Enhancement of EFL learners’ lexical retention: The role of social constructivism

Abstract Vocabulary plays a significant role in the learning process of English as a Foreign Language (EFL) students, especially undergraduates who need to read academic materials and listen to specialized lectures in English. However, retaining lexical items is still a challenging task for many EFL learners. The current study seeks to examine the effectiveness of a constructivist approach in improving EFL learners’ vocabulary retention, mainly meaning recall, as well as investigate their attitude towards this learning method. To achieve this aim, a quasi-experimental design, including a pretest and a posttest, was employed with the participation of 75 EFL Vietnamese undergraduates. The participants were assigned into two groups: the control group (Group 1, N = 37) and the experimental group (Group 2, N = 38). While Group 1 was taught in a conventional way, Group 2 received vocabulary training following the constructivist approach. At the end of the eight-week treatment, a questionnaire was administered to the experimental group to gauge their view on vocabulary learning using constructivism. The results showed that Group 2 retained lexical items more effectively than Group 1. Furthermore, the experimental group had a favorable opinion of learning vocabulary in the constructivist approach. Recommendations for future studies as well as implications for teachers regarding vocabulary and constructivism, were presented in detail.


Introduction
Nation's landmark study (2006) has proved that L2 learners need to know at least 8,000 word families for reading and 6,000 for listening. This highlights the crucial role of vocabulary acquisition and retention for ESL/EFL learners, especially undergraduates, such as those in linguistic fields, who are required to read academic texts and listen to specialized lectures. Therefore, a large body of research has been performed to find optimal ways to aid learners' lexical retention. Many techniques have been proposed and shown to be effective, such as reading while listening (Webb et al., 2013), listening (Jin & Webb, 2020), viewing lectures (Dang et al., 2022), doing tests (Kasahara & Kanayama, 2021), or using input repetition (Duong et al., 2021). With the advancement of technology, modern approaches have been utilized to boost vocabulary size, namely CALL (computer-assisted language learning) (Bi & Shi, 2019;Nguyen et al., 2022;Shi, 2017) or MALL (Mobile-assisted language learning) (Lu, 2008;Nguyen & Le, 2023). These various teaching and learning approaches have been tested and implemented inside the classrooms with the aim of improving L2 acquisition. However, the common problem among these methods is that they seem to place learners in passivity, with little room for creativity and authentic practice.
One of the prominent ways to buoy an active and creative atmosphere in the classroom is the employment of the constructivist approach. This has been widely researched and implemented as it centers on the learner, not the teacher; however, there are two major issues in previous studies. First, most of these implemented models did not adhere to the key elements of constructivism as claimed (Baviskar et al., 2009;O'Connor, 2022). Second, the focus was primarily placed on scientific subjects (Bostock, 1998;Huffman et al., 2003;Hằng et al., 2015). Few works concentrated on how to assist learners in developing their vocabulary via the incorporation of the constructivist approach (Aravin & Bhuvaneswari, 2023;Du, 2013;Lin, 2015). These studies, albeit pertaining to constructivism to some extent, did not fully adhere to the approach, which might have led to questionable findings.
In Du's empirical research (Du, 2013) on using the constructivist model combined with multimedia tools to enhance learners' productive vocabulary, she divided 78 participants majoring in International Trade into two groups: control (G1) and experimental (G2). While the control group undertook traditional training, the experimental group studied vocabulary following the constructivist approach. The results showed that students in G2 scored significantly higher than those in G1 in the posttest despite the similar scorings in the pretest. Moreover, Du also found that most G2 students adopted a positive attitude towards the learning style of constructivism via the analysis of a 5-point Likert scale survey. However, the paper did not clearly give detailed descriptions of the sessions that took place in the constructivist class, raising the question of whether elements of constructivism were adequately applied or not. Another concern is that the researcher let students in G2 watch film clips, write down unfamiliar words, discuss in groups, and organize them to create real-life contexts, which was hard to measure appropriately as there was no guarantee that the words students chose were truly "unfamiliar" or not.
With similar goals, Lin (2015) also conducted an experimental study on 37 students (with a control group of 38 participants) to test the effectiveness of constructivism on vocabulary acquisition. Via the design of a pretest and a posttest, coupled with a detailed procedure of a typical constructivist class session, the author came to the conclusion that the new approach (constructivism) was superior to direct instructional teaching in boosting students' active lexical acquisition. Although the employed method in the study included some features of constructivism, such as group work and class presentations, it did not strictly adhere to the vital constructivist elements as the teacher still tried to control the class in many ways, from giving immediate corrections or feedback to being the center of the lessons. Further, the posttest contained not only multiple-choice questions but also writing, which made the grading process more subjective.
The most recent work about constructivism and vocabulary training was carried out by Aravind and Bhuvaneswari (2023). The authors examined the role of social constructivist theory in promoting English as a second language (ESL) learners' vocabulary acquisition. Via a survey design with a questionnaire administered to 60 undergraduates from different universities, they found that constructivism significantly helped the participants develop their vocabulary. Nonetheless, this outcome originated from the students' self-reported responses to the questionnaire, which was quite subjective. In other words, whether they actually improved their vocabulary acquisition or not remains unclear.
Overall, several gaps in previous research regarding constructivism and vocabulary were already highlighted. To cope with these issues, this study aims to explore whether the constructivist learning approach is effective in improving learners' vocabulary retention and to what extent it is more or less effective than the teacher-centered model. Also, it seeks to investigate learners' attitude towards the implementation of constructivism in vocabulary lessons. This study, consequently, is in search of answers to the following questions: (1) Is the constructivist approach effective in improving learners' vocabulary retention?
(2) To what extent is the constructivist approach more or less effective than the teachercentered method in improving learners' vocabulary retention?
(3) What is learners' attitude towards learning vocabulary via the constructivist approach?
It is significant to conduct the present study for three reasons. First, it contributes further to the literature on constructivism and lexical retention, shedding light on the findings of previous research, especially when scholarship on this topic is relatively limited. Second, it may alter the way vocabulary is taught in the classroom, from a passive mode to a more active style, provoking learners' interest and engagement. Finally, teachers and institutional leaders can modify the curriculum to make vocabulary lessons more relevant to constructivism.

Definition and categories
In the constructivist view, learning is a process of meaning creation; it is how individuals make sense of their experiences (Caffarella & Merriam, 1999). Mvududu and Thiel-Burgess (2012) claim that constructivism is commonly regarded as a method for assessing learners' degree of comprehension and demonstrating that this comprehension can evolve into higher-level thinking. Therefore, constructivism relates to the manner of learning and thought.
In cognitive constructivism, learners are posited as active agents in knowledge construction (Piaget, 1971). Using this framework, students actively reconstruct knowledge in highly nuanced, unique ways, constructing malleable conceptual topologies based on prior knowledge, formal educational experiences, and other cues that aid comprehension. This point of view places emphasis on one's own cognitive processes and labels fellow students as disruptive thinkers if conventional wisdom is challenged (Piaget, 1985).
According to social constructivism, as opposed to cognitive constructivism, knowledge is seen as rooted in peoples' social contexts (Vygotsky & Cole, 1978). This theory holds that knowledge is formed via exposure to local and global cultural contexts and comes from engagement with an increasing variety of communities of practice (Cole, 1990;Scribner, 1985). In addition, Vygotsky's concept of "zone of proximal development" states that emerging mental functions must be supported and quantified through collaborative activities in which students engage in heuristic tasks or problem-solving with the help of more experienced specialists (Vygotsky & Cole, 1978). Students internalize the auxiliary language and strategies employed on the social plane and develop the skills necessary to do these actions without support in the future. A central tenet of the social constructivist stance is that schooling provides students with the linguistic, symbolic, and communicative skills necessary to participate in more complex forms of thought (Olson, 1986).
In the present study, social constructivism is the type under exploration and is referred to as constructivism or the constructivist approach.

Criteria for a constructivist lesson
Constructivism has been widely applied in many fields, yet many researchers and educators are still not well aware of this broad term (Rovegno & Dolly, 2006;Sjøberg, 2010). In fact, it has been wrongly applied in the classroom context due to the missing of certain steps (Baviskar et al., 2009;O'Connor, 2022). To cope with this issue, Gagnon and Collay (2001), as well as Baviskar et al. (2009), proposed essential criteria to which a constructivist lesson should adhere.
In 2001, Gagnon and Collay emphasized six fundamental components of constructivism, including situation, groupings, bridge, questions, display, and reflection. The situation establishes the agenda for student participation by outlining the objectives and tasks. Groupings are the social structures and group interactions that will bring students together during the learning episode's tasks and forms. Before introducing students to a new subject, the bridge is the process of highlighting their prior knowledge. The bridge can redirect their efforts to the new content when they are able to interpret it inside their cognitive maps, values, attitudes, and expectations. The fourth part stimulates, motivates, and integrates student thought and information sharing. During a learning episode, questions are prompts or responses that encourage, extend, or synthesize student thought and communication. This social context allows students to reply to questions (display) made by the teacher, peers, or guests about what Gagnon and Collay refer to as objects of learning. Finally, reflection provides students and instructors with chances to critically consider and discuss their individual and collective learning. This motivates all participants to consolidate their knowledge, apply learning artifacts to other curricular components, and anticipate future learning episodes.
While Gagnon and Collay (2001) characterize the constructivist approach as having six components, Baviskar et al. (2009) distill the activities, structure, content, or setting of a constructivist lesson into four essential elements: eliciting prior knowledge, creating cognitive dissonance, applying knowledge with feedback, and making reflection. The authors illustrate in their research that without a system for extracting students' existing knowledge, new material cannot be presented in a way that allows it to be assimilated into the learner's construct. Similarly, the new material will be dismissed or poorly integrated if the student's attention is not drawn to past knowledge. As a result, eliciting prior knowledge may be considered the first fundamental constructivist classroom prerequisite. The development of cognitive dissonance is the second need. At this level, the student must discriminate between previous knowledge and new information (Inch, 2002;Sewell, 2002). The third component, applying knowledge with feedback, is consistent with the idea of Windschitl (2002) that learners who do not analyze and change existing knowledge in light of new information are more likely to misunderstand or reject it. Aside from validating their inventions, the application assists students in further defining the interconnection of new knowledge to a broader range of settings, thereby permanently integrating the new knowledge. A learning reflection is the fourth aspect. The learner must be aware of the learning after gaining and evaluating the most current knowledge. The presence of the reflection criterion greatly makes the session more constructivist, even if it is not a formal component of the lesson plan.
Obviously, there are certain similarities between the criteria by Gagnon and Collay (2001) as well as Baviskar et al. (2009). In this study, the four criteria stated by Baviskar et al. (2009), were employed as they were more updated.

Constructivism and vocabulary retention
Retention has been referred to as the ability to recall or remember things after a period of time, depending on teaching quality, learners' interest, and material meaningfulness (Richards et al., 2003). Based on this definition, vocabulary retention is supported by a number of prominent theories. First, lexical items should be learned gradually through the course of time, known as distributed learning, not all at a time, regarded as massed learning. In other words, distributed learning has been proven to be more effective than massed learning, especially in L2 vocabulary (Kornell et al., 2009;Nakata & Suzuki, 2019;Nakata & Webb, 2016). This is because absorbing a long list of vocabulary seems to be a daunting task that demotivates learners from acquiring new words effectively. Second, the involvement load hypothesis (Laufer & Hulstijn, 2001) posits that vocabulary learning should include three elements: need, search, and evaluation. Teachers are recommended to give students a problem to elicit their interest, provide them with time and tools to search for the answers, and report their findings which are then evaluated by teachers and others for confirmation. Another theory is the testing effect, which claims that vocabulary gains can be achieved via quizzes or tasks in which learners can test their understanding (Arnold & McDermott, 2013). The explanation is that learners see the gap in their knowledge and try to find ways to solve it, which helps them remember words longer. The connectionist theory also plays an important role in forming and constructing linguistic knowledge. Connectionism pinpoints the relationship between linguistics and authenticity, meaning that learners should be provided with real-world materials (Ellis, 2003;Lewis, 1993) during their lessons. This theory also assumes that information of the lexicon is stored in neural networks, which are fortified through the course of time when learners frequently interact with one another and with authentic materials to make sense of form-meaning relationships (Elman et al., 1996).
Constructivist criteria (Baviskar et al., 2009) correlate well with these theories. First, in each lesson, only a short list of vocabulary is introduced (distributed learning, Kornell et al., 2009), and the previously learned lexical items could be the base for the acquisition of the new words (eliciting prior knowledge). Second, students need to work together (connectionism, Ellis, 2003) to find out the meanings and the usages of the target vocabulary (involvement load hypothesis, Laufer & Hulstijn, 2001) to fill in their knowledge gaps (creating cognitive dissonance). Third, they come to the front to present their findings, making examples using the new words, and receive feedback from others (applying knowledge with feedback) to confirm their information. Finally, making reflections on what has been learned gives students a chance to test whether they appropriately acquired the target vocabulary (testing effect, Arnold & McDermott, 2013).
Previous studies on vocabulary and constructivism (Aravind & Bhuvaneswari, 2023;Du, 2013;Lin, 2015)., as presented earlier, failed to adhere to the constructivist criteria. Therefore, further research is highly needed to examine the effectiveness of the constructivist approach in enhancing EFL learners' vocabulary retention.

The educational setting and participants
This research was conducted in a private university in Vietnam, where the researchers are employed. Since the Summer semester of 2021, the administration of this university has pushed academics to implement the social constructivism approach in teaching and studying applicable subjects, such as English. Therefore, this experimental study was necessary to assess the efficacy of the social constructivist method in boosting students' lexical acquisition in their English class.
To anticipate the required number of participants for the study, an "a priori power analysis" test was performed in the software the G*Power 3.1 (Faul et al., 2009). The result showed that the minimum number was 54 (α = .05, power = .95, ANOVA 2 × 2 between-within-subjects design, medium effect f = .25).
Participants of the study were 75 students majoring in Engineering at a private university in Vietnam, aged 19-20. They came from two different classes, one with 37 students (28 males and 9 females) and the other with 38 students (27 males and 11 females). At the time of the study, they were taking an English course about Academic Reading and Writing to improve language skills for their major. Their English proficiency was equivalent to B1 in the CEFR (Common European Framework of Reference for Languages) as they had already taken a placement test designed by Pearson Education (Ascher & Saslow, 2022). Their textbook was "University Success, Reading, Advanced", equivalent to B2 level (CEFR), covering a wide array of academic topics and critical thinking skills. They were put into two groups, the control group (G1, N = 37) and the experimental group (G2, N = 38). While the control group was taught in a usual way (teacher-centered model), the experimental group was trained using the constructivist approach.
All of the participants agreed to join the research, which was approved by the school's management board.

Instruments
To answer the first and second research questions, a pretest and posttest design was adopted. In the pretest, all participants were shown a list of 64 keywords extracted from the textbook (for each reading text, there is a list of 6-8 keywords in a box, pre-selected by the book authors), and they were required to write down L1 meaning next to each word. If they did not know any item, they could make a guess or skip it. The tests took place in 30 minutes in the classroom, with no discussions or dictionaries allowed. The vocabulary in both tests was exactly the same, but the order of the words was different to avoid any possible impact of testing effect (Arnold & McDermott, 2013).
Afterward, the two researchers worked separately and scored the papers, with 1 for the correct answer (exact or close L1 meaning) and 0 for the wrong answer (wrong L1 meaning or blank). Upon completion, an inter-rater reliability test was run to compare the results, and a high Cohen's Kappa value (k = .95) indicated that the scorings were very similar. Any discrepancies in the scores were solved via discussion between the raters until the agreed final scores were reached for all items. The keywords were presented in Table 1 below: To answer the third research question, a questionnaire using a 6-point Likert scale was employed (1 as Strongly Disagree, 2 as Disagree, 3 as Slightly Disagree, 4 as Slightly Agree, 5 as Agree, and 6 as Strongly Agree). It was designed based on sound research on constructivism (Baviskar et al., 2009;O'Connor, 2022;Vygotsky & Cole, 1978) following robust questionnaire construction guidelines (Dörnyei & Taguchi, 2009). The questionnaire was thoroughly reviewed by two EFL teachers with Ph.D. degrees in TESOL (Teaching English to Speakers of Other Languages) to ensure its content validity.
There were eight items in the questionnaire delivered to the students in the experimental group at the end of the course to measure their view on learning vocabulary using the constructivist approach. These items focused on learners' vocabulary comprehension and memorization (item 1 to item 4), interaction with others (item 5), state of being active (item 6), contribution to the lessons (item 7), and overall satisfaction (item 8).

Descriptions of a typical lesson in two classes
As this study only focused on vocabulary aspects, the following teaching and learning procedures only reflected the corresponding parts. The time for each vocabulary lesson was 45 minutes.

The teacher-centered model
In an ordinary session at the school, teachers delivered the lesson following the teacher-centered model, with the support of technological tools such as a projector, laptops, and Internet access. Although there were interactions in the classroom, the center was still placed on the teacher.
A typical vocabulary session following the teacher-centered model included three main stages. In the first stage, which lasted for about five minutes, the teacher introduced the general topic of the lesson. In the second stage (20 minutes), the teacher showed the word on PowerPoint via a projector, presented its phonetic transcript, read aloud the word, and asked the students to repeat the pronunciation twice or three times. After that, a sentence containing the key word was shown on the slide, and the students were required to guess its part of speech and meaning. The teacher confirmed the information before moving on to the next word. The procedure was repeated until all the target words were presented. In the last stage (20 minutes), the teacher asked the students to complete the blank-filling task in the book using the newly learned words. The students reported their answers, and the teacher confirmed the information. A summary of all the steps can be found in Appendix A.

The constructivist model
The constructivist approach followed the four criteria proposed by Baviskar et al. (2009), which were in line with the six key elements suggested by Gagnon and Collay (2001). Similar facilities were equipped in the classroom as in the other class.
There were five main stages in a typical vocabulary lesson following constructivism. In the first stage, the teacher introduced the general topic of the lesson in three minutes. In the second stage (7 minutes), the teacher tried to elicit students' prior knowledge of the target vocabulary by asking them to report everything they knew about these words (e.g., form, meaning, pronunciation) and wrote down their reported information on the board. In the "creating cognitive dissonance" stage (15 minutes), the teacher asked the students to read the text in the book, highlighted the target words, and work in groups to guess the words' details (e.g., form, meaning, pronunciation). They were allowed to use technological tools to help them find the required information. Following this, the teacher invited each group to come to the front of the class and report their findings while others listened to and gave feedback on their friends' information. The teacher confirmed the answers before proceeding to the next step. In the fourth stage (15 minutes), the students were required to work in groups again and apply the newly learned words by making a conversation. Some groups were invited to present their conversations in front of the class while others listened, asked questions, and/or gave feedback. In the final stage (5 minutes), the teacher asked the students to compare what they just learned with what they had previously known about the target words (on the board), reflecting on their vocabulary acquisition (see Appendix B for a summary of the steps).

Procedures
Data were collected within two months, which was also the English course length (eight weeks of training, three hours a day, three days a week). On the first day of the course, the students in two groups took the same pretest on paper. Then, the control group was trained in a conventional way, while the experimental group was conducted following the constructivist approach. On the last day of the course, both groups took the paper-based posttest, which contained the same keywords as those in the pretest, yet in random order. During the tests, no materials or discussions were allowed. The participants had 30 minutes to provide L1 meanings to the lexical items on the list. After that, students in the experimental group were given ten minutes to complete a short survey on their attitudes toward learning vocabulary via the constructivist approach. The procedure was illustrated in Table 2 as follows:

Data analysis
All of the data were first imported into an Excel file before being processed in SPSS version 27.0 (Statistical Packages for Social Sciences) and R version 4.2.1.
First, the two researchers graded the posttest papers individually with the same criteria as in the pretest scorings and then ran an inter-rater reliability test to check the similarity. Cohen's Kappa (k = .98) revealed that there was a strong agreement between the two raters. After that, the researchers discussed with each other to determine the final scores for items where grading was different.
Next, Shapiro-Wilk tests were run to check the distribution of normality of the scores in the pretest and posttest. Results showed that both pretest scores (W = .53, p < .001) and posttest scores (W = .61, p < .001) were not normally distributed. Then, the Generalized Linear Mixed Model (GLMM) was adopted to examine each approach's effect on participants' vocabulary retention. This method could cope with data of non-normality and, at the same time, address individual differences that other parametric tests, such as ANOVA, MANOVA, or t-tests could not handle (Linck & Cunnings, 2015;Norris, 2015). In the present study, the correct scores were treated as the dependent variable, while groups (control and experimental) and test time (pretest, posttest), were regarded as fixed effects (dependent variables). The random effects were the participants and lexical items, and the type of family in the GLMM was "binomial," as the scores were marked as 1 or 0. Consequently, the fitted model in the study, run in R with the lme4 package (Bates et al., 2014), was as follows: The survey was analyzed using SPSS, with Cronbach's alpha value as an indicator of reliability and consistency and Mean values as indices of participants' attitudes.

Research question 1: is the constructivist approach effective in improving learners' vocabulary retention?
As the pretest scores were of non-normality, a Mann Whitney U Test was performed to check if there was a significant difference between the two groups. The result in Table 3 revealed that the pretest scores in the control and experimental group were statistically similar (Z = −.704, p = .482). Table 4 demonstrated that participants in both groups improved their vocabulary acquisition. Specifically, the posttest scores of the control group (M = .49) were more than doubled the pretest scores (M = .023). Similarly, those in the experimental group almost tripled their scores when comparing the Mean values in the posttest (M = .75) and pretest (M = .24). Therefore, it could be concluded that the constructivist approach was effective in boosting learners' lexical retention. However, whether the differences in the scores between the pretest and posttest as well as between the two groups, were statistically significant or not needed further analysis via the GLMM.

Research question 2: to what extent is the constructivist approach more or less effective than the teacher-centered method in improving learners' vocabulary retention?
As can be seen from Table 5, there was a significant effect on the experimental group (β = 1.150, SE = .062, z = 18.585, p < .001), indicating that students learning in the constructivist approach scored significantly higher than those in the control group. Moreover, there was a significant negative effect of the pretest (β = −1.188, SE = .063, z = −18.825, p < .001). This showed that students in both groups scored much lower on the pretest compared to the posttest. In other words, both groups achieved better vocabulary retention at the end of their course.

Research question 3: what is learners' attitude towards learning vocabulary via the constructivist approach?
Before analyzing the survey, a Scale test was run to examine the reliability. The Cronbach's alpha value was .85, indicating that the study was reliable and internally consistent. The descriptive   analysis was demonstrated below, with Disagreement (Strongly disagree and Disagree), Uncertainty (Slightly disagree and Slightly agree), and Agreement (Agree and Strongly agree). Table 6 illustrates that most participants adopted a positive attitude towards the learning approach of constructivism. In particular, 93.1% believed that they were active in their lessons. In addition, 68.5% of the participants felt that they were able to understand the vocabulary better while 63.2% reported that they could remember the target words more effectively. Moreover, 81.6% thought that they themselves had contributed to the lessons. Overall, more than half of the students (63.1%) were happy with the way they learned vocabulary, and no one was dissatisfied with the learning method used.

Research questions 1 and 2: the effectiveness of the constructivist approach in improving L2 learners' vocabulary retention
Via the quantitative analyses, the present study found that learning vocabulary in the constructivist approach helped students to remember lexical items, especially L1 meanings, effectively. In fact, the experimental group scored much higher in the posttest (M = .75, SD = .43) than in the pretest (M = .24, SD = .43), which was statistically significant (β = −1.188, SE = .063, z = −18.825, p < .001). Furthermore, those learning in the constructivist approach (M = .75, SD = .43) outperformed those following the conventional method (M = .49, SD = .50). The difference in posttest scores between the two groups was substantial (β = 1.150, SE = .062, z = 18.585, p < .001). These outcomes were confirmed via the questionnaire in which more than half of the students in the experimental group believed that they could understand and remember vocabulary better. Such findings align with previous research such as Du (2013) and Lin (2015). To illustrate, Lin (2015) reported that learning vocabulary in the constructivist approach could acquire active lexical items more effectively than those following the conventional teaching method.
There were multiple plausible explanations for these results. First, while the involvement load hypothesis (Laufer & Hulstijn, 2001) emphasizes the need, search, and evaluation for vocabulary gains, the constructivist approach even maximizes the potential of these elements. In fact, the participants were challenged with a problem, or the cognitive dissonance in constructivism (Baviskar et al., 2009), to which they needed to search for answers. In order to solve this issue, they worked together, looked for information with the aid of technological tools, combined with their prior knowledge, and came up with the best ideas. Then, they presented these solutions to their friends and teachers, who gave them feedback (evaluation). These two theories support each other and achieve the synergies that boost learners' lexical gains to a high rate. Another explanation is that neural networks were strengthened via authentic practice, such as using the learned words to make examples about real-life situations or conversations, and via multiple evaluations (from members in groups, from other groups, and from the teacher). This process is well explained by the connectionist theory in the literature review (Ellis, 2003;Elman et al., 1996;Lewis, 1993). In addition, given the description of a typical lesson, it could be inferred that students in the constructivist class were involved in a variety of activities in which they were the key players, not the teacher. With diverse tasks, the participants were exposed to the lexical items more frequently, which might explain why they remembered the words better.
These findings are meaningful as they shed light on the positive role of constructivism in promoting EFL learners' vocabulary, especially in the circumstance that previous research failed to follow the design of the constructivist approach (Baviskar et al., 2009;O'Connor, 2022). Therefore, it could be indicated that constructivism, when applied appropriately, could bring significant results.

Research question 3: learners' attitude towards learning vocabulary in the constructivist approach
Through the analysis of the questionnaire, it was found that the majority of the students in the experimental group adopted an optimistic view of the constructivist approach. Responses to item 8 in the questionnaire revealed that 63.1% of the respondents were satisfied with their constructivist vocabulary lessons. This could be because they believed that constructivism helped them understand lexical items deeply (68.5%), memorize these words longer (63.2%), feel more active (93.1%), be able to interact with other members (73.6%), and have the opportunity to contribute to the lessons (81.6%). These results, again, are in accordance with Du's study (2013). Most of the participants in Du's study (2013) held favorable opinions about learning vocabulary in the constructivist approach. Additionally, the survey in the work by Aravind and Bhuvaneswari (2023) also confirms that most respondents highly valued the role of constructivism in improving their English vocabulary.
The positive attitude of the students in the experimental group could be well explained by constructivism itself which places the learner at the center, not the teacher. Also, constructivism is a learning theory (Richards et al., 2003), which was probably why it was more suitable for learners than the teacher-centered model, a teaching method. Further, the participants not only had a chance to present their ideas but also gave feedback on their friends' ideas, which might make them feel that they created some value in the classroom. These factors possibly shed light on why learners fostered a positive outlook on constructivist vocabulary learning.
Although the obtained results in the present study and previous research were quite similar, the novelty of this current study lies in the robust experimental design strictly following the constructivist criteria (Baviskar et al., 2009;Gagnon & Collay, 2001). As a consequence, its outcomes were of greater reliability and validity.

Conclusion
The aims of this study were to investigate the effectiveness of vocabulary learning in the constructivist approach over that in the teacher-centered model and to explore learners' attitudes towards the implementation of constructivism in their vocabulary lessons. Via the quantitative analyses of an experiment (following the pretest and posttest design) and a survey with the participation of 75 EFL undergraduates in Vietnam, the results showed that learners remembered lexical items more effectively in the constructivist approach. Moreover, it was also found that they had a positive view of this learning method, mainly due to greater comprehension and memorization of the target words as well as the interaction with friends and the contribution they could make to the lessons.
Promising as the present study might be, two limitations need to be mentioned. First, due to time restrictions, no delayed posttests were conducted to examine learners' lexical attrition in both groups. Therefore, it is recommended that future research could conduct another test, about one month after the immediate posttest, to examine whether lexical words could be well retained after such a long time. Second, the sample size was still small, and no qualitative approach was employed to explore participants' perceptions deeply. As a result, other researchers need to include a larger sample size as well as conduct interviews with the participants to gain further insights into the implementation of constructivism in vocabulary lessons.
Despite some limitations, the current study entails important theoretical and practical implications. Theoretically, it lends support to the constructivist theory by proving the positive role of this learning method in actively involving students in the lessons and promoting their lexical retention. Moreover, it shows that constructivism correlates well with other theories (i.e., distributed learning, involvement load hypothesis, testing effect, and connectionism). From the pedagogical perspective, it is crucial that constructivism should be considered and incorporated into the classroom for the learning and teaching of vocabulary. This is due to the positive outcomes this approach brought to EFL learners regarding lexical retention (Aravind & Bhuvaneswari, 2023;Du, 2013;Lin, 2015). Additionally, it is salient that teachers adhere to the procedure and elements of constructivism as proposed by Baviskar et al. (2009) or Gagnon and Collay (2001). Only via the appropriate implementation of this method will students be able to retain lexical knowledge in an effective way. This is because without the activation of prior knowledge or cognitive dissonance, new materials might not be assimilated (Baviskar et al., 2009).