The Innovation of Traditional Sports Culture Dissemination Mode of Ethnic Minorities in the Age of Internet Plus

Ethnic traditional sports is an important part of human sports culture, which is both an expression of cultural forms with national characteristics and a cultural form with quite traditional colors [1-2]. It is both a component of human sports culture and an important content of national traditional history and culture [3]. Ethnic traditional sports are characterized by strong inheritance, ethnicity, creativity, locality and fitness, thus, in today’s ethnic sports designation, ethnic traditional sports are often equated with national sports [4-5].

Chinese national traditional sports culture has a long history, in the process of the construction of a strong sports country, the inheritance and development of good national traditional sports culture can help to better meet the growing demand for physical fitness of the people in various regions, and at the same time, it is of great significance to the construction of a strong sports country and the inheritance of the excellent traditional Chinese culture [6-8]. In the process of China’s historical development for more than five thousand years, a rich variety of national traditional sports culture has been bred, and these regional characteristics of sports culture are deeply imprinted in the people’s memory [9-10].

The development of Chinese national traditional sports culture is a process of historical condensation, but also the process of vertical and horizontal dissemination, in this sense, without dissemination, national traditional sports culture and identity can not be realized. The globalization of sports also indicates the real root of the identity crisis of Chinese traditional national sports culture [11-12]. As the strategy of China’s sports power continues to promote, the state attaches greater importance to the inheritance and development of traditional national sports culture. Especially in the process of China’s sports globalization, the inheritance connotation of national traditional sports culture is the fundamental place to show the unique charm of Chinese sports culture, and plays a role in enriching the content of the world’s sports culture [13-15].

The arrival of the all-media Internet era means a change in the way of cultural communication, and the so-called all-media is the fusion of a variety of media, including traditional media, new media, network media, fusion media and so on. The integration of media is to better promote the development of cultural communication, and in the context of the all-media era, the speed, breadth, and coverage of information dissemination have been greatly improved, not only that, but also more diversified ways of obtaining information, no matter when and where people are able to obtain rich, diversified and personalized information [16-18]. In the context of the rapid development of the all-media era, exploring how to utilize the advantages of all-media to provide high-quality services for the Chinese national traditional sports culture, and how to highly integrate the media with the national traditional sports culture is a topic worth studying [19-20].

The related research on the dissemination of national sports culture is mainly divided into two themes, one is the investigative research on the current situation of national sports culture dissemination, and the other is the analytical research on the specific realization path of national sports culture dissemination strategy. One of the investigative studies is as follows, Wang, J. deeply analyzed the origin, cultural value and communication strategy of traditional Chinese national sports program lion dance, aiming to fully the cultural appeal of lion dance as a national sports program, to enhance the international influence of Chinese national culture, to promote China’s fine national culture to the world, and to show the world the face of China with wisdom and prosperity [21]. Chen, X et al. combined intelligent PLS software and a questionnaire survey to investigate the evolution of the spread of traditional ethnic sports and culture across various network connections, and explained that participation in traditional sports and culture helps to maintain human health, develop community self-awareness, and enhance the sense of national honor [22]. Yin, G et al. suggested that the development and dissemination of traditional ethnic sports are essentially consistent with the development and dissemination of multiculturalism, and therefore explored the dissemination of traditional ethnic sports wrestling in Northeast China in the context of the Belt and Road, and the study provided some novel ideas for scholars in the field [23]. Li, Z et al. Ethnic sports culture is generally disseminated through folklore activities, sports events, and modern media, and empirically explored the current situation of ethnic sports development and dissemination strategies of the Hani (Akha) people, which made a positive contribution to the dissemination and further development of ethnic sports culture [24]. As for the analytical research on the communication path of national sports culture as follows, Li, Y et al. attempted to analyze the development and communication of traditional national sports wushu from the perspective of new media, which contributes effective and scientific suggestions for the communication of digital intangible cultural heritage of traditional wushu culture, promotes the establishment of the traditional wushu inheritance and protection mechanism, and contributes to the sustainable development of the intangible cultural heritage of national sports [25]. Zhang, W. proposes to combine the ethnic sports program Wushu and education for synchronous development, which helps to introduce the good ethnic sports culture into students’ teaching and promotes the formation of correct values of students, as well as the dissemination and good development of the ethnic sports culture [26].

In this paper, the main influences on traditional sports culture communication among ethnic minorities are explored using PLS-based structural equation modeling. Based on the existing theories, the influencing factors of traditional minority sports culture communication are summarized into seven variable factors: individual audience, content ontology, communication subject, environmental object, awareness, understanding and participation, and the research hypotheses are proposed. Then the hypotheses were initially verified through reliability and correlation analysis. Then, based on the simulation fit and path coefficient results, the hypotheses are verified to be valid. Finally, based on the research conclusions and related studies, the innovative path of traditional sports culture dissemination of ethnic minorities in the Internet+ era is designed.

2

Research modeling and study design

In order to explore the innovative path of traditional minority sports culture dissemination methods in the Internet+ era, this paper firstly constructs a structural equation model using the PLS method to analyze the key factors affecting the dissemination of traditional minority sports culture.

2.1

Structural equation modeling based on PLS methodology

2.1.1

Structural equation modeling

SEM is a multivariate statistical analysis method used to discuss the relationship between latent variables (also known as structural variables) and explicit variables (also known as measurement variables) as well as the relationship between latent variables and latent variables. There are two elements of structural quantitative equation modeling: the variables and the relationships between the variables [27].

1)

Categorization of variables

According to the different characteristics of the variables, the variables in structural equation modeling are divided into: latent variables and explicit variables. Latent variables are variables with abstract concepts that cannot be measured directly, but can only be measured indirectly by observing explicit variables, while explicit variables are concrete measurable variables.

According to the different positions of latent variables in the model, they can be divided into two categories: exogenous variables and endogenous variables. Exogenous variables only serve as explanatory variables; they only affect other variables and are not affected by any other variables.

2)

Relationships between variables

As far as the relationship between variables is concerned, a structural equation model includes a structural model and a measurement model.

(1)

Structural Model

A structural model describes the causal relationship between latent variables and the equation can be expressed as: (1) $η = B η + Γ ξ + ζ$

Where: η = (η₁,η₂,…,η_n) is a vector consisting of endogenous latent variables. ξ = (ξ₁,ξ₂,…,ξ_n) is the vector consisting of exogenous latent variables. B(m×m) is the matrix of path coefficients of endogenous latent variables, describing the influence of endogenous latent variables on each other. Γ(m×n) is the matrix of path coefficients of exogenous latent variables describing the influence of exogenous latent variables on endogenous latent variables. ζ(m×1) is the vector consisting of residual terms reflecting the part of η that cannot be explained in the equation.

(2)

Measurement model

The measurement model measures the relationship between the latent variables and their explicit variables, and the equation can be expressed as: (2) $\begin{array}{l} y = Λ_{y} η + ε_{y} \\ x = Λ_{x} ξ + ε_{x} \end{array}$

Where y = (y₁,y₂,…,y_p) is a vector of endogenous explicit variables and is an observation of η. x = (x₁,x₂,…,x_q) is the vector of exogenous explicit variables and is the observation of ξ. Λ_y(p×m) and Λ_x(q×n) are the loading matrices. ε_y(p×1) and ε_x(q×1) are the residual vectors.

2.1.2

PLS modeling approach

There can be two different modeling techniques for parameter estimation and testing of structural equation models, which are LISREL and PLS. In this study, PLS modeling technique was used to achieve structural equation modeling of the factors influencing the way of ethnic minorities’ traditional sports and culture dissemination in the era of Internet+ [28]. The software used is a combination of VisuaPLS and R software for modeling research.

1)

Basic structure

The structural equation model based on PLS modeling also has two types of variables (latent variables and explicit variables) and two types of models (structural model and measurement model). PLS-based measurement models have two forms: reactive models and constitutive models. The equations for the reflective measurement model are given in Equation (2), and the equations for the constitutive measurement model are as follows: (3) $\begin{array}{l} η = π_{η} y + δ_{η} \\ ξ = π_{ξ} x + δ_{ξ} \end{array}$

Where: η,ξ denotes the vector of endogenous and exogenous variables respectively. y,x denotes the vector of explicit variables of endogenous and exogenous variables respectively. π_η,π_ξ is the coefficient matrix. δ_η,δ_ξ is the residual vector.

2)

Parameter estimation

PLS is an iterative algorithm with the following procedure:

(1) Explicit variable centering. Each explicit variable x_ij,y_ij corresponds to n observations, thus there are: (4) $x_{i j} = (x_{i j 1} x_{i j 2} L x_{i j n})$ (5) $y_{i j} = (y_{i j 1}, y_{i j 2}, \dots, y_{i j n})$

First make E(x_ij) = E(y_ij), where i = 1,2,…,k₁, (number of MVs in exogenous LVs) j = 1,2,…,k₂ (number of MVs in endodust LVs). For i,j the definition is the same throughout the iteration.

(2) External approximation: generates an extrinsic estimate of the LV. I.e: (6) $X_{i}^{t + 1} = f_{i}^{t + 1} \sum_{j = 1}^{k_{1}} (ϖ_{i j}^{t} \cdot x_{i j})$ (7) $Y_{i}^{t + 1} = g_{i}^{t + 1} \sum_{j = 1}^{k_{2}} (ϖ_{i j}^{*_{t}} \cdot y_{i j})$

Where: $X_{i}^{t + 1}, Y_{i}^{t + 1}$ is the vector of extrinsic estimates after the tth iteration of latent variables ξ_i and η_i respectively. $ϖ_{i j}^{t}, ϖ_{i j}^{*_{t}}$ are the weights of the manifest variables respectively. $f_{i}^{t + 1}$ and $g_{i}^{t + 1}$ are scalars such that $V a r (X_{i}^{t + 1}) = V a r (Y_{i}^{t + 1}) = 1$ . t is the marker for the iteration.

During the iteration process, there are 3 different methods to determine the weights $ϖ_{k_{j}}$ and obtain the external estimates of the corresponding LVs according to the selection of the external relations.

a) Simple regression (Model A): applied to LV estimation for outward-looking groups of districts. If y_{k_j} is a series of measured variables, ηj is a predictor variable, y_{k_j} = ω_{k_j}η_j + e_{k_j}. Coefficients of covariance (i.e., loadings) are used as weights, i.e., $ϖ_{k_{j}} = C o v [η_{j}, y_{k_{j}}]$ , by iterating the equation until it converges for all $k_{j}, ϖ_{k_{j}}$ , the principal component η_j is obtained, which is the best predictor of the y_{k_j}.

b) Multiple regression (Model B): applied to LV estimation in the inward zone group. If η_i is a predictor variable, y_{k_j} is some column of predictor variables, $η_{j} = \sum_{k_{j}} ϖ_{k_{j}} y_{k_{j}} + d_{j}$ . The multiple regression coefficients between MV and LV are used as weights, i.e., $ϖ_{k_{j}} = {(y_{k_{j}} y_{k_{j}})}^{- 1} y_{k_{j}} η_{j}$ , η_j are the best predictor variables.

c) Combined use of simple and multiple regression (Model C). Different models can be used for different groups of districts in structural equation modeling. For example, model B is used for exogenous LVs and model A for endogenous LVs.

(3) Internal approximation: generating intrinsic estimates of latent variables.

Firstly, the concept of neighboring latent variables needs to be clarified. The so-called neighboring latent variables are those latent variables that have a path link with a certain latent variable in the path diagram. Let a (subscript) represent the neighboring latent variable. Then the intrinsic estimator of the latent variable is: (8) $ξ_{i}^{t + 1} = f_{i}^{*_{t + 1}} \sum_{a} (θ_{i a}^{t + 1} \cdot X_{a}^{t + 1} + λ_{i a}^{t + 1} \cdot Y_{a}^{t + 1})$ (9) $η_{i}^{t + 1} = g_{i}^{* t + 1} \sum_{a} (θ_{i a}^{* t + 1} \cdot X_{a}^{t + 1} + λ_{i a}^{* t + 1} \cdot Y_{a}^{t + 1})$

Where: $ξ_{i}^{t + 1}, η_{i}^{t + 1}$ is the vector of intrinsic estimates after the tth iteration of latent variables ξ_i and η_i respectively. $θ_{i} a^{t + 1}, θ_{i a}^{*_{t + 1}}, λ_{i a}^{t + 1}$ and $λ_{i a}^{* t + 1}$ are the inner weights of the latent variables, respectively. $f_{i}^{* t + 1}$ and $g_{i}^{* t + 1}$ are scalars such that $V a r (ξ_{i}^{t + 1}) = V a r (η_{i}^{t + 1}) = 1$ , T are the markers for the iterations.

Secondly, the method of internal weighting needs to be clarified. There are three main methods of internal weighting: midpoint weighting method, path weighting method and factor weighting method.

a) Midpoint weighting method: this is the original PLS algorithm. This method only utilizes the direction of LV correlation, ignoring the direction of causality and the degree of correlation between LVs.LVs are approximated by the sign-weighted sum of neighboring LVs, similar to the midpoint factor.

b) Factor weighting method: the target LV is weighted by the correlation coefficients between its neighboring LVs, ignoring the causal direction. Therefore, this LV is called the “principal component” of its neighboring LVs.

c) Path-weighting method: the LV is estimated to be the best dependent variable for its predictor variable and the best independent variable for the subsequent predictor variable. In calculating the weighted sum for approximation, the predictor variables of the LV are weighted by the multiple regression coefficients, while the predicted variables of the LV are weighted by the correlation coefficients.

(4) Weight estimation: i.e., determining the weights of the explicit variables, here it is necessary to distinguish between two types of measurement models: reflective and constitutive models.

Reflective model weight estimation: (10) $x_{2 j} = ϖ_{2 j}^{t + 1} ξ_{2}^{t + 1} + ε_{x, 2 j}^{t + 1}$ (11) $y_{i j} = ϖ_{i j}^{* t + 1} η_{i}^{t + 1} + ε_{y, i j}^{t + 1}$

Compositional model weight estimation: (12) $ξ_{l}^{t + 1} = \sum_{j} (ϖ_{1 j}^{t + 1} \cdot x_{1 j}) + δ_{ξ, 1}^{t + 1}$

For reflective models, the loadings become the weights of the explicit variables, while for constitutive models, the regression coefficients become the weights of the explicit variables, and these weights are involved in the next iteration. When the iteration starts, i.e., at t = 0, the initial weights can be assigned arbitrarily.

(5) Judgment of the end of iteration: when a round of estimation is finished, it is necessary to judge whether to end the iteration or not, if the condition of stopping is not reached, the weights of the explicit variables calculated by equations (10)~(12) will be substituted into equations (6)~(7) for the next round of iteration.

The common conditions for stopping the iteration are: (13) $| ϖ_{i j}^{t} - ϖ_{i j}^{t + 1} | < 10^{- 5} and | ϖ_{i j}^{* t} - ϖ_{i j}^{* t + 1} | < 10^{- 5}$ (14) $| (ϖ_{i j}^{t} - ϖ_{i j}^{t + 1}) / ϖ_{i j}^{t} | < 10^{- 5} and | | (ϖ_{i j}^{*} - ϖ_{i j}^{* + 1}) / ϖ_{i j}^{*} | < 10^{- 5}$

(6) Find the value of the latent variables: using the weights determined after iteration, calculate the vector corresponding to each latent variable. I.e.: (15) $ξ_{i}^{T} = \sum_{j = 1}^{k_{1}} (ϖ_{i j}^{T} \cdot x_{i j})$ (16) $η_{i}^{T} = \sum_{j = 1}^{k} (σ_{i j}^{* T} \cdot y_{i j})$

Where: the superscript T indicates the calculation result at the end of the iteration.

(7) Finding load and path coefficients: using the values of the latent variables and the values of the manifest variables, ordinary least squares regression is carried out to calculate the load coefficients and path coefficients, respectively.

2.2

Research hypotheses

2.2.1

Research Hypotheses on Perceptions, Attitudes and Behaviors of Communication Effectiveness

The theory of persuasive effect reveals the process by which information dissemination effectively influences human behavior. In the process of mass communication activities, the audience first has a perceptual reaction to the received information, compares the original opinion with the current opinion according to the level of knowledge and information reserve, and then makes a cognitive change. When the motivation to make a new response is stronger than the motivation to maintain the original response, the audience will change their attitude accordingly, whether they are in favor of or against the existing information. Behavioral implementation is hidden behind the change in attitude, and when the attitude is changed, the corresponding action is generated. The audience will then have to abandon the existing information or forward to stay, etc.. Based on this, the theory of persuasion effect forms a progressive relationship between “cognition-attitude-behavior”. At the same time, relevant studies show that the influence of cognitive, attitudinal and behavioral levels on the communication effect is also increasing step by step.

Based on this, this paper puts forward the following research hypotheses on the communication effect of traditional sports culture of ethnic minorities:

H_1a: Audience’s cognition about traditional minority sports culture positively affects audience’s behavior.

H_1b: Audience’s attitude about traditional minority sports culture will positively influence audience’s behavior.

H_1c: Audience perceptions about traditional minority sports culture will positively influence audience attitudes.

2.2.2

Research hypotheses on individual audiences and the effects of cultural communication

Audiences are selective in their use of media information, and often tend to choose information that is consistent with their own perceptions, while rejecting content that contradicts their own inherent notions. Based on this, this paper puts forward the following research hypotheses on the effect of individual audience on the dissemination of traditional sports culture of ethnic minorities:

H_2a: Individual audience members will positively influence audience perceptions.

H_2b: Individual audience will positively influence audience’s attitude.

2.2.3

Research hypotheses on content ontology and cultural communication effects

Among the factors affecting the dissemination effect of traditional ethnic minority sports culture, the content itself of traditional ethnic minority sports culture has the most direct impact on the dissemination effect.

Therefore, this paper puts forward the following research hypotheses about the content ontology on the communication effect of minority traditional sports culture:

H_3a: The content ontology of minority traditional sports culture positively affects the audience’s perception.

H_3b: The content ontology of traditional ethnic minority sports culture will positively influence the audience’s attitude.

2.2.4

Research hypotheses on the subject of communication and the effects of cultural communication

The main body of communication plays an important role in the process of promoting traditional minority sports culture. In the Internet+ era, the authority of the media, the ease of use and smoothness of the communication platform, and the diversification of the communication channels will all have a certain impact on the cultural communication effect. Based on this, this paper presents the following research hypotheses regarding the impact of communication subjects on the communication of traditional minority sports culture:

H_4a: The communication subject will positively influence the audience’s cognition.

H_4b: The communication subject will positively influence the audience’s attitude.

2.2.5

Research hypotheses on environmental objects and cultural communication effects

Social cognitive theory suggests that behavior is not only governed by the individual’s subjective needs, but also by environmental conditions. In the process of promoting minority traditional sports culture, its dissemination effect will also be affected by the environment.Based on this, this paper presents the following research hypotheses regarding the impact of environmental factors on the spread of minority traditional sports culture:

H_5a: Environmental objects will positively affect the audience’s cognition.

H_5b: Environmental objects will positively affect audience’s attitude.

2.2.6

Conceptual modeling of structural equations

Based on the structural equation modeling method of PLS, combined with the previous research hypotheses, this paper uses AMOS23.0 software to draw the complete path structural equation model of the influencing factors of the dissemination effect of the traditional sports culture of ethnic minorities as shown in Figure 1. The model consists of a total of 7 latent variables and 11 paths.

2.3

Study design

2.3.1

Scale Description and Variable Measurement

In this paper, structural equation modeling is used to verify the influence of individual audience, content ontology, communication subject, and environment object on communication effects. In order to ensure the reliability and validity of the observational measurements, the content of the scale is adopted as much as possible from the mature scales that have been used in domestic and international literature.

The scale in this paper adopts a five-level Likert scale with 21 observational variables. The latent and observed variables in the scale and their descriptions are shown in Table 1.

Table 1.

Description of variables

Variable classification	Latent variables	Observed variables
Independent variables	Audience individual	Educational level (AUD1)
		Hobbies and interests (AUD2)
		Media literacy (AUD3)
		Cultural literacy (AUD4)
	Content ontology	Understandability (CONT1)
		Interestingness (CONT2)
		Multimedia capability (CONT3)
		Integration (CONT4)
	Communication subject	Authority (COMM1)
		Ease of use (COMM2)
		Fluency (COMM3)
		Diversification of communication channels (COMM4)
	Environmental object	Institutional environment (ENV1)
		Cultural atmosphere (ENV2)
		Media technology (ENV3)
Dependent variables	Degree of cognition	Know of (COG1)
	Have gained some understanding (COG2)
	Degree of understanding	Agree (UND1)
	Degree of understanding	Be interested (UND2)
	Degree of participation	Comment (PAR1)
	Degree of participation	Share (PAR2)

2.3.2

Questionnaire design and data collection

Based on the scale descriptions and variable measurements described above, and in conjunction with the needs of this study, this paper designed the measurement items for the research variables.The first part of the questionnaire is the preface, which mainly explains the background and purpose of the research.The second part is the demographic characterization of the respondents, which mainly includes gender, education level, age, occupation, income, and whether they use digital communication media. The third part is the influencing factors of the traditional sports culture dissemination effect of ethnic minorities, and 15 measurement questions were set based on the observation variables. The fourth part is the communication effect of traditional sports culture of ethnic minorities, and 6 measurement questions were set based on the observation variables. The answers were set on a five-point Likert scale from 5 to 1 representing completely agree, agree, uncertain, disagree, and completely disagree, respectively. To ensure that the sample is representative and comprehensive, snowball sampling and quota sampling in non-probability sampling were used in this paper. The questionnaires were distributed on November 8, 2023 after a small pre-test and correction, and were collected one week later. A total of 404 responses were collected. Twenty-one invalid responses were deleted and a total of 383 valid responses were obtained.

The results of the descriptive statistics of the sample are shown in Table 2. There was no significant difference in the gender of the respondents. In terms of education, 89.30% of respondents had a bachelor’s degree or higher, indicating that the survey respondents with higher cultural literacy were more concerned about and recognized the traditional sports culture of ethnic minorities. Occupation is more dispersed, indicating that the survey questionnaire covers a wider range. In addition, 89.03% (341) of the respondents answered “yes” to the question of whether or not they use digital media, indicating that new media are widely used in cultural communication.

Table 2.

Sample description statistics

Item	Category	Frequency (copies)	Percentage (%)
Gender	Male	158	41.25
Gender	Female	225	58.75
Educational background	College and below	41	10.70
	Undergraduate	289	75.46
	Master	45	11.75
	Doctor and above	8	2.09
Age	<20	75	19.58
	20-25	128	33.42
	26-35	130	33.94
	36-45	36	9.40
	>45	14	3.66
Occupation	Teacher	42	10.97
	Pupil	168	43.86
	Researchers	44	11.49
	Administrative personnel	82	21.41
	Other	47	12.27
	Total	383	100.00
Income	1000-3000 yuan	141	36.81
	3000-5000 yuan	74	19.32
	5000-10000 yuan	108	28.20
	10000-20000 yuan	42	10.97
	>20000 yuan	18	4.70
Whether to use digital media	Yes	341	89.03
Whether to use digital media	No	42	10.97

2.3.3

Common method bias test

The influential factor model of the dissemination effect of traditional sports culture of ethnic minorities constructed in this paper involves a number of variables, and in order to ensure the rationality and scientificity of the study and to avoid the influence of a single sample source on the variables, it is necessary to carry out a common method bias test (CMV) on the collected data before analyzing them by using structural equation modeling.

In this paper, Harma’s one-factor test was used to factor analyze the 27 topics, and seven factors with eigenvalues greater than 1 were extracted by using unrotated principal component analysis through SPSS software, with a cumulative explained variance of 72.46%, of which the explained variance of factor 1 was 40.375%, which is less than the 50% criterion, indicating that the influence factor model does not have common method bias.

3

Empirical analysis of the model of cultural communication influencing factors

3.1

Reliability and validity analysis

3.1.1

Reliability analysis

Before the empirical analysis, the data in the required scale is analyzed for reliability, the reliability analysis is mainly through the reliability and authenticity of the measured scale to reflect the stability of the test results, the level of the reliability results also indicates the size of the stability of the measured scale, it is generally believed that the Cronbach’s coefficient is greater than 0.7 indicates a good degree of reliability. In this paper, Cronbach’s Alpha reliability coefficient is used to test the degree of consistency of the variables in the questionnaire study for each measurement item.

The results of the reliability analysis of each scale are shown in Table 3. The results of the analysis of the measurement data show that the coefficients of Cronbach’s Alpha of the variables audience individual, content ontology, communication subject, environment object, cognition, comprehension, and participation in the study of this paper are 0.921, 0.912, 0.926, 0.935, 0.874, 0.907, 0.904, which are all greater than 0.7, indicating that the variables have good internal consistency reliability.

Table 3.

Reliability analysis of each scale

Measurement dimension	Item	CITC	Cronbach’s Alpha after deleting the item	Alpha coefficient
Audience individual	AUD1	0.776	0.905	0.921
	AUD2	0.831	0.884
	AUD3	0.772	0.904
	AUD4	0.748	0.908
Content ontology	CONT1	0.805	0.895	0.912
	CONT2	0.748	0.893
	CONT3	0.746	0.894
	CONT4	0.779	0.887
Communication subject	COMM1	0.756	0.893	0.926
	COMM2	0.785	0.884
	COMM3	0.834	0.925
	COMM4	0.789	0.929
Environmental object	ENV1	0.802	0.927	0.935
	ENV2	0.774	0.930
	ENV3	0.818	0.925
Degree of cognition	COG1	0.745	0.819	0.874
Degree of cognition	COG2	0.783	0.785	0.874
Degree of understanding	UND1	0.808	0.864	0.907
Degree of understanding	UND2	0.759	0.886	0.907
Degree of participation	PAR1	0.805	0.857	0.904
Degree of participation	PAR2	0.782	0.869	0.904

3.1.2

Exploratory factor analysis

Validity analysis is an important element in empirical research, validity refers to whether the measured results can reflect the content to be examined, the more the measurements match with the content to be examined, the higher the validity.KMO value represents the validity, the KMO is more than 0.8 means high validity, 0.7-0.8 means good validity, 0.6-0.7 means the validity is acceptable, and less than 0.6 means poor validity. The whole scale in the questionnaire was firstly tested and the results of KMO value and Bartlett’s test of sphericity are shown in Table 4. The KMO value is 0.894 and the Bartlett’s test of sphericity value is significant (Sig.<0.05), which indicates that the validity of the overall sample data of the scale collected is better and it is suitable to be analyzed by factor analysis.

Table 4.

Inspection table of overall scale KMO and Bartlett’s test

A measure of adequacy of the Kaiser-Meyer-Olkin sample		0.894
Bartlett’s sphericity test	Approximate chi-square	6127.83
	df	502
	Sig.	0.000

Next, this paper will extract the factors using principal component analysis and maximum variance method, and the resulting variance explained ratio is shown in Table 5.

Table 5.

Variance explanation ratel

Factor number	Feature root			Rotational front difference interpretation rate			Explanation rate of variance after rotation
Factor number	Feature root	Variance explanation rate	Cumulation	Feature root	Variance explanation rate	Cumulation	Feature root	Variance explanation rate	Cumulation
1	13.504	41.525%	41.525%	13.504	41.525%	41.525%	3.862	12.431%	12.431%
2	2.317	6.848%	48.373%	2.317	6.848%	48.373%	3.834	12.352%	24.783%
3	2.120	6.521%	54.894%	2.120	6.521%	54.894%	3.825	12.314%	37.097%
4	1.959	6.374%	61.268%	1.959	6.374%	61.268%	3.706	12.247%	49.344%
5	1.706	5.529%	66.797%	1.706	5.529%	66.797%	3.121	9.928%	59.272%
6	1.583	5.182%	71.979%	1.583	5.182%	71.979%	3.004	9.613%	68.885%
7	1.241	4.157%	76.136%	1.241	4.157%	76.136%	2.305	7.251%	76.136%

As can be seen from Table 5, the factor analysis extracted a total of seven factors with eigenroot values greater than 1. The variance explained ratio of these seven factors after rotation is 12.431%, 12.352%, 12.314%, 12.247%, 9.928%, 9.613%, and 7.251%, respectively, and the cumulative variance explained ratio after rotation is 76.136%.

In order to explore the correspondence between each factor and the question items, this paper uses the maximum variance rotation method (varimax) to perform the rotation, and the factor loading coefficients after the rotation are shown in Table 6.

Table 6.

Factor load coefficient after rotation

Variables	Factor load coefficient							Common degree (common factor variance)
Variables	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5	Factor 6	Factor 7	Common degree (common factor variance)
AUD1	0.779	0.133	0.216	0.179	0.106	0.103	0.105	0.727
AUD2	0.808	0.151	0.188	0.112	0.088	0.118	0.153	0.758
AUD3	0.794	0.134	0.167	0.156	0.196	0.116	0.122	0.758
AUD4	0.797	0.148	0.139	0.186	0.121	0.145	0.085	0.755
CONT1	0.170	0.775	0.175	0.152	0.092	0.115	0.084	0.771
CONT2	0.154	0.792	0.198	0.209	0.146	0.162	0.091	0.755
CONT3	0.143	0.777	0.176	0.163	0.134	0.146	0.117	0.727
CONT4	0.201	0.763	0.171	0.151	0.141	0.168	0.168	0.734
COMM1	0.165	0.176	0.765	0.133	0.097	0.174	0.115	0.753
COMM2	0.159	0.194	0.782	0.151	0.175	0.153	0.069	0.753
COMM3	0.177	0.192	0.785	0.089	0.109	0.139	0.099	0.739
COMM4	0.191	0.144	0.785	0.163	0.159	0.113	0.114	0.733
ENV1	0.157	0.141	0.127	0.782	0.143	0.144	0.131	0.735
ENV2	0.153	0.175	0.109	0.779	0.166	0.129	0.141	0.742
ENV3	0.148	0.147	0.172	0.776	0.158	0.134	0.147	0.763
COG1	0.129	0.164	0.154	0.171	0.794	0.141	0.082	0.774
COG2	0.133	0.093	0.189	0.136	0.795	0.162	0.158	0.766
UND1	0.138	0.155	0.171	0.142	0.162	0.786	0.158	0.734
UND2	0.113	0.178	0.154	0.151	0.152	0.782	0.167	0.751
PAR1	0.152	0.182	0.138	0.182	0.155	0.149	0.783	0.801
PAR2	0.152	0.179	0.182	0.215	0.162	0.190	0.791	0.749

From Table 6, it can be seen that all the research items correspond to a common degree value higher than 0.7, which means that there is a strong correlation between the research items and the factors, and the factors can extract the information effectively.

3.1.3

Validation factor analysis

Validated factor analysis (CFA) is a research method used to measure whether the correspondence between factors and measurement items (scale question items) remains consistent with the researcher’s predictions [29].

In order to further measure the relationship between the questionnaire items and the measurement factors, this paper conducts a validated factor analysis for 7 factors as well as 21 analysis items. In this paper, the correlation between the factors and the analyzed items is demonstrated by calculating the factor loading coefficients, which are shown in Table 7. It can be seen that the absolute value of the standardized loading coefficient of each measurement relationship is greater than 0.8 and shows significance (p<0.05), which means that there is a better measurement relationship.

Table 7.

Factor load coefficient

Latent variables	Observed variables	Non-standard load coefficient	Std. Error	z (CR value)	p	Standard load factor
Audience individual	AUD1	1.000	-	-	-	0.811
	AUD2	0.986	0.036	25.137	0.000	0.826
	AUD3	1.01	0.044	25.25	0.000	0.837
	AUD4	0.976	0.040	24.54	0.000	0.822
Content ontology	CONT1	1.000	-	-	-	0.827
	CONT2	1.014	0.041	25.361	0.000	0.821
	CONT3	1.006	0.042	25.717	0.000	0.836
	CONT4	1.023	0.041	25.434	0.000	0.828
Communication subject	COMM1	1.000	-	-	-	0.809
	COMM2	1.037	0.043	24.212	0.000	0.805
	COMM3	1.028	0.041	24.446	0.000	0.819
	COMM4	1.027	0.043	24.216	0.000	0.803
Environmental object	ENV1	1.000	-	-	-	0.842
	ENV2	1.014	0.034	26.767	0.000	0.835
	ENV3	0.946	0.035	25.641	0.000	0.811
Degree of cognition	COG1	1.000	-	-	-	0.828
Degree of cognition	COG2	0.999	0.042	24.69	0.000	0.821
Degree of understanding	UND1	1.000	-	-	-	0.821
Degree of understanding	UND2	0.967	0.048	22.691	0.000	0.803
Degree of participation	PAR1	1.000	-	-	-	0.806
Degree of participation	PAR2	1.028	0.046	23.219	0.000	0.832

In order to explore the aggregation validity (convergent validity) among the factors and question items, this paper utilizes AVE (Average Variance Extraction) and CR (Combined Reliability) to test the aggregation validity, and the results of the test of aggregation validity are shown in Table 8.

Table 8.

Test of aggregation validity

Factor	Average variance extraction AVE value	Combined reliability CR value
Audience individual	0.689	0.924
Content ontology	0.682	0.920
Communication subject	0.674	0.916
Environmental object	0.693	0.921
Degree of cognition	0.686	0.905
Degree of understanding	0.671	0.897
Degree of participation	0.677	0.868

From Table 8, it can be seen that the AVE values corresponding to a total of seven factors are all greater than 0.6, and the CR values are all higher than 0.8, which means that the data of this analysis have good convergent (convergent) validity.

To further validate the item validity of the questions, the discriminant validity between the factors was analyzed in this paper, and the results of the discriminant validity test are shown in Table 9.

Table 9.

Test of discriminative validity

Factor	Coding	AUD	CONT	COMM	ENV	COG	UND	PAR
Audience individual	AUD	0.832
Content ontology	CONT	0.493	0.824
Communication subject	COMM	0.491	0.458	0.821
Environmental object	ENV	0.485	0.511	0.504	0.834
Degree of cognition	COG	0.429	0.457	0.496	0.426	0.833
Degree of understanding	UND	0.461	0.491	0.477	0.495	0.466	0.825
Degree of participation	PAR	0.464	0.469	0.494	0.492	0.467	0.512	0.827

As can be seen from Table 9, the AVE square root values of the seven variable factors of individual audience, content ontology, communication subject, environmental object, awareness, understanding, and participation are 0.832, 0.824, 0.821, 0.834, 0.833, 0.825, and 0.827, respectively, which are greater than the maximum values of the absolute values of the correlation coefficients between the corresponding factors of 0.493, 0.511, 0.504, 0.495, 0.512, and 0.512.This indicates good discriminant validity among the seven variable factors.

3.2

Correlation analysis

Correlation analysis is primarily used to determine how closely two variables are related and is used as a preliminary test of research hypotheses.In this paper, the mean values of the latent variables were calculated and the Pearson correlation coefficient test was performed using SPSS28.0. [30]. The hypothesis testing based on correlation analysis is shown in Table 10. Where ** indicates significant correlation at 0.01 level (two-sided).

Table 10.

Hypothesis testing based on correlation analysis

Research hypothesis	Correlation variable	Correlation coefficient
H_1a	Degree of cognition←→Degree of participation	0.339**
H_1b	Degree of understanding←→Degree of participation	0.291**
H_1c	Degree of cognition←→Degree of understanding	0.278**
H_2a	Audience individual←→Degree of cognition	0.354**
H_2b	Audience individual←→Degree of understanding	0.249**
H_3a	Content ontology←→Degree of cognition	0.415**
H_3b	Content ontology←→Degree of understanding	0.298**
H_4a	Communication subject←→Degree of cognition	0.376**
H_4b	Communication subject←→Degree of understanding	0.379**
H_5a	Environmental object←→Degree of cognition	0.325**
H_5b	Environmental object←→Degree of understanding	0.461**

The data in Table 10 show that there are different degrees of correlation between all the latent variables in the research hypotheses, and they are positively correlated, i.e., the hypotheses have been preliminarily verified.

3.3

Structural equation modeling analysis

3.3.1

Model fit analysis

When using structural equation modeling for hypothesis testing, it is important to ensure the structural validity of the data. Good model fit is essential for conducting structural equation modeling analysis. In this study, the cardinal degrees of freedom ratio (CMIN/DF), root mean square of approximation error (RMSEA), goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), value-added goodness-of-fit index (IFI), and comparative fitness-of-fit index (CFI) were selected to evaluate the model fit. The measurements, acceptable range, and degree of fit of each index are shown in Table 11. The CMIN/DF was 1.724<3, the RMSEA value was 0.051<0.08, the GFI was 0.872>0.8, and the values of AGFI, IFI, and CFI were 0.853, 0.942, and 0.941, respectively, and the goodness-of-fit measurements were all within the general research criteria, so the model was well-fitted.

Table 11.

Model fitting index

Fitting index	Acceptable range	Measured value	Degree of fit
CMIN/DF	<3	1.724	Good
RMSEA	<0.08	0.051	Good
GFI	>0.8	0.872	Good
AGFI	>0.8	0.853	Good
IFI	>0.8	0.942	Good
CFI	>0.9	0.941	Good

3.3.2

Structural Equation Analysis and Hypothesis Testing

The correctness of the path between latent variables during structural equation modeling analysis. It is usually considered that to test the correlation between two latent variables, the ideal values of the indicators of standard error, critical ratio, and p-value must be achieved. Among them, the ideal value of standard error S.E. is S.E. ≥ 0. The ideal value of critical ratio C.R. is C.R. > 2. And the conditions for determining the ideal value of the indicator of P value for hypothesis testing are: when P ≥ 0.05, it means not significant. When 0.01≤P<0.05, it means generally significant. When P<0.01, it indicates highly significant.

The constructed structural equation model and the standard paths in the research hypotheses are shown in Figure 2, including the role of cognition on engagement (H_1a), the role of comprehension on engagement (H_1b), the role of cognition on comprehension (H_1c), the role of individual audience on cognition (H_2a), the role of individual audience on comprehension (H_2b), the role of content ontology on cognition (H_3a), the role of content ontology’s role on comprehension (H_3b), communication subject’s role on cognition (H_4a), communication subject’s role on comprehension (H_4b), environment object’s role on cognition (H_5a), environment object’s role on comprehension (H_5b).

The specific structural equation modeling path coefficients are shown in Table 12.

Table 12.

Structural equation model path coefficient

Research hypothesis	Path	std.	Unstd.	S.E.	C.R.	P	Whether the hypothesis holds or not
H_1a	COG→PAR	0.285	0.175	0.063	3.216	0.003	YES
H_1b	UND→PAR	0.206	0.114	0.056	2.312	0.028	YES
H_1c	COG→UND	0.220	0.113	0.041	2.852	0.007	YES
H_2a	AUD→COG	0.384	0.267	0.073	4.112	0.000	YES
H_2b	AUD→UND	0.012	0.012	0.077	0.152	0.893	NO
H_3a	CONT→COG	0.323	0.302	0.071	4.316	0.000	YES
H_3b	CONT→UND	0.156	0.135	0.063	2.264	0.025	YES
H_4a	COMM→COG	0.204	0.231	0.097	2.505	0.018	YES
H_4b	COMM→UND	0.227	0.385	0.184	2.153	0.034	YES
H_5a	ENV→COG	0.275	0.571	0.192	3.165	0.003	YES
H_5b	ENV→UND	0.388	0.474	0.095	4.986	0.000	YES

As can be seen from Table 12, except for the non-significant relationship between brand communication body and brand emotion, all other path relationships pass the significance test and are statistically significant.

1) The standardized coefficient of awareness on engagement is 0.285, P<0.05, and awareness has a significant positive effect on engagement, so hypothesis H_1a is valid.

2) The standardized coefficient of comprehension on engagement is 0.206, P<0.05, which indicates that comprehension has a significant positive effect on engagement, so hypothesis H_1b is valid.

3) The standardized coefficient of cognition on comprehension is 0.220, P<0.05, indicating that cognition has a significant positive effect on comprehension, so hypothesis H_1c is valid.

4) The standardized path coefficient of individual audience on cognition is 0.384, P<0.05, indicating that individual audience has a significant positive relationship on cognition, so hypothesis H_2a is valid.

5) The standardized path coefficient of individual audience to comprehension is 0.012, P>0.05, indicating that individual audience to comprehension is not significant, so hypothesis H_2b is not valid.

6) The standardized coefficient of content ontology to awareness is 0.323, P<0.05, which indicates that there is a significant positive correlation between content ontology to awareness, so hypothesis H_3a is valid.

7) The standardized coefficient of content ontology to comprehension is 0.156, P<0.05, indicating a significant positive relationship between content ontology to comprehension, thus hypothesis H_3b holds.

8) The standardized coefficient of communication subject on comprehension is 0.204, P<0.05, indicating a significant positive relationship between communication subject on comprehension, so hypothesis H_4a is valid.

9) The standardized coefficient of communication subject to comprehension is 0.227, P<0.05, indicating that communication subject to comprehension has a significant positive correlation, so hypothesis H_4b is valid.

10) The standardized coefficient of environment object on cognition is 0.275, P<0.05, which indicates that environment object has a significant positive correlation on cognition, so hypothesis H_5a is valid.

11) The standardized coefficient of environmental object on comprehension is 0.388, p<0.05, which indicates a significant positive relationship between environmental object on comprehension, thus hypothesis H_5b is valid.

4

Design of Diversified Communication Paths for Traditional Sports Culture of Ethnic Minorities

In the era of “Internet Plus”, new media tools, as an important carrier of economic development, can effectively promote the development of various industries, and the traditional sports culture of ethnic minorities has also gained new communication channels in the new era. The disseminator and audience groups are not subject to time constraints and spatial limitations in the process of information exchange, and realize resource sharing at any time and any place, so as to realize the innovative development of dissemination pathways while maintaining the connotation of traditional culture. Statistics show that network technology has become an important driving force for the rapid dissemination of traditional ethnic minority sports culture, but at the same time, social software is prone to problems such as information bias, lack of content and accuracy, which need to be improved in the later dissemination process of traditional ethnic minority sports culture.

1) Building an educational resource platform of “Internet + traditional ethnic minority sports culture”.

Education and training are key to the further development of traditional sports culture in ethnic minorities. During the field investigation, this project found that many school teachers believe that traditional ethnic minority sports culture has not been fully integrated with new media technology in the education process, and that young people are resistant to the current passive teaching. The popularization of “Internet +” technology can improve the teaching environment, stimulate the enthusiasm of young people to learn, and for the dissemination of traditional ethnic minority sports culture with certain special characteristics, the construction of the “Internet +” dissemination platform can optimize the allocation of resources, and effectively improve the quality of dissemination. Improve the quality of dissemination.

The traditional mode of sports teaching is the teacher on the playground to explain the movements of young people, instilling theoretical knowledge, due to the limitations of the educational place, the teaching content is boring and monotonous, the teaching form is single, resulting in young people in the process of learning not to know, and it is difficult to understand the connotation of the course, the teaching of traditional sports culture of ethnic minorities must be fully integrated with the digital resource base and other advanced carriers, so that young people can actively participate in the learning. Learning, not only in the classroom learning to obtain knowledge, but also timely access to cultural information, at the same time compared with the traditional teaching content, young people are more interested in multimedia teaching, with better quality of teaching audio-visual video is more concerned about.

At this stage, some physical education teachers are not clear about the connotation of traditional sports culture of ethnic minorities, especially young physical education teachers pay less attention to this, resulting in poor teaching quality, multimedia teaching methods can be used through the network platform, in the form of “cloud classroom” and other forms of teachers and students to strengthen the learning, and at the same time, in the teaching practice, physical education teachers can create a virtual environment through VR technology to enhance the experience of young people. Technology to create a virtual environment, enhance the youth experience, enrich the teaching content, in recent years, more and more physical education teachers in the teaching of the introduction of game teaching method, to strengthen the cultural knowledge of young people, action skills, project norms and other aspects of education, to stimulate the interest of young people to learn at the same time, and effectively improve the quality of teaching.

2) Construction of “Internet + traditional ethnic sports culture” industrial culture brand

Traditional minority sports culture in the dissemination process is also inseparable from the brand building, the specific forms of industrial performance can be divided into the following three: First, with traditional minority sports culture features of the athletic performance and project competition. Secondly, fitness and leisure content with the characteristics of traditional minority sports culture.Thirdly, the supplies have the characteristics of traditional sports culture of ethnic minorities.

At this stage, many regions carry out ethnic tourism and sports projects according to local conditions, and cultural communicators also organize ethnic traditional sports and culture halls from a professional point of view, such as the She ethnic traditional sports and culture communicators in Zhejiang, Fujian and other places to open martial arts training courses, etc. Although the ethnic traditional sports and culture industry has certain shortcomings in the process of development, and the overall economic benefits are not good, with the existence of low social influence, low number, low profitability and lack of support from family members. Although the traditional ethnic sports and culture industry has some shortcomings in the development process, the overall economic benefits are not good, there are low social influence, the number of colleges is small, the profitability is low, and family members do not support, etc., the “Internet +” can promote the development of the traditional ethnic sports and culture industry through the diversified development mode, comprehensively integrating the online mode and the offline mode, improving the operation of the local traditional sports and culture industry of local ethnic minorities and realizing the double harvest of economic benefits and social benefits.

Ethnic minority traditional sports culture not only has rich cultural connotation, but also has better health, fitness, leisure effect, the modern society people are living in a faster pace of life, work pressure and life pressure, for health and fitness, leisure and entertainment requirements are higher. Minority traditional sports culture communicators can integrate minority traditional sports culture with the construction of health museums and fitness museums in business planning, give full play to cultural characteristics, and create a comprehensive and contemporary business industry platform. For example, through the way of network publicity, more people can be aware of the traditional sports culture of ethnic minorities have the function of health, fitness and leisure function, and actively carry out the traditional sports culture of ethnic minorities, private tutoring, network teaching, community exchanges, etc., to explore more potential learners, and to transform the traditional closed, passive teaching mode into an open, proactive teaching mode, so that the connotations of the traditional sports culture of ethnic minorities can be used to promote the development of the traditional sports culture of ethnic minorities, and to promote the development of the traditional sports culture of ethnic minorities. The connotation of traditional sports culture can be developed globally, and at the same time, it also provides a guarantee for the survival of cultural institutions and cultural communicators, and strengthens the foundation for the sustainable development of traditional sports culture of ethnic minorities.

Through new media technology, cultural communication and tourism events can be combined, and organizers can innovate the events through scientific planning, benign development, efficient cooperation, etc. With the help of new media technology, the tourism industry can be driven in the process of the events to form a certain well-known tourism, sports and culture brand, and at the same time, it can also realize the sales and promotion of products such as national costumes, traditional handicrafts, cultural shirts and sports equipment, making the overall competitiveness of the industry chain enhance. At the same time, it can also realize the sales and promotion of products such as ethnic clothing, traditional handicrafts, cultural shirts, sports gears, etc., making the overall competitiveness of the industrial chain increase. Generally speaking, traditional sports culture of ethnic minorities can be orderly promoted under the industrialization development mode, innovate industrialization development bases with high-quality products, and realize the integrated development of tourism, competitions, products and services.

5

Conclusion

Based on the PLS method, this paper realizes the structural equation modeling of the influence elements of minority traditional sports culture communication, and on this basis, proposes the innovative path of minority traditional sports culture communication in the Internet+ era.

Structural equation modeling is used to explore the influence of individual audience, content ontology, communication subject and environment object on the communication effect (including awareness, understanding and participation).The coefficient scores of Cronbach’s alpha of the seven research variables are all greater than 0.8, the KMO value is 0.894, the value of Bartlett’s test of sphericity is significant (Sig. < 0.05), and the explanatory rate of the rotated cumulative equation reached 76.136%, indicating that the variables have good reliability and structural validity. Meanwhile, the corresponding AVE values of the seven factors were all greater than 0.6, the CR values were all higher than 0.8, and their AVE square root values were all greater than the maximum of the absolute values of the corresponding correlation coefficients among the factors, indicating that the seven variables have good convergent and discriminant validity among the factors. In addition, there are different degrees of correlations among the latent variables in the research hypotheses, all of which are significantly positively correlated at the 0.01 level, which preliminarily validates the research hypotheses.

The measures of the cardinal degrees of freedom ratio (CMIN/DF), root mean square of approximation error (RMSEA), goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), value-added goodness-of-fit index (IFI), and comparative fitness-to-fit index (CFI) of the constructed structural equation model were 1.724, 0.051, 0.872, 0.853, 0.942, and 0.941, which are within the general research criteria, thus the model fit goodness of fit. Among the research hypotheses, only the hypothesis that there is no significant correlation (P=0.893>0.05) between the latent variables in H_2b (individual audience → comprehension) does not hold, while the rest of the research hypotheses hold (P<0.05). The three factors of the environment object, communication subject, and content ontology all contribute positively to awareness and comprehension, while the individual audience only contributes positively to awareness.Based on the relevant conclusions, this paper proposes an innovative path to promote and optimize the communication effect of traditional sports culture among ethnic minorities.

Funding:

Guangdong Province Philosophy and Social Science Planning “Chaozhou Culture Research Special Project”: “Research on the Documentation and Digitization Inheritance Mechanism of Traditional Folk Sports in Chaoshan Area” (GD23CZZ07); Guangdong Province Philosophy and Social Science Program “Study on the Mechanism and Path of Inclusion, Sharing and Winning of Ethnic Traditional Sports in Guangdong, Hong Kong and Macao Greater Bay Area” (GD24CTY02).

Lingua:: Inglese

Frequenza di pubblicazione:: 1 volte all'anno
Argomenti della rivista:: Scienze biologiche, Scienze della vita, altro, Matematica, Matematica applicata, Matematica generale, Fisica, Fisica, altro

Feed RSS della rivista

The Innovation of Traditional Sports Culture Dissemination Mode of Ethnic Minorities in the Age of Internet Plus

Ling Qin

Shaoyu Chen

Xiang Fu

Ming Luo

Pubblicato online: 21 mar 2025

Ricevuto: 29 ott 2024

Accettato: 02 feb 2025

DOI: https://doi.org/10.2478/amns-2025-0621

Parole chiaveStructural equation modeling, PLS method, Traditional minority sports culture, Influencing factors

© 2025 Ling Qin et al., published by Sciendo

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

Parole chiave
Structural equation modeling, PLS method, Traditional minority sports culture, Influencing factors