The Impact of Students’ Learning Strategies and Teachers’ Strategy Instructions on English Literacy-Driven Achievement
Author: Chao-Yun Lee (Master Program of Professional Teacher, National Taichung University of Education), Jin-Chang Hsieh (Research Center for Testing and Assessment, National Academy for Educational Research)
Vol.&No.：Vol. 71, No. 2
Date：June 2026
Pages：35-71
DOI：https://doi.org/10.6209/JORIES.202606_71(2).0002
Abstract：
Purpose
The research is to explore the influence of learning strategies and teacher’s strategy instructions on English literacy-driven achievement in the context of 12-year basic education. The study has two goals. The first goal is to investigate the relationship between English literacy-driven achievement and learning strategies including memory, cognition, inference and metacognition strategies. The second goal is to test whether teacher’s strategy instructions affect the means of English achievement in the class and adjust the impact of learning strategies on English achievement.
Literature Review
The researches have indicated that learning strategies affect students’ English grades. These studies use Pearson correlation and description statistics to demonstrate the relationship between grades and learning strategies. Additionally, the studies reveal that teachers’ strategy instructions influence on English achievement. A semi-structure interview is applied to explore the impact of teachers’ strategy instructions on English grades. However, no research has investigated how learning strategies and teachers’ strategy instructions affect achievement within the same framework. Furthermore, contextual effects have not been considered when examining the relationship between learning strategies, teachers’ strategy instructions and achievement.
Method
1. Participants
The participants are the seventh-grade students who were the first cohort of 12-year basic education in Taiwan in 2020. There are 2264 students including 1184 males and 1080 females. Additionally, there are 219 teachers from 216 schools.
2. Variable
In this research, the dependent variable is student’s English literacy-driven achievement. The independent variables include student’s learning strategies and teacher’s strategy instructions. Student’s learning strategies consist of four variables: memory, cognition, inference and metacognition strategy. Similarly, the teacher’s strategy instruction consists of memory, cognition, inference and metacognition strategy instruction.
3. Instrument
(1) Memory strategy: The scale contains six items. The factor loading ranges from .77 to .85. The composite reliability is .92.
(2) Cognition strategy: The scale contains six items. The factor loading ranges from .67 to .80. The composite reliability is .89.
(3) Inference strategy: The scale contains eight items. The factor loading ranges from .71 to .83. The composite reliability is .93.
(4) Metacognition strategy: The scale contains ten items. The factor loading ranges from .69 to .81. The composite reliability is .93.
(5) English Literacy-Driven Achievement: The assessment includes 93 reading comprehension question and 89 listening questions.
(6) Memory strategy instruction: The scale contains two items. The factor loading ranges from .64 to .81. The reliability is .76.
(7) Cognition strategy instruction: The scale contains three items. The factor loading ranges from .57 to .78. The reliability is .69.
(8) Inference strategy instruction: The scale contains three items. The factor loading ranges from .63 to .82. The reliability is .76.
(9) Metacognition strategy instruction: The scale contains three items. The factor loading ranges from .68 to .78. The reliability is .76.
4. Data Analysis
In this study, HLM is applied. To use HLM, data is weighted and centered. Additionally, five models are evaluated.
Results
1. Descriptive Statistics
Pearson correlation between achievement and learning strategies is .29, .30, .42, and .35 for memory cognition, inference and metacognition strategy respectively. All of correlations are significant. Additionally, the correlation between achievement and learning strategy instructions is .02, .06, .07, .06 for memory cognition, inference and metacognition strategy instruction respectively. However, cognition, inference, and metacognition strategy instruction show significant correlations, while only memory strategy instruction is not significant.
2. HLM Analysis
(1) Model 1: Intercept-Only Model
The result shows that interclass correlation is .29 which is higher .05. Additionally, reliability of the mean is .8 which is higher .7. Therefore, the further analysis should be applied.
English Literacy-Driven achievement = β0j + γij
β0j = γ00 + μ0j
(2) Model 2: Means-As-Outcomes Model
The analysis reveals that the coefficient of memory, cognition, and metacognition strategy instruction are not significant. However, the coefficient of inference strategy instruction is significant (γ03 = 5.81). In addition, the variance of random variable (μ0j) is significant indicating that next step is needed.
English Literacy-Driven achievement = β0j + γij
β0j = γ00 + γ01memory strategy instruction + γ02cognition strategy instruction + γ03inference strategy instruction + γ04metacognition strategy instruction + μ0j
(3) Model 3: Incorporating Individual-Level Explanatory Variables While Treating The Slope As A Fixed Coefficient
The result shows that coefficients of inference (γ30) and metacognition (γ40) strategy are significant (γ30 = 16.32; γ40 = 5.10). However, at the intercept of second level, inference strategy instruction (γ03) is significant (γ03 = 5.83).
English Literacy-Driven achievement = β0j + β1j memory strategy + β2j cognition strategy +
β3j inference strategy + β4jmetacognition strategy + γij
β0j= γ00 + γ01memory strategy instruction + γ02cognition strategy instruction + γ03inference strategy instruction + γ04metacognition strategy instruction + μ0j
β1j = γ10
β2j = γ20
β3j = γ30
β4j = γ40
(4) Model 4: Random Coefficient Model
The results of the random coefficient model are very similar to model 3. The coefficients of inference (γ30) and metacognition (γ40) strategy are significant (γ30 = 17.11; γ40 = 4.60). Moreover, inference strategy instruction (γ03) is also significant (γ03 = 6.36). However, the variance of μ1j, μ2j, μ3j, and μ4j are not significant.
English Literacy-Driven achievement = β0j + β1j memory strategy + β2j cognition strategy +
β3j inference strategy + β4jmetacognition strategy + γij
β0j= γ00 + γ01memory strategy instruction + γ02cognition strategy instruction + γ03inference strategy instruction + γ04metacognition strategy instruction+μ0j
β1j = γ10+μ1j
β2j = γ20+μ2j
β3j = γ30+μ3j
β4j = γ40+μ4j
(5)Model 5: Slopes-as-Outcomes Model (Full Model). In this model, teachers’ strategy instructions are added to predict the slopes of students’ learning strategies. The results show that the cross-level interaction terms (γ₁₁-γ₄₁) are not significant, indicating that teachers’ strategy instructions do not moderate the effects of students’ learning strategies on English literacy-driven achievement. In addition, only the variance of the intercept (μ₀j) remains significant. Therefore, Model 4 is selected as the final model.
Although D2 decreases as model complexity increases, the variances of the random slopes (μ₀j-μ4j) in the random coefficient model are not significant. Therefore, the random coefficient model (Model 4), rather than the more complex slopes-as-outcomes model, is chosen as the final model.
Conclusion
The study highlights the importance of learning strategy and strategy instruction in the context of 12-year basic education. Students’ inference and metacognition strategies have a significant impact on English Literacy-Driven achievement, while memory and cognition strategies do not. Regarding teacher’ strategy instruction, inference strategy instruction has positive effect on the mean of English achievement in the class. In contrast, memory, cognition, and metacognition strategy instruction do not have impacts on English achievement in the class. Furthermore, regardless of which strategies teacher choose, the coefficients of students’ learning strategies on English achievement are not moderated.
Through analysis of HLM, the role of teachers, students, and learning achievements are depicted more clearly compared to previous literature. Based on the result, the following suggestions are proposed. First, students should improve their inference and metacognition strategy. Second, teachers should guide students in developing inference strategy to enhance overall English achievement in the class.
Besides, other topics and issues warrant further investigation. In the study, only junior high school students were included. However, to gain the deeper understanding of learning strategies, further research should also include elementary school students to examine the effects of learning strategies from an early stage. Furthermore, this study focused solely on listening and reading. However, English proficiency encompasses listening, reading, writing and speaking. Therefore, further research should consider all four skills.
Keywords：
metacognition strategy, English academic achievement, memory strategy, inference strategy, cognition strategy