Analysis of the Value of Japanese Language Education and Expansion Path of Foreign Language Learning at the University Level under the Foreign Language Policy of the New College Entrance Examination: An Investigation Based on LISREL Structural Equation Modeling

Since China’s reform and opening up, it has begun to draw on various advanced experiences from abroad. Exchanges with countries around the world have become more and more frequent, and Japan has slowly become one of China’s larger economic partners. Some aspirants who returned from studying in Japan, based on the accumulation of their study and life in Japan, not only transferred Japanese culture to China, but also introduced Japanese education and advanced technology into China [1-4], and Japanese language education was also carried out in China. Under the new gaokao foreign language policy, China gradually reduces the traditional emphasis on foreign languages, which is mainly manifested in a more relaxed state, in which students are more free to learn foreign languages, and more emphasis is placed on the practicality of foreign languages rather than test scores [5-8].

Foreign language learning is a dream and a challenge for many students. Whether it is for the purpose of further education, work or personal interest, mastering a foreign language is very beneficial, especially for college students, the results of their foreign language learning is of great significance to their growth and future development [9-12], but many students encounter various difficulties and frustrations in the process of learning foreign languages, and expanding the path of foreign language learning has become an important way to learn foreign languages [13-14]. During the university period, students’ personal time is more sufficient, for students learning foreign languages have more time and energy to put on foreign language learning [15-16], generally expanding the path of foreign language learning in addition to listening, speaking, reading and writing to realize, but also through the development of learning plans, create a foreign language learning atmosphere and other ways to achieve, which not only plays an important role in the learning of Japanese language, but also for English, Russian, French and other foreign languages have obvious effects [17-20].

This paper outlines the evolution of the foreign language policy of the new college entrance examination based on relevant information, including the multilingual program and the provision of two examination opportunities for foreign language subjects.Through the intercultural knowledge test paper and regression analysis, the value of Japanese language education in cultivating intercultural awareness and its economic value are explored.The LISREL structural equation model was selected, and the advantages of the model and the modeling process were outlined. The components of the model were analyzed, and family factors, learning environment, cultural interest and foreign language learning expansion were selected as relevant variables to construct an exploratory model of foreign language learning expansion path at university level. Substituting the actual data into the model results in arithmetic results that correspond to the fit of the model.

2

Foreign Language Policy for the New Higher Education Entrance Examination

2.1

“Multilingualism Plus” - a cultural powerhouse

Foreign language education planning is a special field of language planning, which mainly decides what kind of language is introduced into the language ecosystem of the country, what scale it reaches, who uses it under what circumstances, how it is used, how to acquire foreign language competence, etc. In terms of the type of planning, it is mainly a matter of status and image planning, which has a lesser impact on ontological planning of foreign languages. Since the beginning of this century, many countries in the world have implemented foreign language capacity building as a national strategic plan. Foreign language education in the United States is mainly Spanish and French, but after the “9/11” incident, the “Critical Language Proficiency” program was launched, and five languages, including Japanese and Arabic, were listed as key language programs related to national security, with obvious political and ideological factors. China launched the “Belt and Road” initiative in 2015, in order to achieve language paving and people-to-people communication with more than 60 countries along the route, the country’s major universities implement the “multilingual +” foreign language major planning, such as Peking University offers 40 minor language and cultural courses, including 32 languages in countries along the Belt and Road. The number of languages currently taught at Shanghai International Studies University (SISU) has reached 46. Beijing Foreign Studies University (BFSU) plans to offer 100 foreign languages, covering the official languages of all countries along the “Belt and Road”. Despite the rapid development of multilingual education in recent years, there is still a gap in the number of languages compared with the United States, and it is still far from being able to meet the requirements of a cultural power.

2.2

Language Examination Reform Policy

The examination and enrollment system is the basic national education system, and language examination policy is one of the important elements of language policy, especially language education policy. Since the resumption of the college entrance examination, the State has made a number of adjustments to the foreign language examination policy (mainly the foreign language examination policy), such as the ever-changing weighting of the scores of the foreign language subject in the college entrance examination, which has gone through a period of seven years, from the resumption of the college entrance examination to the fact that the scores are counted in the total scores of the college entrance examination according to 100 per cent. During this period, the government issued a number of language policy-related documents, such as the early documents issued by the State Council, proposing that the proportion of points awarded to foreign language subjects will be increased year by year. At this stage, it is clearly proposed to deepen the reform of the examination and enrollment system, including “enhancing the relevance of the college entrance examination to senior high school learning …… and providing two examination opportunities for foreign language subjects ……” and so on. Under the influence of national macro-policies, provinces and cities have also introduced more detailed language examination reform policies.

2.3

Educational value of the Japanese language

2.3.1

Value of Intercultural Awareness Cultivation

After the implementation of the foreign language policy of the new college entrance examination, the current status of intercultural awareness development among Y high school Japanese learners was investigated by means of an intercultural knowledge test paper.

The frequency histogram of the total score, the frequency histogram of the level of intercultural knowledge in Japanese and the normal distribution curve are shown in Figure 1 using the statistical software SPSS29.0. By observing the process of distributing the intercultural knowledge test paper, the researcher found that Japanese learners were interested in some parts of the questionnaire, and that students answered the questions slowly and took longer to answer the questions than the researcher expected. By counting the valid questionnaires, the researcher used Excel and SPSS29.0 to analyze the questionnaires descriptively and with frequency statistics. Its findings are as follows: 400 questionnaires were distributed and 399 valid questionnaires were recovered, of which the highest score was 100, the lowest score was 0, and the mean score was 59.57, which is less than 60, with a standard deviation of 14.595. The distribution of scores was further examined. 0-60 was 13.25%, 65-80 was 22.75%, and 85-100 was 64%. This shows that the students have a high level of intercultural knowledge.

The statistics of students’ mastery of each topic in the intercultural knowledge test paper are shown in Figure 2. Questions 1~5: Behavioral interactions. Questions 6~9: Clothing, food, housing and transportation. Questions 10~13: Festival celebrations. Questions 14~17: Cultural practices. Questions 18~21: country profile. The study participants’ years of Japanese language study were all 3 years or more, and the longest period of Japanese language study was 6 years, indicating that the study participants had a certain foundation of Japanese language study and knowledge accumulation, and were not new to the Japanese language. According to the data, in terms of behavioral interactions, the number of correct answers to all questions except question 5 was above 50%.Question 5 examined the differences between Chinese and foreign communication, and the percentage of correct answers was only 40.3%, which indicates that students do not know enough about the differences between Chinese and foreign cultures in communication.In clothing, food, housing, and transportation, more than 50% of the questions were correct.Clothing, food, housing, and transportation are closely related to our daily lives, and students have a good grasp of the cultural knowledge of clothing, food, housing, and transportation.Festivals and celebrations are mainly examined in major Chinese and Japanese festivals. Question 10 examines the understanding of the customs of the Lantern Festival, a traditional Chinese festival, and the number of correct answers to the rest of the questions is over 50%, which shows that students have a certain degree of understanding of the major festivals in Japan. Students’ mastery of cultural practices was the worst. Questions 14-17 were answered correctly by 25.1%, 43.9%, 35.7%, and 34.2% respectively, all below 50%.The questions on cultural practices with the poorest mastery mainly examined the cultural differences between China and Japan regarding dragons, the meanings of numbers, slang, and the cultural meanings of colors in other countries. The same words have very different meanings when placed in different cultural contexts. It is important for students to understand the meanings of phrases in different contexts in order to achieve true intercultural communication. In the country profile, the number of all correct answers is above 50%, which shows that students have a certain knowledge of the country profile of Japan and have developed a cross-cultural awareness.

2.3.2

Economic value of Japanese language education

Multiple regression modeling is a commonly used method in statistics to study the relationship between one or more independent variables and a dependent variable. Unlike a simple linear regression model that involves only one independent variable, a multiple regression model can consider the effects of multiple independent variables at the same time After analyzing the basic descriptive statistics of the annual salary and income levels in the Chinese labor market, we first conduct a regression analysis of the labor market starting annual salary returns for the Japanese-language proficiency of Chinese college graduates. In this study, the starting annual salary is the actual pre-tax annual salary in the first year of work after graduation, and the logarithm of the starting annual salary is used as the dependent variable in the regression model, while the independent variable is Japanese language ability, which is subdivided into listening and speaking ability and reading and writing ability in this study, and regressed to see their impact on the starting annual salary level, and the regression model does not involve the variables related to years of work or experience, as the starting annual salary is not affected by years of work. Since starting year salary is not affected by years of work experience, no variables related to years of work experience or experience are included in this regression model.

In terms of the return on starting salary for Japanese listening and speaking ability, only basic demographic indicators and Japanese listening and speaking ability are included in model (1), which shows that Japanese listening and speaking ability has a significant effect on starting salary at the 0.01 statistical level, and that each point of improvement in listening and speaking ability increases starting salary by 0.8 percent. After that, we gradually added variables such as type of undergraduate college, work region, type of work unit, years of parental education, and highest education to supplement the regression model, and the value of R-squared was gradually increased after the adjustment, and the explanatory strength of the overall model was gradually increased from 10% to 32%, and finally model (4) was obtained. The regression results are shown in Table 1. In terms of work geography, there is no significant difference in the Northeast compared with the West, and the annual salary level in the Central region is 15.6% higher than that in the West, while that in the East is 27.7% higher. The annual salary level is significantly influenced by the number of years of parental education, which is significant when compared to family background and ability, with each additional year of parental education increasing it by 27.4%.

Table 1.

Japanese ability to pay for the Labour market

	(1)	(2)	(3)	(4)
Gender male	0.212*** [0.082]	0.132* [0.082]	0.140** [0.070]	0.144* [0.065]
Japanese ability	0.009*** [0.003]	0.007*** [0.004]	0.005** [0.004]	0.006* [0.004]
Working field Central region		0.240*** [0.083]	0.202** [0.083]	0.156* [0.078]
Eastern region		0.370*** [0.094]	0.311*** [0.097]	0.277*** [0.099]
Northeast region		-0.098 [0.145]	-0.043 [0.135]	0.045 [0.140]
Job type		0.140 [0.122]	0.211* [0.125]	0.229* [0.120]
Parents are subject to education		-0.412** [0.204]	-0.328* [0.185]	-0.274* [0.154]
The highest degree (benchmark)		-0.253*** [0.072]	-0.257*** [0.072]	-0.193*** [0.074]
Constant term			0.036*** [0.014]	0.032** [0.013]
R2	0.112	0.295	0.313	0.350
Adj-R2	0.102	0.237	0.294	0.322
N	200	200	200	200

3

Exploration of Foreign Language Learning Expansion Paths Based on the LISREL Model

3.1

Superiority of the LISREL model

LISREL stands for Linear Structural Relationship, and the LISREL model is often used to analyze the relationship between a series of unknown coefficients, including the relationship between latent and measured variables, which can improve the accuracy of the research on the path of foreign language learning expansion. This study constructs the proposed model to cover four latent variables, which are family factors, cultural interest, learning environment, and foreign language learning.

1)

Latent variables can be effectively explored

In the process of social science research, the concepts of many indicators are vague, and the use of a single indicator measurement is prone to conceptual bias. The LISREL model is a superior method for enhancing the study’s accuracy.On the flip side, it also addresses the issue of multicollinearity of independent variables in multiple regression analysis (MRA). When the dependent variable is affected by a wide range and there is interdependence between independent variables, the results obtained using multiple regression analysis may be unexplained, and the results obtained may not be consistent with common sense and other problems. The LISREL model first extracts interdependent and highly relevant indexes to form a relatively independent factor in the analysis, which makes up for the shortcomings of multiple regression analysis.

2)

Easy to analyze the causal relationship

Although path analysis can also calculate direct and indirect effects, it can only be used to calculate the effect between significant variables. Once potential variables are involved, it cannot be used. And it is limited in this role, the requirements for information are more stringent, the use of many assumptions, which involves only one-way arrows in the path analysis system, that is, it is not possible to use the data to determine the causal relationship of variables. Of course, the determination of causality is mainly affirmed through the knowledge of this specialty, only the LISREL model enhances the researcher’s thinking and understanding of the deeper issues of causality, and plays a good role in helping.

3)

Model fitting is more reasonable

Due to the inclusion of systematic expertise in the design process, it breaks through the traditional limitation of relying too much on the mathematical and scientific perspectives to find the relationship between the variables, thus making the fitting of the LISREL model, which can be a process of cross-corroboration, or a process of theoretical model confirmation.

3.2

Construction of the LISREL model

Structural equation modeling is the use of collected data to predict the parameters in the model and how well the entire model fits the actual data. In the process, the model is continually adjusted until it fits the observed data. Currently, there are only two major techniques available to construct structural equation models. One is the covariance structural analysis method of maximum likelihood estimation, often referred to as “hard modeling”, such as LISREL, and the other is the partial least squares analysis of variance method, referred to as “soft modeling”, such as PLS.

3.2.1

Overview of the LISREL model

From the previous analysis, it can be seen that structural equation modeling can be divided into the measurement model and the structural model. The LISREL model is also composed of two parts: the structural equation model and the measurement equation model.

The structural equation modeling describes the relationship between the hidden variables (LV) in the system as η = Bη + Γξ + ξ where η = (η₁,η₂,⋯,η_m) denotes the vector of endogenous hidden variables, ξ = (ξ₁,ξ₂,⋯,ξ_n) denotes the vector of exogenous hidden variables, ζ denotes the vector of residuals, and B(m×n) and Γ(m×n) denote the matrices of path coefficients for η and ξ . Each factor (η₁.η₂,⋯,η_m) in η corresponds to its endogenous vector and each factor ξ = (ξ₁,ξ₂,⋯,ξ_n) in ξ corresponds to its exogenous vector. The hidden variables are interconnected through linear equations with the two path coefficient matrices B(m×n) and Γ(m×n) and the residual vector ξ . Where B(m×m) the path coefficient matrix describes the effect of the exogenous vector on the endogenous vector, Γ(m×n) the coefficient matrix describes the effect between the endogenous vectors, and ξ is used to describe the other factors in the model.

The measurement equation model describes the relationship between the manifest variable (MV) and the hidden variable (LV) with the equation: (1) $\begin{array}{l} y = Λ_{y} η + ε_{y} \\ x = Λ_{x} ξ + ε_{x} \end{array}$ where y = (y₁,y₂,⋯,y_p) and x = (x₁,x₂,⋯,x_p) are the explicit variables for η & ξ respectively, Λ_γ(p×m) and Λ_x(q×n) denote the loading matrix, and ε_y and ε_x denote the residuals. The hidden variables are defined by the explicit variables through measurement equation modeling. Connect η to y variables and ξ to x variables. Explicit variables y and x are connected to the hidden variables η and ξ phases through linear equations with coefficients λ_γ and λ_x and measurement error terms ε_γ and ε_x, respectively.

Structural equation modeling and measurement equation modeling in LISREL require the following conditions to be met.

Assumptions in the structural equation modeling: (2) $E (ζ) = E (ζ ζ') = 0$

Assumptions in the measurement equation modeling: (3) $E [ε_{y} ε_{x}] = E [η ε_{y}^{'}] = [η ε_{x}^{'}] = 0$ (4) $E [ε_{y}] = E [ε_{x}] = E [η ε_{x}^{'}] = [ξ ε_{y}^{'}]$ (5) $E [ζ ε_{y}] = E [ζ ε_{x}] = 0$

Standardization: (6) $\begin{array}{l} E [η] = [ξ] = 0 \\ E [x] = y = 0 \\ V a r (x) = V a r (y) = 1 \end{array}$ where: ξ is the exogenous hidden variable, n is the endogenous hidden variable, X is the explicit variable of ξ, Y is the explicit variable of n, λ is the coefficient between the explicit variable and the hidden variable, 8 is the covariance matrix between Φξ and ξ, Θ is the covariance matrix between δ, Λx describes the relationship matrix between X and ξ, Ay describes the relationship matrix between Y and η, and δ describes the measurement error of X. ε is the observation error of Y, Θε is the covariance matrix between ε, ζ is the potential disturbance, Ψζ is the covariance matrix between ζ , β is the regression coefficient between the endogenous hidden variables, B is the matrix of regression coefficients between the endogenous hidden variables, γ is the regression coefficients between the exogenous and the endogenous hidden variables, and Γ is the matrix of regression coefficients between the exogenous and the endogenous hidden variables.

It is expressed by the regression equation as: (7) $\begin{array}{l} η_{1} = γ_{11} ζ_{1} + γ_{12} ζ_{2} + ζ_{1} \\ η_{2} = β_{21} η_{1} + γ_{21} ζ_{1} + ζ_{2} \end{array}$ (8) $\begin{array}{l} x_{1} = λ_{11} ζ_{1} + δ_{1} \\ x_{2} = λ_{21} ζ_{1} + δ_{2} \\ x_{3} = λ_{31} ζ_{1} + δ_{3} \\ x_{4} = λ_{42} ζ_{2} + δ_{4} \\ x_{5} = λ_{52} ζ_{2} + δ_{5} \\ x_{6} = λ_{62} ζ_{2} + δ_{6} \end{array}$ (9) $\begin{array}{l} y_{1} = λ_{11} η_{1} + δ_{1} \\ y_{2} = λ_{21} η_{1} + δ_{2} \\ y_{3} = λ_{31} η_{1} + δ_{3} \\ y_{4} = λ_{42} η_{2} + δ_{4} \\ y_{5} = λ_{52} η_{2} + δ_{5} \\ y_{6} = λ_{62} η_{2} + δ_{6} \end{array}$

The matrix equation is expressed as: (10) $x = Λ_{χ} ζ + δ$ (11) $y = Λ_{y} η + δ$ (12) $η = B η + Γ ζ + ζ$ $$\eta = {\rm{B}}\>\eta + \Gamma \>\zeta + \zeta $$

The vector form is: (13) $[\begin{matrix} x_{1} \\ x_{2} \\ x_{3} \\ x_{4} \\ x_{5} \\ x_{6} \end{matrix}] = [\begin{matrix} λ_{11} & 0 \\ λ_{21} & 0 \\ λ_{31} & 0 \\ 0 & λ_{42} \\ 0 & λ_{52} \\ 0 & λ_{62} \end{matrix}] [\begin{matrix} ζ_{1} \\ ζ_{2} \end{matrix}] + [\begin{matrix} δ_{1} \\ δ_{2} \\ δ_{3} \\ δ_{4} \\ δ_{5} \\ δ_{6} \end{matrix}]$

The vector form is: (14) $[\begin{matrix} y_{1} \\ y_{2} \\ y_{3} \\ y_{4} \\ y_{5} \\ y_{6} \end{matrix}] = [\begin{matrix} λ_{11} & 0 \\ λ_{21} & 0 \\ λ_{31} & 0 \\ 0 & λ_{42} \\ 0 & λ_{52} \\ 0 & λ_{62} \end{matrix}] [\begin{matrix} η_{1} \\ η_{2} \end{matrix}] + [\begin{matrix} δ_{1} \\ δ_{2} \\ δ_{3} \\ δ_{4} \\ δ_{5} \\ δ_{6} \end{matrix}]$ (15) $[\begin{matrix} η_{1} \\ η_{2} \end{matrix}] = [\begin{matrix} 0 & 0 \\ β_{21} & 0 \end{matrix}] [\begin{matrix} η_{1} \\ η_{2} \end{matrix}] + [\begin{matrix} γ_{11} & γ_{12} \\ γ_{21} & 0 \end{matrix}] [\begin{matrix} ζ_{1} \\ ζ_{2} \end{matrix}] + [\begin{matrix} ζ_{1} \\ ζ_{2} \end{matrix}]$

3.2.2

Methods of estimating LISREL parameters

LISREL model diagram rules: ellipses indicate exogenous and endogenous hidden variables, and rectangles indicate explicit variables corresponding to exogenous and endogenous hidden variables. Unidirectional arrows indicate causal relationships between different variables, while bidirectional arrows indicate correlation relationships between different vectors. Labeling rules for coefficients: causality is the effect of the beginning of a single arrow on the end of a single arrow, e.g., γ₁₂ indicates the effect of ξ₁₂ on η_t. The correlation variable refers to the correlation among the variables at the ends of the double arrows. The single arrow itself represents an equation with the variables at the beginning of the arrow at the left end of the equation and the algebraic sum of the variables at the end of the arrow at the right end of the equation.

The object of LISREL evaluation is to apply fitting to achieve parameter estimation when the difference between the explicit variable sample covariance matrix S and the total covariance matrix Σ is minimized, and there exists θ that satisfies $F = F (S, \sum^{Λ})$ minimization. When estimating using the LISREL method, let Σ and S be P -dimensional vectors, and when the number of selected samples n→∞, then the explicit variable sample covariance matrix S converges to the total covariance matrix Σ. Assuming that the values of $\sqrt{n} S$ are finite and asymptotically obeying the normal distribution, the covariance matrix Γ(p×p) is also asymptotically asymptotically. The expression for θ is Σ = Σ(θ). The estimated values of each parameter value are substituted into the equation to obtain $\sum^{Λ} = \sum (\hat{θ})$ . Minimization consists mainly of: selection of initial values, iterative process and stopping criteria.

3.2.3

Tests and corrections of the LISREL model

Whether the model construction is valid or not needs to be further tested, the main test indicators for LISREL modeling are:

The chi-square value test whose formula is: (16) $x^{2} = (N - 1) \times F$ where F is the minimum value of the fitted function.

The degree of freedom of the chi-square value is df = (p + q)(p + q + 1)/2–t, p + q denotes the number of measurable variables and t denotes the minimum number of parameters that must be estimated for all paths.

Goodness-of-fit index GFI

The degree of total fit can be evaluated by applying the Goodness of Fit Index GFI: (17) $G F I = 1 - 1 - F_{\min} / F_{0}$

The denominator and numerator represent the pre-fit and post-fit objective function values, respectively.

The total fit can be evaluated by applying the positive goodness-of-fit index (AGFI) with the expression: (18) $A G F I = 1 - [(k / d f) (1 - G F I)]$

In the formula: (19) $K = 1 / 2 (p + q) (p + q + 1)$

LISREL is a systems theory approach that combines mathematical and statistical methods with grounded theory. The use of LISREL allows for a comprehensive analysis of the system. When the model fails to fit the data, the model needs to be revised. The purpose of correcting the model is to find the defects of the initial model and achieve the purpose of modifying the model. When correcting the model, it is necessary to consider not only the statistical indicators, but also the theoretical basis of its model. If only from the mathematical perspective, the theoretical support is of low validity and lacks practical guidance.

3.3

Exploration of Foreign Language Learning Expansion Paths

In this study, 450 full-time undergraduate students majoring in English in the first to third years of a comprehensive university were used as subjects. In the preset model, there is a positive path between family factors and learning environment and cultural interest, i.e., it is assumed that the family’s recognition and acceptance of the target language will create a favorable learning atmosphere for the learners, which will in turn increase their interest in learning the target language and culture. This hypothesis is based on the results of foreign experiments, which indicate a positive correlation between family factors and learning motivation.

The study consisted of two phases. The first phase consisted of interviews and open-ended written feedback to observe and summarize the subjects’ descriptive personal information and their initial foreign language learning situation. The second stage is a self-administered questionnaire, which adopts a Likert scale, and the questions are set with reference to the relevant experimental questionnaires in foreign countries, and some of the questions are added, deleted and amended with the results of the qualitative analysis in the first stage, in an attempt to achieve the comprehensiveness of the questionnaire content and the scientificity of the questionnaire setting. Finally, 24 measurable indicators were identified, as shown in Table 2. Before the formal experiment, questionnaire prediction was conducted, and the overall reliability values of the questionnaire (Cronbach’s α) were all over 0.811. The feedback results of the first stage were analyzed qualitatively by the content analysis method, and the quantitative questionnaire data of the second stage were analyzed by descriptive statistics using SPSS21.0, and the structural equations of the LISREL model were validated and corrected using Amos21.0. Model validation and correction.

Table 2.

Measurable Settings and internal reliability values

Variable	Cronbach’s α	Mean	Standard deviation
Foreign language learning development	0.942	7.093	2.301
Learning environment	0.905	8.054	2.035
Cultural interest	0.879	6.184	2.140
Family factor	0.811	6.284	2.063

The final structural model and standardized path coefficients are shown in Figure 3. The following conclusions can be drawn from the completeness of the final structural model and the interaction of the latent variables:

1) The results showed that family factors had a significant impact on the learning environment and foreign language learning (P<0.001), and the results showed that the influence of family factors on foreign language learning (0.92) was much greater than the impact on the learning environment (0.59), indicating that various related factors in Chinese families, such as parental expectations, social status and economic conditions, had a more significant impact on the intrinsic motivation of foreign language learning than the external environment.

2) The foreign language learning paths of Chinese college foreign language learners are closely related to other extraneous variables, with high similarity to previous studies. At the same time, Chinese learners’ paths of expanding their foreign language learning also show a number of dynamic characteristics, which are worth exploring in depth. In the Chinese educational context, parental encouragement and guidance are the catalysts for converting positive factors in the external environment into internal instrumental motivation, which is also the basis for the successful construction of foreign language learning paths. Thus, we should not limit ourselves to the study of the learning environment at the cognitive context stage, but should also start from the influence of learners’ psychology and personality, and explore the important role of the family as the primary unit of society in the construction of learners’ foreign language value identity and personality.

In addition, it is worth noting that the questionnaire data in this experiment show that the mean of the relevant sub-dimensions of “learning environment” is generally abnormally high, such as “teacher teaching style”, “peer relationship”, “foreign language classroom atmosphere”, “teaching equipment” and other hardware facilities and human factors in the learning environment, as well as “successful learning experience” and “learning interest” and other learning experience factors. The above mean values show that the participants attach great importance to the foreign language learning environment and the resulting learning experience, and turn the positive factors into the driving force for the development of foreign language learning.

4

Conclusion

This paper focuses on analyzing the value of Japanese language education through intercultural knowledge test papers and regression analysis. LISREL structural equation modeling was established to obtain relevant data in the form of questionnaires, which were then substituted into the model to analyze the path relationship between family factors, learning environment, cultural interest and learning a foreign language. The study shows that:

1) The New College Entrance Examination policy of “Multilingualism Plus” has enabled students to gain knowledge of the country of Japan and develop cross-cultural awareness. Japanese listening and speaking skills have an impact on the starting annual salary of graduates at a 0.01 level of significance.

2) Family factors, learning environment, and cultural interests can significantly affect foreign language learning. Therefore, reasonable improvement of family factors, learning environment, and cultural interests can effectively expand the path of foreign language learning.

Funding:

This research was supported by the Jiangxi Provincial Education Science “14th Five Year Plan” Project 2023 “Difficulties and Breakthroughs in Japanese Language Learning for College Students under the New College Entrance Examination Foreign Language Policy” (Project Number: 23YB316).

Langue:: Anglais

Périodicité:: 1 fois par an
Sujets de la revue:: Sciences de la vie, Sciences de la vie, autres, Mathématiques, Mathématiques appliquées, Mathématiques générales, Physique, Physique, autres

RSS Feed de la revue

Analysis of the Value of Japanese Language Education and Expansion Path of Foreign Language Learning at the University Level under the Foreign Language Policy of the New College Entrance Examination: An Investigation Based on LISREL Structural Equation Modeling

Hongju Yan

Publié en ligne: 17 mars 2025

Reçu: 22 oct. 2024

Accepté: 06 févr. 2025

DOI: https://doi.org/10.2478/amns-2025-0308

Mots clésRegression analysis, Intercultural knowledge, LISREL structural equation, Japanese language education

© 2025 Hongju Yan, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Mots clés
Regression analysis, Intercultural knowledge, LISREL structural equation, Japanese language education