A Study on English Language Teachers’ Lexical Attrition in Chinese Contexts

Introduction

Foreign language (FL) proficiency and pedagogical capabilities are vital attributes for second language/foreign language (L2/FL) teachers (Choi & Lee, 2016), particularly in the context of English as a foreign language (EFL) education. Compared to teachers in other subjects, the uniqueness of FL teachers’ professional competence primarily lies in their proficiency in the target language as bilinguals (Chen & Goh, 2011), with their target language proficiency serving as the foundation for effective teaching (Braine, 2010). Stakeholders such as teachers, students, parents, project managers, and policymakers universally consider a teacher’s proficiency in the target language as a fundamental quality for excellence in teaching (Braine, 2010). Insufficient language proficiency is a major obstacle faced by non-native language teachers, impacting employment, teaching, and professional development (Braine, 2010; Chen & Goh, 2011). Without regular interaction with native speakers, engage in language communication and learning, or actively undergo language training, teachers may rely heavily on textbooks, leading to declining language proficiency (Choi & Lee, 2016). This challenge is exacerbated in regions like China with limited exposure to authentic English environments (Wenfeng & Gao, 2008). Given these factors, it is essential to investigate whether EFL teachers experience language attrition after entering the workforce and identify the underlying factors contributing to this phenomenon.

For a long time, a widely held assumption has been that a language will fade from memory once it ceases to be used, practiced, or studied, regardless of whether it is a second language (L2), FL, or the first language (L1) (Mickan et al., 2020; Schmid, 2004, 2022; Schmid & Mehotcheva, 2012). The fundamental causes of this phenomenon and the seemingly unstoppable nature of the language attrition process have been enduring questions in the field of language science (Mickan et al., 2021). Language attrition, defined as “the loss of language skills and knowledge by those who have studied and then discontinued the use of the language over time” (Freed, 1982, p. 1), has gradually emerged as a topic of interest in the fields of applied linguistics and psycholinguistics (Kriz & Chromy, 2020). Language attrition is not equivalent to language forgetting in the conventional sense; language attrition primarily emphasizes that language knowledge may not necessarily be entirely lost from memory, nor is it irreversibly concealed or obliterated by newly acquired knowledge. Instead, less frequently used or unused knowledge becomes less interconnected, rendering it temporarily less accessible to language users (de Bot et al., 2004). Previous research on language attrition has primarily focused on L1 attrition (Schmid, 2022; Schmid & Dusseldorp, 2010; see also Schmid & Köpke, 2019). Investigations into L2/FL attrition have been relatively limited (Mehotcheva & Köpke, 2019), and the field of L2/FL attrition remains less extensive and theoretically developed compared to L1 attrition (Schmid, 2022). More recently, there has been a significant body of research dedicated to the phenomenon of FL attrition (Jessner et al., 2021; Mickan et al., 2020, 2023), with a particular focus on FL acquired or learned exclusively within formal classroom settings, devoid of any substantive immersive learning experiences (Schmid, 2022). These languages are typically acquired or learned through intentional efforts but do not typically constitute a substantial part of daily life outside of the educational context (Mehotcheva & Köpke, 2019). A key aspect of FL attrition research is to identify the factors leading to the decline of a FL (Kim & Kim, 2022). Abbasian and Khajavi (2011) contend that the duration of unusing FL is a critical factor in FL attrition, with longer interruptions of contact with L2/FL leading to more severe attrition. Mickan et al. (2020), drawing from a modified retrieval-induced forgetting paradigm in the field of psychology, have substantiated that competition from more used L1 drives FL attrition. While the majority of research on FL attrition has primarily focused on language learners, such as university students, there has been limited attention given to FL attrition among FL teachers. Although FL teachers are commonly thought to possess a strong resistance to FL attrition (Alharthi, 2015), primarily due to their roles as consistent FL learners (Choi & Lee, 2016; Richards et al., 2013), existing literature and preliminary surveys suggest that they experience ongoing and significant FL lexical attrition (Alharthi, 2015). Unfortunately, none of the studies that have explored lexical attrition have addressed the overlooked aspect of why EFL teachers appear to forget vocabulary. Horwitz (1996) argued that despite assuming FL teachers possess a high level of FL proficiency, they are still considered language learners. Previous research has suggested that FL teachers must acquire the skill of establishing a teacher-student learning community and continuously enhance their own FL proficiency through lifelong learning (Choi & Lee, 2016). Additionally, it is important to recognize that language learning or mastery is an ongoing and never-ending process (Richards et al., 2013), and many non-native speakers must actively maintain their FL proficiency to prevent attrition (Schmid, 2023). This is particularly crucial for FL teachers, especially in the context of EFL education in China, where EFL teachers remain the primary facilitators of students’ FL learning, especially in the context of basic education where exam preparation is a primary focus (Wenfeng & Gao, 2008). If FL teachers experience attrition in their language proficiency, it can impact the quality of their FL instruction and, consequently, result in adverse effects on students’ FL learning experiences (Alharthi, 2015). FL teachers, being the primary role models for FL learners, need to maintain and continually enhance their FL proficiency (Richards et al., 2013). Thereby, identifying and addressing FL attrition among teachers can greatly assist FL educators in better planning their career development. In the long term, given the emphasis on English for its potential to promote sustainable economic growth and national progress (Xie et al., 2023), identifying the factors influencing FL attrition can offer a fresh perspective and knowledge for the formulation of EFL teaching policies (Schmid, 2022). This includes advocating for the establishment of standards for EFL teachers’ language proficiency, regulating pre-service training programs for EFL teachers, prompting educational authorities to prioritize the language proficiency of junior high school EFL teachers, conducting targeted training, and implementing monitoring and assessment measures. This holds potential significance for enhancing the sustainable development of EFL education. However, there is currently limited consideration of FL attrition and maintenance as issues in policies or teaching methods (Schmid, 2022), particularly regarding the maintenance of teachers’ language proficiency (Choi & Lee, 2016).

Given the notably higher attrition rate in the vocabulary domain relative to other aspects of language knowledge, coupled with the inherently open-ended nature of vocabulary itself and the intricate interplay among FL teachers’ vocabulary proficiency, overall language proficiency, and professional competencies (Ni & Jin, 2020b), especially within the context of language proficiency differentiation, vocabulary proficiency emerges as a core component (Alqahtani, 2015). This study will primarily focus on exploring the factors influencing post-service lexical attrition among junior high school EFL teachers in China and identifying effective approaches to mitigate lexical attrition. Specifically, this research aims to determine the current status of lexical attrition among in-service EFL teachers after ceasing formal instruction, the potential influencing factors contributing to lexical attrition, and how to enhance the lexical attrition status of EFL teachers.

Literature Review

Causes of L2/FL Lexical Attrition

Researchers have extensively investigated the reasons behind FL attrition and conducted numerous studies to understand and address the identified problems associated with it (Schmid, 2022). It has been suggested that various factors contribute to the attrition of lexical items and pose challenges to vocabulary acquisition (Ni & Jin, 2020a). These factors can be broadly categorized as intra-lexical traits and extra-lexical traits (Alharthi, 2015; Jin & Ni, 2011). Intra-lexical features are related to the form, frequency, class, and meaning of words. The intra-lexical features are related to the word’s form, frequency, class, and meaning. That is, the attrition of vocabulary may be easy due to various factors, including grammatical class, similarity to L1, word length, part of speech, and concreteness, cognates, abstractness or image ability of meaning (Milton, 2009). In addition to the effects of intra-lexical factors, several extra-lexical factors can further accelerate the extent of lexical attrition among L2 learners. These factors include the salience of frequency in input, the erosion of time, language contact, initial proficiency, length of exposure to the language, age, attitude, and motivation (Ni, 2009). Furthermore, as highlighted by Schmid (2022), FL attrition is a developmental experience that spans the lifetime of FL learners, occurring within a single generation rather than across different generations. Therefore, changes in language environment, occupation, family situation, cultural context, and decisions to learn other languages can contribute to language attrition (Schmid, 2022). Additionally, between-language competition can also lead to attrition (Mickan et al., 2020). Köpke and Genevska-Hanke (2018) also further discuss the sensitivity of language systems to language context based on a 30-year review of attrition research. Previous studies indicate that vocabulary attrition is a common experience for language learners, which can negatively impact language proficiency (Montrul, 2009). While it may intuitively appear that frequency of use, initial language proficiency, and emotional factors such as motivation and attitude play significant roles in language attrition and retention, identifying the specific factors (e.g., length of time, amount of contact, attitudes) that facilitate or impede attrition and to what degree, as well as how to maintain or regain language proficiency, is challenging due to individual differences among FL learners (Mehotcheva & Mytara, 2019; Schmid, 2022). Consequently, the field lacks empirical evidence and fails to consider the interactions between factors. It is worth noting that there is a scarcity of research specifically addressing the impact of attrition in FL teachers’ FL proficiency on their professional development (Alharthi, 2015), potentially leading to significant hurdles in their career progression (Choi & Lee, 2016). Since FL teachers are also considered FL learners (Horwitz, 1996) and serve as exemplary committed advanced learners (Valmori & De Costa, 2016), investigating the causes of lexical attrition among EFL teachers systematically may contribute to understanding the impact of EFL teachers’ lexical attrition on language teaching and learning. Therefore, this study aimed to explore issues related to attrition in word knowledge among EFL teachers. Previous studies indicated that L2/FL attrition has two distinct characteristics: it is initially triggered by the disuse of the L2 and occurs within the lifespan of an L2/FL user (Ni & Jin, 2020a). Thus, this study exclusively focuses on the extra-lexical factors of L2 lexical attrition. The study specifically aims to investigate the following overarching research questions: (1) To examine the current state of lexical attrition experienced by in-service EFL teachers in China. (2) To identify the factors contributing to the lexical attrition of instructed FL among in-service EFL teachers.

Materials and Methods

Participants

Participants were 105 full-time EFL teachers (77 females/28 males) from 43 different junior high schools in China. Convenience sampling was employed, and a printed questionnaire was administered to EFL teachers who were participating in a national in-service training program for young in-service EFL teachers in junior high school. Out of 135 respondents, 105 provided valid responses after deleting those questionnaires with the same answer for all items. The participants are all native Chinese speakers with experience in teaching English at the junior high school level. All participants were between the ages of 23 and 29 (Mean = 26.41; SD = 1.48) and held a bachelor’s degree in English. Based on the incubation periods identified in previous attrition onset studies (Mehotcheva, 2010; Ni, 2009), the process of L2 lexical attrition can be categorically divided into three stages based on the duration of time that participants stop using certain learned FL words. These stages are the early stage (within 2 years after disuse), the middle stage (from the second year to the sixth year), and the final stage (6 years later). This process exhibits a rapid-stable-rapid pattern, where vocabulary experiences a notably swift attrition in the early stage, remains relatively stable in the middle stage, and then undergoes accelerated attrition in the final stage. Hence, we accounted for the incubation period of lexical attrition among EFL teachers. The average time since these teachers graduated from college is 3.24 years (ranging from 2 to 6 years) (SD = 1.53). Since graduation, their self-reported frequency of using college-level English averages 0.34 (SD = 0.48), signifying minimal or rare use of college-learned English (Ni & Jin, 2020b). On average, the participants had been teaching EFL for 3.15 years (ranging from 1 to 6 years) (SD = 1.15). Furthermore, all teachers were exposed to these two textbooks, namely Integrated English Skills 2 and Integrated English Skills 3, during their university education. These textbooks were published by the High Education Press in China. All participants were required to take the Test for English Majors-Band 4 (TEM-4) during their university years in China. TEM-4 is a high-stakes English proficiency test for English-major undergraduates in China (Fan et al., 2020), established in 1991 by the National Advisory Commission on Foreign Language Teaching in Higher Education (NACFLT, 2004). It is designed to assess the comprehensive English abilities of second-year students, with limited testing opportunities and highly authoritative results. As it effectively distinguishes between students of varying proficiency levels and reflects their true abilities (C. Li, 2012), TEM-4 is considered crucial for graduation in some universities and job-seeking in China (Fan et al., 2020). The exam includes a dedicated section for assessing vocabulary skills, and students who pass the exam are expected to have a certain level of vocabulary knowledge. Therefore, the TEM-4 serves as an effective standard for evaluating vocabulary proficiency (Table 1). Regarding score reporting, individual test-takers scoring 60 or above receive a certificate from NACFLT, indicating their level of performance as “excellent” (score 80 or above), “good” (score between 70 and 79), or “pass” (score between 60 and 69) (NACFLT, 2004). Based on the teachers’ performance on the TEM-4 (Mean = 67.58, SD = 6.16), the overall TEM-4 scores for the teachers did not reach the “Good” level, despite 69 (65.71%) teachers successfully passing TEM-4. Additionally, 55 (52.38%) teachers do not consider themselves advanced English learners. Ten teachers (referred to as A, B, C, D, E, F, G, H, I, and J to ensure anonymity) were purposefully selected, representing a variety of characteristics including educational background, academic titles, gender, and age (ranging from 24 to 28). They each had 3 to 5 years of experience in EFL instruction at different junior high schools. To maintain participant privacy and avoid potential response bias, we provided a brief introduction to the interview topic. This approach was chosen to encourage authentic and spontaneous responses that genuinely reflect the participants’ perceptions (Creswell, 2014). Prior to administering the questionnaire and conducting teacher interviews, participants were informed about the study’s purpose and assured that the data collected would be used solely for academic purposes. They were also assured that their identities would remain confidential and their data anonymized. Participants willingly completed the questionnaire and participated in the interviews.

OPEN IN VIEWER

Table 1.

TEM-4 Performance of Subjects.

	Score range	Frequency	Percent
Valid	40–49	8	7.62
	50–59	28	26.67
	60–69	67	63.81
	70 or above	2	1.90
	Total	105	100.00

Instruments

The Word Self-Report Scale

We conducted an assessment of lexical attrition among EFL teachers using a self-report scale. Initially, we sampled words from the vocabulary lists of the two textbooks Integrated Skills of English 2 and Integrated Skills of English 3 to ensure that participants were familiar with these words. These textbooks contained a total of 502 words. In the context of vocabulary selection, a geometric sequence is used to choose words from the textbooks’ vocabulary list to ensure that the selected words are distributed in a way that is somewhat uniform or representative. Then, 97 words were selected using a geometric sequence, which included some proper nouns. Proper nouns were excluded, leaving us with 45 words, totaling 18 nouns, 12 verbs, 9 adjectives, 4 adverbs, and 2 conjunctions. These 45 selected test words also appeared in the Common Vocabulary List for English Majors in Higher Education based on the Teaching Syllabus for English Majors in Higher Education (2000 Edition). All surveyed teachers indicated that they had studied these 45 words during their university years. To develop the word self-report scale, detailed information from Wesche and Paribakht’s (1996) study was consulted, with a preference for meaning-translation. Three optional items were employed to measure lexical attrition (Table 2). Participants were instructed to choose option A if they believed they had not learned the word, option B if they believed they had learned the word but could not recall its meaning, and option C if they believed they had learned the word and could either translate it into Chinese or provide a synonym in English.

OPEN IN VIEWER

Table 2.

The Word Self-Report Scale.

Please answer the following questions according to the following requirements.

1. If you select option A, please indicate A next to the word. 2. If you select option B, please indicate B next to the word. 3. If you select option C, please indicate C next to the word and translate it into Chinese or provide its synonyms in English. Option A: I have not previously learned this word. Option B: I have learned this word, but I cannot recall its meaning. Option C: I have learned this word, and I believe it means (synonym or translated word).

According to Wesche and Paribakht (1996), the measurement guidelines can be summarized as follows:

(1) Choosing option A, “Retrieval Failure,” will not award any points to the participants.

(2) Selecting option B, “Difficult Retrieval,” will earn 0.5 points.

(3) Opting for option C and providing a correct explanation, “Right Retrieval,” will yield two points.

(4) Selecting option C but offering an incorrect interpretation, “Wrong Retrieval,” will grant one point.

The total score was 90, calculated as 45 multiplied by 2. All the responses were categorized into four groups:

A: Retrieval failure,

B: Difficult retrieval,

CW: Wrong retrieval,

CR: Right retrieval.

Questionnaire

With the assistance of two experts who have extensive experience in researching language attrition, we developed a questionnaire in Chinese to investigate the lexical attrition of EFL teachers. The questionnaire consisted of two parts: the first part gathered demographic information such as age, gender, and teaching experience, while the second part comprised 34 descriptive items (which were later revised to 21 items) examining the causes of lexical attrition among EFL teachers. To determine the items related to EFL teachers’ lexical attrition in the questionnaire, we considered the input provided by teachers regarding their experiences with language attrition. In accordance with Ma et al. (2023), item statements within a questionnaire are achieved through diverse methodologies encompassing the integration of qualitative data as well as existing scales. The process of item derivation for the present questionnaire followed a sequential procedure, whereby qualitative data collection served as the initial phase. Thus, before deciding on the questionnaire items, interviews were conducted with teachers first, focusing on the factors influencing their lexical attrition. Most of the EFL teachers interviewed explicitly or implicitly indicated that their vocabulary has suffered attrition due to years of not using the words they had learned in college. We summarized and categorized their responses based on the keywords they mentioned, which allowed us to identify five main themes related to the causes of EFL teachers’ lexical attrition (see Table 3). These themes include language contact, initial language proficiency, teachers’ concerns about maintaining language proficiency, teachers’ attitudes toward language learning, and artificially-intervened language use.

OPEN IN VIEWER

Table 3.

Lexical Attrition Profiles of the Interviewees.

Factors of lexical attrition	No.	Interviewees
Limited exposure to previously learned English words.	8	Teachers A, B, D, E, F, G, H, J
Predominantly using Chinese during teaching.	7	Teachers A, C, D, F, G, I, J
Decreased English proficiency.	5	Teachers B, D, E, H, I
The more words one knows, the less attrition occurs.	6	Teachers A, C, D, E, G, H
Lack of motivation to maintain EFL proficiency.	4	Teachers B, F, G, I
Insufficient awareness of the need to improve EFL proficiency.	6	Teachers B, C, D, E, H, I
Lack of interest or passion in English learning.	4	Teachers C, D, F, J
Limited time to expand English knowledge.	5	Teachers A, E, F, H, I
Lack of English proficiency tests for teachers.	7	Teachers B, C, D, E, H, I, J
Inadequate training to enhance English knowledge.	7	Teachers A, B, D, F, G, H, I

Then, we conducted a thorough literature review to establish a theoretical framework for the sources related to EFL teachers’ lexical attrition, including the concept of lexical attrition (Freed, 1982), the features of lexical attrition (Ni & Jin, 2020a), and the causes of language attrition (Alharthi, 2015; Mehotcheva, 2010; Ni, 2009; Schmid, 2022). Based on this literature review and data from teachers’ interview, we developed a set of 34 items to assess EFL teachers’ lexical attrition, which were organized into three parts. The first part measured the frequency of English use, including teachers’ natural exposure and artificially-intervened exposure to English after graduation. Participants self-assessed their exposure using a five-point scale (never, rarely, occasionally, and often, and always). The second part focused on self-assessment of English proficiency and consisted of two sections. The first section assessed participants’ initial English proficiency using items adapted from the can-do statements (Gardner et al., 1987). To ensure the reliability of the can-do statements, a correlation analysis was conducted between self-assessed initial English proficiency and scores on the TEM-4 (r = .64, p < .01), indicating that the can-do statements could reflect teachers’ initial English proficiency. The second section assessed participants’ emotional attitudes towards EFL learning. The third part gathered information about participants’ efforts and concerns regarding the maintenance and development of their English proficiency. Participants were required to choose the most appropriate option based on their experiences and perceptions. To avoid any bias or misunderstanding, the questionnaire was presented in Chinese. Participants used a 5-point Likert scale (ranging from 1 for “strongly disagree” to 5 for “strongly agree”) to indicate their agreement with each statement. This process resulted in a comprehensive questionnaire comprising 34 items. The questionnaire was administered to four experts in order to assess its relevance as a measure of surface validity. Consensus was reached among all participants to incorporate all items and sections, with the additional recommendation to undertake a pilot study to test the reliability of the initial 34-item questionnaire. The pilot study indicated that the questionnaire’s Cronbach’s alpha was .788, above the acceptable level of .70, demonstrating satisfactory measurement of factors related to teachers’ lexical attrition. Finally, the 21-item questionnaires and word self-report scales were administered in face-to-face sessions.

Interview

Initially, 10 teachers were interviewed to identify factors influencing lexical attrition among teachers, guiding the development of the questionnaire. These interviews, each lasting about 30 min, were conducted by the first author in Chinese Mandarin to ensure comprehensive expression. Tencent Meeting and WeChat facilitated the interviews. Transcripts were independently coded by two English major graduate students, discrepancies were resolved using an analytical framework, and interviewees reviewed findings for clarity and reliability. The interviews served two purposes: informing questionnaire design and aiding in the interpretation of data obtained from the questionnaire and word self-report scale. The interview questions were as follows: (1) Have you experienced any attrition in your English vocabulary after completing formal learning? If yes, what do you believe are the causes of your lexical attrition? (2) Do you believe that EFL teachers’ language attrition can negatively impact foreign language teaching and learning? (3) How do you think EFL teachers should avoid lexical attrition to maintain and improve their English language skills?

Results

Results of the Words Self-Report Scale on Lexical Attrition

The descriptive statistics for the word self-report scales are summarized in Table 4. The overall mean score, highest score, and lowest score on these scales were 63.12 (SD = 9.53), 83.80, and 47.10, respectively. These scores exhibited significant variation, indicating a diverse dispersion and distribution pattern. The word self-report scale’s validity and reliability were found to be acceptable. The results show that teachers scored low to moderate on the word self-report scale (Mean = 63.12, SD = 9.53), implying a significant attrition of English vocabulary among junior high school EFL teachers after their career commencement. For option A, I have not previously learned this word, the mean score was very low (Mean = 2, SD = 2.43), indicating that only a small number of the 45 words had been completely forgotten by the participants. This suggests that most of these words were still within the teachers’ basic recollection. The mean score for option CR was 22 (SD = 4.58), indicating that participants could not correctly retrieve a substantial number of words, showcasing a significant loss in the words they had learned during their college years. Additionally, the mean scores for options B and CW were 13 (SD = 5.40) and 8 (SD = 3.44), respectively, indicating that teachers remembered having encountered these words before but struggled to extract their meanings. This suggests that EFL teachers had experienced significant attrition in the English words they once learned. Notably, the mean score for option A (Mean = 2) was significantly lower than CR (Mean = 20), CW (Mean = 6), and B (Mean = 11), indicating that the majority of the words were still retained in the teachers’ memory. In summary, the above findings collectively point to a substantial attrition in English vocabulary among junior high school EFL teachers after they begin their careers. However, it is noteworthy that these eroded words have not been completely forgotten but rather persist in the teachers’ memories.

OPEN IN VIEWER

Table 4.

Descriptive Statistics of the Word Self-Report Scales (n = 105).

Retrieval Category	No. of words	Minimum	Maximum	Mean	SD
A (Retrieval failure)	45	0	9	2	2.43
B (Difficult retrieval)	45	0	24	13	5.40
CW (Wrong retrieval)	45	1	16	8	3.44
CR (Right retrieval)	45	14	32	22	4.58
Total score of the word scales		47.10	83.80	63.12	9.53

Results of the Questionnaire on Lexical Attrition Factors

Exploratory Factors Analysis

To extract factors from the 34-item questionnaire and identify the underlying factors that explain the covariation among the measured variables, we conducted an exploratory factor analysis (EFA) using principal component analysis with Varimax rotation. This method was chosen primarily to maximize the variance on each factor, thereby simplifying the factor loadings structure and making the results easier to interpret (Thompson, 2004). The objective was to determine whether the five factors examined in this study adequately represented the factors influencing EFL teachers’ lexical attrition. The iteration rotation process was performed nine times until convergence was achieved. The results revealed a Kaiser–Meyer–Olkin (KMO) value of 0.81 (>0.5), indicating sample size and data collection met the requirements for conducting factor analysis (R. H. Pearson & Mundform, 2010). Bartlett’s test of sphericity was significant (Approx. χ² = 1,427.97, df = 300, p < .01), further confirming the suitability of the data for factor analysis. Next, principal component analysis with Varimax orthogonal rotation was conducted to extract five latent factors with eigenvalues greater than 1.00. These factors explained a cumulative variance contribution rate of 65.19% (Table 5), providing substantial coverage of the factors under investigation. In the social sciences, a cumulative variance contribution rate exceeding 60%, with the first two components accounting for around 40%, is generally considered satisfactory for EFA (R. H. Pearson & Mundform, 2010). Eight items (Items 9, 13, 16, 17, 22, 29, 30, and 33) exhibited factor loadings below 0.40 and were thus deemed inadequate. Additionally, five items (2, 3, 6, 8, and 14) did not demonstrate sufficient internal consistency and were excluded from further analysis. Consequently, the revised questionnaire comprised 21 items. The factor loadings, eigenvalues, and variance explanatory rates of these items are presented in Table 5. The items are arranged in descending order based on their factor loadings. Notably, the absolute values of the factor loadings for each item exceeded .40. The factor loadings of the 21 items ranged from .52 to .90, indicating that the questionnaire generally reflected the content related to the five factors: teachers’ language contact, teachers’ initial language proficiency, teachers’ concerns about improving language proficiency, teachers’ attitude toward language learning, and teachers’ artificially-intervened language use.

OPEN IN VIEWER

Table 5.

Exploratory Factor Analysis Results.

Items	Factor loading	Eigenvalues	Variance percentage
Language contact		7.70	32.08
The frequency of practicing English listening	.90
The frequency of practicing English writing	.83
The frequency of practicing English speaking	.81
The frequency of composing emails in English	.76
The frequency of engaging in English conversation	.74
The frequency of acquiring new English vocabulary	.73
The frequency of watching English movies	.65
The frequency of reading English articles	.62
Initial Language proficiency		3.45	14.38
Self-assessment of English reading skills	.83
Self-assessment of English writing skills	.82
Self-assessment of English listening skills	.71
Self-assessment of English speaking skills	.63
Teachers’ concerns about language proficiency		1.90	7.91
The importance of enhancing English proficiency	.84
The importance of reviewing English knowledge	.82
The importance of expanding English knowledge	.52
Teachers’ attitude toward language learning		1.59	6.64
Confidence in English learning	.82
Interest in English learning	.77
Positive attitude toward English learning	.64
Artificially-intervened language use		1.00	4.18
The frequency of English training sessions for teachers	.78
The frequency of English proficiency tests for teachers	.67
The frequency of English lectures/seminars for teachers	.61
Extraction method: principal component analysis.Rotation method: Varimax with Kaiser normalization.

Reliability Analysis

To assess the internal consistency of the revised 21-item questionnaire, the reliability analysis was conducted using Cronbach’s alpha. As presented in Table 6, all Cronbach’s alpha values fell within the range of .730 to .916, surpassing the threshold value of .7 commonly accepted as indicating satisfactory internal consistency (Shrestha, 2021). The overall Cronbach’s alpha value for the entire questionnaire was .883. These findings indicate that the questionnaire demonstrates reasonably adequate reliability in measuring the latent variables, suggesting that the 21-item scale adequately captures the five principal factors of Chinese EFL teachers’ lexical attrition.

OPEN IN VIEWER

Table 6.

Reliability Statistics by Cronbach’s Alpha.

Main factors	Cronbach’s alpha	No. of items
Language contact	.916	8
Initial language proficiency	.825	4
Teachers’ concerns about language proficiency	.730	3
Teachers’ attitude toward language learning	.766	3
Artificially-intervened language use	.816	3
Total	.883	21

Descriptive Statistics and Correlations

The questionnaire employed a 5-point Likert scale ranging from “1 (Strongly disagree)” to “5 (Strongly agree),” with scores between 1 and 2 indicating low level of measured variables, 2 and 3 suggesting moderate level of measured variables, 3 and 4 representing moderate to high level of measured variables, and scores above 4 signifying high level of measured variables, following the criteria established by C. C. Li (2018). The score range, maximum, minimum, mean values, standard deviations, and correlation coefficients can be noted in Table 7. It is shown that teachers had moderate scores on their perception of concerns about language proficiency (M = 3.31, SD = 0.708) and their attitude toward language learning (M = 3.24, SD = 0.872), but their perception of language contact (M = 3.01, SD = 0.788) and artificially-intervened language use (M = 3.05, SD = 785) were relatively lower. Their perception of initial language proficiency was moderately high (Mean = 3.14, SD = 0.749), slightly above the threshold of 3. There was a significantly positive correlation between these five IVs (Independent Variables) and scores on the word self-report scales (p < .05), which indicated that the IVs were responsible for increasing the degree of teachers’ EFL lexical attrition.

OPEN IN VIEWER

Table 7.

Descriptive and Correlation Coefficients of Variables (n = 105).

		Descriptive					Correlation coefficients
Variables		Score range	Min	Max	M	SD	LC	TAEL	TCLP	AILU	ILP
DV	SWSS	0–90	47.10	83.80	63.12	9.53	.51**	.41**	.28**	.31**	.55**
IVs	LC	1–5	1.13	5	3.01	0.788	—	.48**	.16	.60**	.41**
	TAEL	1–5	1.32	5	3.24	0.872		—	.16*	.40**	.46**
	TCLP	1–5	1	5	3.31	0.708			—	-.12	.26*
	AILU	1–5	1	5	3.05	0.785				—	.28*
	ILP	1–5	1	5	3.14	0.749					—

Note. SWSS = scores on words self-report scale; LC = language contact; TAEL = teachers’ attitude toward English learning; TCLP = teachers’ concerns about language proficiency; AILU = artificially-intervened language use; ILP = initial language proficiency; DV = dependent variable

*

p < .05. **p < .01.

Multiple Regression Analysis

To perform the multiple linear regression analysis, we first calculated the factor scores. These scores were computed using the regression method and then standardized to ensure that they had a mean of 0 and a standard deviation of 1. This standardization facilitates comparison and interpretation of the factor scores alongside other data. The results showed that all factor scores had a mean of 0 and a standard deviation of 1, indicating successful standardization. In the subsequent multiple regression analysis, we used these standardized factor scores as independent variables. These factor scores provide a simplified latent structure that aids in predicting the dependent variable. This approach allows us to analyze the unique contribution of each factor to the dependent variable. Multiple Regression Analysis was conducted to determine the best linear combination of these five factors for predicting teachers’ EFL lexical attrition. Then, we may also learn the explanatory power and predictive ability of these five factors that caused lexical attrition. Statistical assumptions, such as the normal distribution of residuals and the non-linear correlation between predicted variables and residuals, were all met in the analysis. The regression method of “enter” showed that the combination of the IVs significantly predicted teachers’ scores on the word self-report scales; in other words, predicted the degree of teachers’ EFL lexical attrition, F (5,99) = 14.32, p < .05, with all of them significantly contributing to the prediction (p < .05) (Table 8). The beta weights, presented in Table 8, suggested that initial language proficiency and language contact contribute most to predicting scores on the word self-report scales. The R² value was .42, which indicated that 42.00% of the variance in scores on subjects’ word self-report scales was explained by the model. The standardized regression formulation is subjects’ scores on the word self-report scales = 0.36 × initial language proficiency + 0.29 × language contact + 0.13 × teachers’ concerns about language proficiency + 0.08 × teachers’ attitude towards English learning + 0.03 × artificially-intervened language use (Table 8).

OPEN IN VIEWER

Table 8.

Multiple Linear Regression: Important Statistics (n = 105).

Variables		R	R 2	Adjusted R2	F (5,99)	β	t (99)	Tolerance	VIF
DV	SWSS	.648	.420	.390	14.32*
	LC					.29	2.75*	0.63	1.59
	TAEL					.08	0.80*	0.67	1.50
IVs	TCLP					.13	1.60*	0.85	1.18
	AILU					.03	0.25*	0.62	1.61
	ILP					.36	3.94*	0.71	1.41

Note. VIF = variance inflation factor.

*

p < .05.

Discussion

The present study conducted a comprehensive empirical investigation into the phenomenon of EFL teachers’ lexical attrition. The findings of this study provide evidence of EFL teachers’ lexical attrition occurring after the completion of formal language learning. The identified factors contributing to this attrition include language contact, initial language proficiency, concerns about maintaining language proficiency, teachers’ attitude toward language learning, and artificially-intervened language use. The findings of this study identified the following issues worthy of further discussion.

First, the overall results of our study suggested that there is significant attrition in EFL teachers’ vocabulary after beginning their teaching careers, and that confirmed our prediction, which assumed EFL teachers’ lexical attrition exhibits similarities to that observed in other FL learner groups in China. Our finding was in line with the assertions of previous researchers, who have noted that various aspects of language are susceptible to attrition among language learners within 3 to 6 years after the cessation of formal training (de Bot, 2004; Jessner et al., 2021; Mehotcheva, 2010; Ni, 2009; Schmid, 2022, 2023; Schmid & Mehotcheva, 2012). As professionals with prolonged exposure to the FL, FL teachers are typically regarded as having more opportunities to engage with FL compared to individuals in other occupations (Horwitz, 1996). However, our research findings confirm that EFL teachers experience a certain degree of lexical attrition. While a minority of teachers may succeed in maintaining or improving their EFL vocabulary through unwavering efforts, overall, junior high school EFL teachers exhibit a declining trend in their English vocabulary proficiency after the commencement of their careers, a phenomenon that is understandably concerning. These research findings align with previous studies (Choi & Lee, 2016; Valmori & De Costa, 2016), emphasizing that, despite the expectation for FL teachers to maintain a high level of FL proficiency, FL development is indeed a prolonged journey. FL teachers, especially non-native language instructors, are perpetual learners (Horwitz, 1996) who must consistently uphold their language proficiency and continuously expand their language knowledge. This reiterates the enduring nature of language development and the ongoing necessity for language learning, even within the community of language educators (Braine, 2010). The conflict between language acquisition and attrition is a fundamental aspect of language development (Jessner, 2003). Pre-service education primarily focuses on language acquisition, with proficiency progressing from basic to advanced levels. However, once in their professional roles, English teachers’ work, social lives, and family commitments limit their time for language learning and practice (Wichadee, 2012). Additionally, FL education in secondary schools in China is often constrained by exam-oriented teaching, leading to instructional content and activities mostly revolving around textbook content, and lacking genuine language interaction and communication (Wenfeng & Gao, 2008). This mechanistic teaching approach has, to some extent, dampened English teachers’ professional enthusiasm, learning interest, and breadth of expertise, resulting in a gradual decline in their vocabulary proficiency. According to the descriptive analysis, the words categorized as “B” (Difficult Retrieval) and “CW” (Wrong Retrieval) in this word test were found to be temporarily difficult to recall, even though they were not retrieved correctly. An explanation for this observation can be attributed to the Saving Paradigm Hypothesis (Schmid, 2006), which suggests that forgetting is akin to misplacing something rather than permanently losing it. In other words, when information becomes temporarily inaccessible, forgetting occurs (Weltens & Grendel, 1993), implying that lexical attrition in teachers, similar to other FL learners, is not a case of forgetting but rather a temporary retrieval difficulty. The findings from the interviews further support this conclusion:

(1) I primarily use Chinese in my teaching. The words I frequently use are limited to the textbooks I teach from. This can have an impact on my English proficiency. I often feel uncertain about how much of the foreign language skills I acquired are still retained. (Teacher D, a 27-year-old female)

(2) What frustrates me the most is that I know for sure that I learned these English words, but I just can’t recall them. It really bothers me. (Teacher G, a 30-year-old male)

Second, this study employed a self-designed questionnaire to investigate lexical attrition among EFL teachers in Chinese contexts. Through EFA, it identified five key factors contributing to EFL teachers’ lexical attrition: teachers’ language contact, initial language proficiency, teachers’ concerns about maintaining language proficiency, teachers’ attitude toward language learning, and artificially-intervened language use. The related items for each factor can be found in Table 5. These findings align with previous research that has identified language attrition among FL learners, attributed to personal, psychological, and environmental factors (Mickan et al., 2023; Schmid, 2022). However, earlier research primarily focused on language learners or non-language teacher groups and overlooked investigations into language attrition among language teachers. Furthermore, early research overlooked the fact that non-native language teachers themselves are considered advanced language learners (Horwitz, 1996), and when they do not engage in regular FL exposure and training, their FL proficiency can also decline (Braine, 2010). Existing literature has solely emphasized the effects of maintaining and improving FL teachers’ FL proficiency on students’ FL learning outcomes (Choi & Lee, 2016; Valmori & De Costa, 2016). However, it has not directly indicated that FL teachers experience attrition in their FL proficiency after commencing their careers, and the direct causes of potential FL attrition have also not been explored. This study identified five factors influencing the lexical attrition of EFL teachers, highlighting the complexity of this phenomenon of teachers’ language attrition. Our findings support previous research, emphasizing the role of language contact in the daily lives of English teachers and its impact on vocabulary attrition (Jessner et al., 2021; Mehotcheva & Mytara, 2019; Mickan et al., 2023; Schmid, 2022). Despite their profession providing continuous exposure to English, various factors limit the quality and quantity of this exposure as teachers juggle different responsibilities (Wichadee, 2012). Additionally, in line with previous research, this study confirms the critical role of initial language proficiency in teachers’ lexical attrition (Hansen, 2011; Mehotcheva & Mytara, 2019; Schmid, 2011, 2022), underscoring the importance of pre-service language education and proficiency assessment for prospective English teachers. Also, teachers’ concerns about maintaining proficiency are crucial, as the fear of skill loss can motivate active language learning and practice, thereby reducing language attrition. Attitudes toward language learning also affect attrition, with positive attitudes contributing to maintenance. Artificially-intervened language use may help slow down language attrition, emphasizing the need to provide teachers with more lectures, training, and interventions regarding language proficiency. The findings reveal the multifaceted nature of lexical attrition among EFL teachers, with factors like language contact, initial proficiency, concerns, attitudes, and teaching practices all playing significant roles. These factors were also found to be positively related to teachers’ scores on word self-report scales, indicating their predictive value for the extent of lexical attrition. Hence, a conclusion drawn from the findings is that maintaining a positive attitude towards EFL learning can significantly impact teachers’ commitment to preserving and enhancing their language competency. The motivation to improve language proficiency leads to increased exposure to language knowledge, while heightened language contact contributes to higher FL proficiency. Ultimately, a higher frequency of language use promotes language output and helps counteract language attrition. This aligns with the input-proficiency-use cycle proposed by B. Z. Pearson (2007). Recognizing the relationship among these factors is essential for designing effective interventions and strategies to support EFL teachers in maintaining language proficiency throughout their careers.

Third, the results of the multiple regression analysis underscore the importance of initial language proficiency and language contact as the most influential factors in predicting lexical attrition among EFL teachers. This aligns with previous research findings that emphasize the significance of discontinuing language contact and the level of initial language proficiency in foreign language retention or attrition (Alharthi, 2015; Hansen, 2011; Mickan et al., 2023; Ni, 2009; Schmid, 2022). A clear pattern emerges when comparing teachers’ current vocabulary proficiency to their initial language proficiency, revealing significant attrition. Teachers with higher initial proficiency experience less lexical attrition, indicating a greater level of confidence in their language abilities (Schmid, 2022). These findings, to some extent, resonate with Stanovich’s (1986) notion that “the rich get richer,” suggesting that learners with a stronger foreign language vocabulary have a greater capacity to acquire new vocabulary and related knowledge while losing less of their existing vocabulary. Assessing teachers’ language contact primarily relied on examining their extracurricular FL learning behaviors. The data indicate that teachers with more intense professional activities (both in terms of classroom hours and extracurricular language usage) and higher frequencies of engagement in extracurricular activities such as reading, communication, and writing tend to have higher existing vocabulary levels and greater resistance to attrition in their post-career vocabulary proficiency. This aligns perfectly with the statement made in Jessner et al.’s (2021, p. 25) study that “The factor that would appear most obvious as exerting the strongest influence in attrition is (lack of) language use and contact: It stands to reason that the less a language is used, the more likely it is to deteriorate and vice versa.” Therefore, for EFL teachers, it is only natural that the focus of current teacher training should be on their present language proficiency, increasing their language exposure opportunities, and enhancing their vocabulary.

Conclusion

The current study serves as a pioneering exploration into the phenomenon of lexical attrition among EFL teachers in China, offering valuable insights and laying the groundwork for future investigations into language attrition among FL teachers. Our findings underscore the significant lexical attrition experienced by EFL teachers following the completion of their formal language training. This lexical attrition patterns among Chinese junior high school EFL teachers exhibit significant similarities with those of other FL learners. Factors such as teachers’ language contact, initial language proficiency, concerns about maintaining language proficiency, attitude toward language learning, and artificially-intervened language use frequency can all contribute to the post-career lexical attrition among teachers. These findings offer profound insights into language attrition, carrying both practical and pedagogical implications for the continuous professional development of EFL teachers and pointing the way towards future research directions. Also, they serve as a valuable reference for forthcoming studies focusing on language attrition among FL teachers. First, our study unequivocally confirms that the phenomenon of lexical attrition among EFL teachers is not only real but also bears a resemblance to attrition observed in other FL learner groups. Considering the pivotal role EFL teachers play as primary facilitators of English language acquisition in China and their status as successful advanced learners, our findings substantiate the theoretical proposition put forth by Choi and Lee (2016). This proposition underscores the critical importance of both perceived linguistic competence and pedagogical expertise in optimal English teaching. It is essential to consider these factors when comprehending the professional state, growth, and practices of English teachers. Second, the identification of the five factors contributing to lexical attrition among EFL teachers not only deepens our understanding of the challenges they encounter in maintaining perceived linguistic competence but also underscores the intricate complexities associated with being a language teacher. Additionally, it is indispensable for the development of interventions aimed at mitigating lexical attrition. Educational institutions can draw on these insights to provide external support, such as offering training courses on foreign language knowledge, providing language learning resources, fostering peer support, organizing academic workshops, and providing ongoing guidance (Russell, 2012). Moreover, offering academic incentives to teachers can motivate them to actively engage in language learning and enhance their awareness of the significance of their own language proficiency, a potential boon. Third, given that language contact and initial language proficiency have the most pronounced impact on teachers’ lexical attrition, in order to facilitate maximum English language usage among teachers and increase teachers’ English exposure, it is crucial to ensure that pre-service teachers, first and foremost, acquire and maintain at least a minimum level of language proficiency. Ensuring they have a solid language foundation, schools and educational institutions can actively create opportunities for English exposure. This includes increasing opportunities for English use within the educational environment and encouraging teachers to participate in international educational projects or exchange activities, thereby assisting them in maintaining language fluency. Simultaneously, teachers themselves can embark on self-directed learning during the attrition phase (Xu, 2010). Activities like watching movies, reading books, or engaging in written correspondence, such as letters and emails, naturally contribute to language retention (Mehotcheva, 2010). Lastly, recognizing the pivotal role of teachers’ attitudes and motivation in reducing language attrition, schools can institute regular language assessment and feedback mechanisms. These mechanisms help teachers gauge their current proficiency levels in foreign languages, fostering reflective practices. This reflective process entails a critical self-examination of their professional development, self-assessment of language attrition, recognition of its adverse effects, and ultimately, enhancement of their foreign language proficiency (Akbari, 2007). Such an approach effectively mitigates the extent and pace of foreign language attrition.

However, this study has inherent limitations that suggest directions for future research. First, the sample is not diverse enough. This study exclusively surveyed junior high school teachers. However, the absence of data from primary and university teachers limits the generalizability of the research findings. For instance, university teachers, without the pressure of standardized testing, might employ more English in their classrooms, potentially leading to variations in lexical attrition compared to junior high school teachers. Further investigations are warranted to assess if the findings and implications of the present study would apply to similar and dissimilar samples of subjects. Second, while this study examined teachers’ lexical attrition, it did not explore its long-term effects, or its impact on teacher efficacy. Future research should delve deeper into these aspects for a more comprehensive understanding of the phenomenon. Third, this study does not encompass all the factors contributing to EFL teachers’ lexical attrition. Language attrition is a complex and dynamic process influenced by psychological, physiological, linguistic, and sociological factors. Further research is needed to explore additional factors that affect lexical attrition among EFL teachers. Lastly, it would be worthwhile to explore whether there are temporal variations in the lexical attrition among EFL teachers.