distance | -0.075 | 0.013 | -5.850 | < .001 | -0.111 |
Sex (Male) | -0.059 | 0.033 | -1.790 | 0.073 | -0.068 |
Student Age | -0.004 | 0.002 | -1.930 | 0.054 | -0.036 |
Table 5: Linear Regression for Predictors Affecting Writing Gains
Since initial proficiency was the most meaningful predictor of subsequent proficiency gains, we report our findings for both the numerical and categorical interpretation of proficiency gains to facilitate comparisons of our results with other studies. Thus, in addition to comparing numerical representations of proficiency based on Rasch modelling fair averages, we present, for example, the percentage of Novice Low-Mid students who progressed from a score between 1.00 and 1.99 to a score from 2.00 to 2.99 during the semester of interest. We determined this percentage for both speaking and writing and for all six proficiency levels (Novice Low-Mid to Advanced Low). As demonstrated in Table 6, a large majority of Novice students advanced one sublevel in both speaking and writing while only about half of the Intermediate Low students advanced one complete sublevel. Nevertheless, many fewer students who were pretested at Intermediate Mid and Intermediate High made the same progress while no Advanced Low students progressed to Advanced Mid during the semester:
|
| Pretest | Posttest | Gain Scores |
Gain of Sublevel | |||
|
| M | SD | M | SD | M | SD | |
Speaking | Novice Low-Mid | 1.61 | 0.46 | 2.54 | 0.75 | 0.92 | 0.78 | 80% |
| Novice High | 2.61 | 0.30 | 3.30 | 0.71 | 0.69 | 0.66 | 68% |
| Intermediate Low | 3.52 | 0.29 | 4.03 | 0.76 | 0.51 | 0.73 | 48% |
| Intermediate Mid | 4.45 | 0.30 | 4.69 | 0.77 | 0.24 | 0.74 | 35% |
| Intermediate High | 5.38 | 0.28 | 5.22 | 0.67 | -0.16 | 0.69 | 12% |
| Advanced Low | 6.33 | 0.23 | 5.24 | 1.04 | -1.09 | 1.00 | 0% |
|
|
|
|
|
|
|
|
|
Writing | Novice Low-Mid | 1.47 | 0.43 | 2.24 | 0.83 | 0.77 | 0.94 | 70% |
| Novice High | 2.63 | 0.33 | 3.46 | 0.78 | 0.83 | 0.77 | 71% |
| Intermediate Low | 3.52 | 0.33 | 4.11 | 0.78 | 0.59 | 0.78 | 55% |
| Intermediate Mid | 4.43 | 0.30 | 4.57 | 0.76 | 0.13 | 0.78 | 28% |
| Intermediate High | 5.36 | 0.28 | 5.11 | 0.80 | -0.25 | 0.82 | 14% |
| Advanced Low | 6.30 | 0.24 | 5.35 | 0.78 | -0.95 | 0.78 | 0% |
Table 6: Descriptive Statistics for Speaking and Writing by Proficiency
5.4 A Posteriori Interviews
Though not part of the original research design, to better understand and contextualise the unexpected findings of stagnation and regression of some of the highest-proficiency learners observed in this study, brief a posteriori interviews were conducted with the Program Director (55-year old male with more than 30 years of experience), the Curriculum Coordinator (40-year old male with nearly 20 years of experience) and the two instructors of the highest proficiency students (both females in their early 30s with nearly ten years of experience).
One administrator confirmed that many of the highest-proficiency learners face unique challenges as they complete their study at an IEP. He noted that distractions for higher-proficiency students can include preparing for the TOEFL test, completing university applications, the need to earn additional resources for tuition at a new school, extra time needed to make housing arrangements and moving their residence, as well as additional distractions for those who are married or have children. One teacher observed that as soon as students feel they have achieved their immediate IEP goals, they often adjust their focus toward additional schooling or career advancement. One of the teachers observed that even students who were once diligent and fully engaged in every class will often become less focused before the end of their last semester once they are admitted into a university or get a promotion at their job.
Another administrator noted that some preoccupied students may not take the posttest seriously during their last semester or may skip it altogether as they adjust their focus to other priorities. Such a phenomenon could substantially impact how language gains are interpreted if scores from the highest performing students are missing or incomplete. If widespread, interpretations of such data could be systematically skewed. Insightfully, one administrator with significant experience at two other IEPs conceded that, in the institutions in which he worked, it was common for many of the highest performing students not to take their final exams in their last semester. This could be one reason students may appear to stagnate or regress as seen in other studies (e.g., Isbell et al., 2019; Winke et al., 2022). Further research may be needed to help clarify actual proficiency gains of students with advanced proficiency at the conclusion of their IEP study.
6 Discussion
The results of the first research question addressed the comparability of FSI language categories and the ASJP scores as measures of linguistic distance. Though both measure linguistic distance, they are determined in different ways. Whereas the FSI language categories are based on empirical test data from actual language learners, the ASJP scores are derived from phonetic differences identified through automated lexical comparisons. Nevertheless, the strong correlation suggests that the two seem to measure the same construct. Though we recognise that future results could vary somewhat from those observed here, this finding suggests that one of these measures might function adequately as a proxy for the other. It also suggests that though FSI categories were specifically devised for native English speakers learning foreign languages, it may be appropriate to use the FSI categories as a bidirectional measure of linguistic distance with utility for students learning English from one of the listed L1s.
The findings of this study showed that gains were similar for both speaking and writing, and that on average, students could expect to progress approximately one third of a full proficiency sublevel. Nevertheless, at the lowest proficiency levels, most students advanced a full level though such progress was much less common at higher proficiencies. While most lower proficiency students in this study progressed to the next sublevel within one semester, on average, most students in foreign language programmes tend to take approximately one year to reach the same gains (Thompson, 1995; Winke et al., 2022). This finding suggests that, at least at lower proficiency levels, students may progress more quickly in intensive than extensive programmes (e.g., Collins et al. 1999). After the Intermediate Mid level, however, the IEP students in this study did not progress as quickly.
The second research question also addressed whether students make the same progress in both speaking and writing. Unlike Thompson (1996) and Bernardt et al. (2016), we did not find that writing developed more quickly than speaking. Our data suggests that both skills developed at approximately the same rate. That is, many students in the lower proficiency levels advanced approximately one sublevel each semester, while students in the more advanced levels appeared to stagnate or even regress in both skills. Since differences in speaking and writing in other settings could be attributed to curricular inconsistencies or differences in how language gains were measured, additional study may be needed to further clarify this.
The third research question addressed whether students in an intensive program progress in proficiency gains at the same rate as students in an extensive programme. Our results suggested that some learners in an intensive program may progress up to two or three times more rapidly than learners in an extensive program – at least at lower proficiency levels. These findings seem consistent with other recent studies indicating that students in intensive programmes tend to make faster progress than students in extensive programmes (Liu & Yin, 2021; Serrano, 2022).
Nevertheless, our findings also show that, on average across all proficiency levels, the intensive students examined in this study progressed about one third of a full proficiency sublevel, which is the functional equivalent of advancing one proficiency sublevel over the course of one year. Thus, examining all learners at the same time without regard for proficiency in the comparison of language gains by programmes could oversimplify the phenomena and make intensive and extensive programmes appear to be more similar than they might actually be. Although the comparison provided here can only be considered tentative since no experiment was conducted, these findings provide enough evidence to warrant more careful experimental studies that ensure the comparability of groups, that effectively control for extraneous variables, and that take proficiency levels into account.
The fourth research question discussed the extent to which linguistic distance, initial proficiency, sex, and age might affect proficiency development in speaking and writing over one semester. Although evidence abounds that males and females differ in their processing and approaches to language learning (e.g., Arabski & Wojtaszek, 2011; Dionne et al., 2003; Green & Oxford, 1995), these differences appeared to have no noticeable effect on language gains in this study. As Van der Slik et al. (2015) has suggested, it is possible that whatever advantages females may have had in younger years (e.g., Boyle 1987; Burstall 1975; Ehrlich 1997; Główka 2014), may become less beneficial once learners reach adulthood.
Although learner age had no noticeable effect on writing gains, there was a meaningful but minor effect for age on speaking where younger adult learners advanced more than their older adult counterparts. While more research is needed in this area, these findings may suggest that the strong advantages that children have in language acquisition (e.g., Sierens et al., 2019; Xue et al., 2021) may persist in a much weaker form into adulthood and continue to decrease with age. Although it is unclear why age had a small effect on speaking but not writing, it is possible that this difference may be due to the nature of writing itself. Unlike speaking, which can be acquired naturally in the L1 and the L2, writing must be taught and learned explicitly. Many aspects of writing – and academic writing in particular – are governed by conventions requiring a great deal of study and practice. In this context, potential age-related advantages for younger language learners might be overshadowed by cognitive strategies, motivation to improve in the use of writing conventions, and the fact that most writing tasks allow for more monitoring and revision than speaking tasks.
Linguistic distance had a minor effect on both speaking and writing. Although many studies suggest a strong relationship between linguistic distance and language development (e.g., Bernhardt et al., 2015; Chiswick & Miller, 2005; Hart-Gonzalez & Lindemann, 1993; Juffs, 2020), generally, our findings seem more consistent with those of other scholars who observed linguistic distance to be a meaningful but weak factor in language development (e.g., Elder & Davies, 2010; Isbell et al., 2019). There could be several reasons for this:
The effects of linguistic distance may be more obvious for single features of the second language that differ from the L1, such as the pronunciation of a single sound or specific vocabulary (e.g., Georgiou, 2021; Juffs, 2020).
It could be that abundant language practice across skills (DeKeyser, 2020), as was the case in the examined IEP, might soften the effects of language distance.
Moreover, the general measures of speaking and writing proficiency used in this study may not be granular enough to capture stronger effects of linguistic distance without more longitudinal data.
The initial proficiency level of the learners proved to be the most substantial predictor of language gains we observed for both speaking and writing. That is, on average, the lower the initial proficiency, the greater were the gains that could be expected. For speaking, the relationship between initial proficiency level and speaking gains over one semester was a perfect negative correlation. For writing, the relationship was similar, r(4) = -0.943, p = .005. This result suggests that proficiency may be useful as a general predictor of language gains for both speaking and writing. These findings seem consistent with other studies examining overall language development in foreign language programmes, where higher proficiency level students tended to stagnate or even regress (e.g., Isbell et al., 2019; Winke et al., 2022).
Since this study included observations of systematic stagnation or even regression of proficiency in high-proficiency students, we discuss a few possible reasons for this. As proficiency grows, assessments tend to require progressively more complex linguistic demands which often carry a greater cognitive load that could result in lower scores as suggested by Joyce (2019). Another possibility could be tied to changes in motivation due to increased anxiety to perform well as suggested by Weng and Liu (2024). Moreover, Wardani et al. (2020) noted that, while students who are motivated to learn will be successful, advanced proficiency learners may be less motivated to improve if they perceive that their language skills are adequate. In addition, Baker (2019) examined IEP student experiences that affect motivation such as money matters, loneliness, juggling homework and test preparation, anxiety, and so forth. Higher-proficiency students may have more difficulty with such factors than lower-proficiency students, particularly as they face new complexities in their preparations to transition to new employment or further study in an English-medium university.
Meanwhile, the results of this study may be valuable to programme administrators in two specific ways:
We encourage programme administrators to be inquisitive and observant regarding students in their program with advanced proficiency. Doing so may facilitate the development of useful hypotheses regarding proficiency gains in advanced learners and lead to important refinements to curricula. Advanced proficiency students may feel that developing an array of skills needed for success in university coursework is more pressing than general proficiency gains. Thus, it may be important to adapt the curriculum to ensure they can reach these goals while continuing to help them recognise how language development will help them use the language successfully in academic and professional contexts.
Program administrators should find ways to help students and teachers develop realistic expectations about language development. While motivating lower proficiency students with the prospects of positive proficiency gains, those with higher proficiencies may also benefit from a greater understanding of the kinds of language development that is typical at their proficiency level. While recognising that large proficiency gains may not be reasonable for every learner, teachers of advanced students may look for creative ways to engage them to ensure that learning is as meaningful as possible and well suited to meeting their specific needs. At the same time, appropriate steps should be taken to ensure that aggregated data from summative assessments of those finishing their IEP experience be as accurate as possible.
7 Limitations and Future Research
The results of this study need to be interpreted within the context of its limitations. Several limitations are described below along with recommendations for future research.
The first issue is related to the student participants. Though data for this study came from a robust number of language learners, the majority of the students were from a Spanish L1 background. While the variance associated with different L1s as defined by language distance was appropriately accounted for in the statistical analyses, this preponderance of participants from a single L1 may have had an impact on this study’s results and their interpretations. In future study, this line of research would be greatly strengthened if a wider range of L1s were included with more equal distribution across L1s.
Another limitation is associated with our comparisons between intensive and extensive programs. Despite an extensive pool of data from an intensive language program that was analysed in this study, observations and claims associated with extensive language programmes were drawn from other studies. Though this literature was robust and informative leading to valuable insights, our comparisons should be considered tentative since this study was observational rather than experimental. However, these findings provide enough evidence to warrant more careful experimental studies that ensure the comparability of groups and that effectively control for extraneous variables. On a related note, though all student production analysed in this study was from the same curriculum, and their speaking and writing were rated and analysed using the exact same procedures, no attempt was made to examine teacher effect. This should also be considered in future research along with any other context-dependent factors.
Finally, though compelling reasons for a lack of measurable language development in some higher-proficiency learners were presented, uncertainty for the precise reasons for the observed stagnation and regression remain. Of particular importance is the need to accurately differentiate poor language performance due to actual changes in language ability from lower scores arising from missing data or half-hearted assessment performance from distracted or preoccupied students. Additional quantitative and qualitative evidence in subsequent research should be gathered to help clarify this phenomenon.
8 Conclusion
This study examined the progress of 2,325 Novice to Advanced ESL students in the productive skills of speaking and writing over the course of one 15-week semester and determined whether linguistic distance, proficiency, sex, or age affected their progress. Results show that, on average, students made similar progress in speaking and writing. Results also suggest that age may have a slight negative effect on speaking gains, and that linguistic distance may have a slight negative effect on both speaking and writing gains.
Nevertheless, the effect of language proficiency seems much more substantial, with large gains for lower-proficiency learners compared to higher-proficiency learners who slowed or regressed in their proficiency. These findings have important implications for pedagogical expectations, especially that learners with a more advanced proficiency may have different needs and motivations from lower-level learners. We encourage teachers and administrators to use this information to set realistic expectations for the proficiency development of their students. We also recommend exploring the motivation of higher-level students and determine what additional support or resources they may need as they prepare to complete their IEP studies.
Appendices
Appendix A
Appendix B
References
American Council on the Teaching of Foreign Languages. (2012). ACTFL proficiency guidelines 2012. https://www.actfl.org/sites/default/files/pdfs/public/ACTFLProficiencyGuidelines2012_FINAL.pdf
Alpine Testing Solutions. (2020a). Examination of the ACTFL Oral Proficiency Interview - Computer® (OPIc) in Arabic, English, and Spanish for the ACE Review - Part B: Statistical Analysis and Evidence of Validity. Alpine Testing Solutions.
Alpine Testing Solutions. (2020b). Examination of the ACTFL Writing Proficiency Test® (WPT) in English, Russian, and Spanish for the ACE Review - Part B: Statistical Analysis and Evidence of Validity. Alpine Testing Solutions.
Arabski, P. J., & Wojtaszek, A. (Eds.). (2011). Individual learner differences in SLA. https://ebookcentral-proquest-com.proxy.lib.utk.edu
Baker, A. W. (2019). Self-regulation in transition: A case study of three English language learners at an IEP [Unpublished master’s thesis]. Brigham Young University. https://scholarsarchive.byu.edu/etd/7497
Bernhardt, E., Molitoris, J., Romeo, K., Lin, N., & Valderrama, P. (2015). Designing and sustaining a foreign language writing proficiency assessment program at the postsecondary level. Foreign Language Annals, 48(3), 329-349.
Boyle, J. P. (1987). Sex differences in listening vocabulary. Language Learning, 37(2), 273-284.
Burman, D. D., Bitan, T., & Booth, J. R. (2008). Sex differences in neural processing of language among children. Neuropsychologia, 46(5), 1349-1362. https://doi.org/10.1016/j.neuropsychologia.2007.12.021
Burstall, C. (1975). Factors affecting foreign-language learning: A consideration of some recent research findings. Language Teaching, 8(1), 5-25.
Chan, V. (2021). Nature or nurture? A critical analysis of gender differences in second and foreign language learning. Social Sciences and Education Research Review, 8(1), 26-41.
Chiswick, B. R., & Miller, P. W. (2005). Linguistic distance: A quantitative measure of the distance between English and other languages. Journal of Multilingual and Multicultural Development, 26(1), 1-11.
Collins, L., Halter, R. H., Lightbown, R. M., & Spada, N. (1999). Time and the distribution of time in L2 instruction. TESOL Quarterly, 33, 655-680.
Corder, S. P. (1979). Linguistic distance and the magnitude of the language learning task. Studies in Second Language Acquisition, 2(1), 27-36.
Corder, S. P. (1981). Error Analysis and Interlanguage. Oxford University Press.
Cox, J. G., & Sanz, C. (2015). Deconstructing PI for the ages: Explicit instruction vs. practice in young and older adult bilinguals. International Review of Applied Linguistics in Language Teaching, 53(2), 225-248.
Danesh, J., & Shahnazari, M. (2020). A structural relationship model for resilience, L2 learning motivation, and L2 proficiency at different proficiency levels. Learning and Motivation, 72, 101636. https://lao.ca.gov/2004/english_learners/021204_english_learners.htm
DeKeyser, R. (2020). Skill acquisition theory. In B. VanPatten, J. Williams, G. D. Keating, & S. Wulff (Eds.), Theories in second language acquisition (pp. 83-104). Routledge.
DeKeyser, R. (2018). The critical period hypothesis: A diamond in the rough. Bilingualism: Language and Cognition, 21(5), 915-916. https://doi.org/10.1017/S1366728918000147
Dionne, G., Dale, P. S., Boivin, M., & Plomin, R. (2003). Genetic evidence for bidirectional effects of early lexical and grammatical development. Child Development, 74, 394-412.
Eckes, T. (2011). Introduction to many-facet Rasch measurement: Analyzing and evaluating rater-mediated assessments. Lang.
Ehrlich, S. (1997). Gender as social practice: Implications for second language acquisition. Studies in Second Language Acquisition, 19, 421-446.
Elder, C., & Davies, A. (2010). Performance on ESL Examinations: Is There a Linguistic Distance Effect? Language and Education, 12(1), 1-17. https://doi.org/10.1080/09500789808666736
Fattahi, N., & Nushi, M. (2021). The effect of gender and language proficiency on the metaphor use in the writing of TEFL students. Asian-Pacific Journal of Second and Foreign Language Education, 6(1), 19.
Foreign Service Institute. (n.d.). Foreign Language Training. U.S. Department of State. https://www.state.gov/foreign-language-training/
Georgiou, G. (2021). Interplay between perceived cross-linguistic similarity and L2 production: Analyzing the L2 vowel patterns of bilinguals. Journal of Second Language Studies, 4, 48-64.
Główka, D. (2014). The impact of gender on attainment in learning English as a foreign language. Studies in Second Language Learning and Teaching, 4(4), 617-635.
Goertler, S., Kraemer, A., & Schenker, T. (2016). Setting evidence-based language goals. Foreign Language Annals, 49, 434-454.
Green, J. M., & Oxford, R. (1995). A closer look at learning strategies, L2 proficiency, and gender. TESOL Quarterly, 29(2), 261-297.
Hart-Gonzalez, L., & Lindemann, S. (1993). Expected achievement in speaking proficiency, 1993. School of Language Studies, Foreign Services Institute. Department of State.
Hubert, M. D. (2013). The development of speaking and writing proficiencies in the Spanish language classroom: A case study. Foreign Language Annals, 46, 88-95.
Isbell, D. R., Winke, P., & Gass, S. M. (2019). Using the ACTFL OPIc to assess proficiency and monitor progress in a tertiary foreign languages program. Language Testing, 36, 439-465.
Jackson, F. H., & Kaplan, M. A. (1999). Lessons learned from fifty years of practice in government language teaching. In J. E. Alatis & A. Tan (Eds.), Georgetown University Round Table on Languages and Linguistics (pp. 71-87). Georgetown University Press.
Jaekel, N., Ritter, M., & Jaekel, J. (2023). Associations of students’ linguistic distance to the language of instruction and classroom composition with English reading and listening skills. Studies in Second Language Acquisition, 45(5), 1287-1309.
Joyce, P. (2019). The relationship between L2 listening proficiency and L2 aural language processing. PASAA, 57(1), 9-32.
Juffs, A. (2020). Aspects of Language Development in an Intensive English Program. New York: Routledge.
Knoch, U., & Chapelle, C. A. (2017). Validation of rating processes within an argument-based framework. Language Testing. https://doi.org/10.1177/0265532217710049
Levenshtein, V. I. (1965). Binary codes for the correction of deletions, insertions and substitutions of symbols. Doklady Akad Nauk SSSR, 163, 845-848.
Linacre, J. M. (2018). FACETS computer program for many-facet Rasch measurement (Version 3.81.0). Winsteps.com
Liu, X., & Yin, J. (2021). A review on the effects of instructional time and teacher quality on language learning performance. Open Access Library Journal, 8, e7834. https://doi.org/10.4236/oalib.1107834
Maghsudi, M., Sharifi, E., & Abedi, S. (2015). The effect of gender on foreign language learning. International Journal of Educational Investigations, 2(2), 162-166.
McNamara, T. F. (1996). Measuring second language performance. Longman.
Open Doors. (2024). Open Doors 2023 Fast Facts. Retrieved from https://opendoorsdata.org/fast_facts/fast-facts-2024
Porte, G., & McManus, K. (2018). Doing replication research in applied linguistics. Routledge.
Serrano, R. (2011). The time factor in EFL classroom practice. Language Learning, 61, 117-145.
Serrano, R. (2022). A state-of-the-art review of distribution-of-practice effects on L2 learning. Studies in Second Language Learning and Teaching, 12, 355-379.
Sierens, S., Slembrouck, S., Van Gorp, K., Agirdag, O., & Van Avermaet, P. (2019). Linguistic interdependence of receptive vocabulary skills in emergent bilingual preschool children: Exploring a factor-dependent approach. Applied Psycholinguistics, 40(5), 1269-1297. https://doi.org/10.1017/S0142716419000250
Thompson, I. (1996). Assessing foreign language skills: Data from Russian. The Modern Language Journal, 80, 47-65.
Van der Ploeg, M. (2023). Language learning never gets old: Learning a new language in later life [Thesis fully internal (DIV), University of Groningen]. University of Groningen. https://doi.org/10.33612/diss.803512740
Van der Slik, F. W. (2010). Acquisition of Dutch as a second language: The explanative power of cognate and genetic linguistic distance measures for 11 West European first languages. Studies in Second Language Acquisition, 32(3), 401-432.
Van der Slik, F. W., Van Hout, R. W., & Schepens, J. J. (2015). The gender gap in second language acquisition: Gender differences in the acquisition of Dutch among immigrants from 88 countries with 49 mother tongues, PloS One, 10/11: e0142056.
Viera, C., & Arispe, K. (2021). Can Spanish programs evaluate oral proficiency gains for all learners? An empirical analysis of traditional and nontraditional Spanish majors’ oral proficiency development. Foreign Language Annals, 55, 832-852.
Wardani, A. D., Guawan, I., Kusumaningrum, D. E., Benty, D. N., Sumarsono, R. B., Nurabadi, A., & Handayani, L. (2020). Student learning motivation: A conceptual paper. ECPE 2020. https://dx.doi.org/10.2991/assehr.k.201112.049
Weissberg, R. (2006). Connecting speaking and writing in second language writing instruction. University of Michigan Press.
Weng, F., & Liu, X. (2024). Exploring second language students’ language assessment literacy: Impact on test anxiety and motivation. Frontiers in Psychology, 15, 1289126.
Wichmann, S., Holman, E. W., & Brown, C. H. (Eds.). (2022). The ASJP Database (version 2.0) [Data set]. https://asjp.clld.org/
Winke, P., Zhang, X., Rubio, F., Gass, S., Soneson, D., & Hacking, J. (2020). The proficiency profile of language students: Implications for programs. Second Language Research and Practice, 1, 25-64.
Xinxin, C., Xiaolan, L., & Ahmed, M. (2022). Relationship of the linguistic distance to English ability of a country. arXiv preprint arXiv:2211.07855.
Xue, J., Hu, X., Yan, R., Wang, H., Chen, X., & Li, M. (2021). Onset age of language acquisition effects in a foreign language context: Evidence from Chinese–English bilingual children. Journal of Psycholinguistic Research, 50(2), 239-260. https://doi.org/10.1007/s10936-019-09637-y
Zughoul, M. R., & Taminian, L. (1984). The linguistic attitudes of Arab university students: Factorial structure and intervening variables. International Journal of the Sociology of Language, 50, 155-179.
Authors:
K. James Hartshorn (corresponding author)
Brigham Young University
4061 JFSB, Provo, UT 84602 USA
James_hartshron@byu.edu
+1 (801) 422-4034
https://orcid.org/0000-0002-0629-7410
Wendy Baker-Smemoe
Brigham Young University
4057 JFSB, Provo, UT 84602 USA
wendy_baker@byu.edu
https://orcid.org/0000-0001-9740-6303
Matthew Millar
Brigham Young University
4064 JFSB, Provo, UT 84602 USA
Benjamin McMurry
Brigham Young University
4064 JFSB, Provo, UT 84602 USA
ben_mcmurry@byu.edu
https://orcid.org/0000-0002-5628-3295
______________________________________________
1This represents two learners who were bilingual in multiple Slavic languages.
2 The IEP reports gathering rating data across multiple semesters with the following MFRM results. Speaking infit means square range: .90 to .99, reliability range: .91 to .97. Writing infit means square range: .81 to .90, reliability range: .74 to .94.