Research on Data Science-based Digital Empowerment Path in the Integration of Civic and Political Education and Dual Innovation Education in Colleges and Universities

In terms of the curriculum system of Civic-Creative Integration, based on the theoretical courses, practical courses and extension courses, we constantly optimize the curriculum system and establish a scientific and reasonable Civic-Creative Integration Curriculum System that adapts to the needs of students’ all-round development. In terms of theoretical courses, it builds a curriculum system in which compulsory courses and elective courses, online courses and offline courses complement each other, realizes the integration and penetration of ideological and political education and innovation and entrepreneurship education curriculum system, and meets the multifaceted needs of college students. In terms of practical courses, it strengthens the important position of entrepreneurship practice courses as the carrier of ideological and political education, realizes the organic unity of theoretical learning and practical operation, and reasonably builds a practical course system to enhance the practical ability of college students and meet the demands of social entrepreneurial talents. In addition, by setting up a second degree or a minor, the student-oriented innovation and entrepreneurship minor course is specially designed as an extension course of the curriculum system to serve the personality development of college students with entrepreneurial interests and to open up innovation and entrepreneurship “professional” education.

In terms of the teaching content of Civic-Creative Integration, the main line of in-depth integration is to establish moral values, integrate socialist core values, ethical and moral values, awareness of the rule of law, honesty and trustworthiness, and mental health education into the innovation and entrepreneurship courses, so as to build a “Civic-Creative Integration” course content system. As for the theoretical courses, on the basis of sorting out the teaching contents of the courses and taking into account the characteristics of the innovation and entrepreneurship education courses, we identify the integration space between the Civics courses and the Civics of the courses, and integrate the teaching contents. As for the practical courses, through systematic construction, combined with the content of the Civic and Political courses, the top-level design of the teaching content is strengthened, and innovative concepts and teamwork are integrated into the whole practical teaching process, avoiding the tendency to emphasize entrepreneurial skills, entrepreneurial project proposals, entrepreneurial practice projects, and so on, in order to realize the effective integration between the two.

In terms of teaching methods of Civic-Creative Integration, on the one hand, we explore new teaching modes such as inspirational teaching, case teaching, online and offline mixed teaching, flipped classroom, and diversified teaching such as “in-class + out-of-class” and “on-campus + out-of-campus” by means of the project-based system, activity-based system, and microclassroom, and so on. On the other hand, projects such as Civic Integration Excellence Courses, Demonstration Courses, and Famous Teachers’ Forums are carried out to promote the integration of Civic Integration in the already existing theoretical courses, seminars and sharing, and tournaments and other work sessions.

On the one hand, we make full use of the existing learning resources, such as the “Learning Strong Country” APP, the official website of Party history learning and education, the red thematic education base, the main theme of film and TV dramas, as well as the local “two courses”, the Party’s history learning platform and other platforms, to carry out in-depth integration of the Civic and creative thinking. Theme education. On the other hand, it has issued guidelines for the teaching of ideology and politics in the curriculum, built a library of elements and cases of ideology-creation integration with modern information technology as a tool, and increased students’ participation by means of case discussions and the presentation of curricular projects.

Regression analysis is an important branch of mathematical statistics, which is a powerful tool for exploring the causal relationship between various types of data. Logistic regression is one of the important methods of multivariate categorical analysis, and it is a powerful tool for dealing with multivariate complex variables and unconventional mathematical variables, especially some dichotomous problems, such as the problem between “disease” and “non-disease”, and the problem of “occurrence” and “non-occurrence” [19].

Mathematically speaking, the probability of an event occurring p may be any value in the interval [0,1], if 1 means that the event occurs, and 0 means that the event does not occur, then the event has only two states, “0” and “1”, which are discrete logical values, which can only be obtained by the transition function, and the ideal transition function is: 6P(Y=1|x)={ 0,z<0 0.5,z=βTx=0 1,z>0

In the above equation, x is a vector of explanatory variables affecting the occurrence of the event, β is a vector of regression parameters, z is a function of the explanatory variables x, and the value of function z will determine whether the outcome of the event is “occurrence” or “non-occurrence”.

Functions in conventional mathematical models are generally continuously differentiable, and the probability p of the event we want to obtain may be any value on the interval [0, 1], so an alternative function is needed to map a continuously differentiable mathematical function model to a set {0, 1} of only two logistic values. The alternative function for analyzing binary classification given by mathematicians CoxD.R et al (1970) is the logistic function (or Sigmoid function): 7g(z)=11+exp[−z]

The logic function image is shown in Figure 1(a), from which it can be seen that the logic function image is a monotonically increasing bounded “S” curve on interval 1, the fast track of the function value is close to [−∞, +∞] when the independent variable tends to positive infinity, and the function value quickly approaches 0 when the independent variable tends to negative infinity, which is exactly in line with the characteristics of the probability of event occurrence in the interval [0,1], which is very important for solving the binary classification problem.

For the independent variable obeys the binomial distribution (“0” or “1”) of the two-distribution problem, the probability of y = 1 and y = 0 are respectively: 8P(y=1)=p 9P(y=0)=1−p

p is the probability of y occurring and 1 − p is the probability of y not occurring, and the ratio of the two is the odds (odds), which refers to the ratio of the probability of an event occurring to the probability of it not occurring. Then the log odds are: 10ln(odds)=lnp1−p

From equations (6) and (7), we know that at z > 0, g(z) > 0.5, P(Y = 1|x) = 1, when the event occurs. And there is p < 1 − p when the event does not occur, i.e., p < 0.5, P(Y = 1|x) = 0, g(z) < 0.5, z < 0, g(z) can be used to describe the probability of the event occurring p. Thus: 11ln(odds)=lnp1−p=z=βTx

The image of the log odds function is shown in Fig. 1(b), and when the probability of event p is in the interval [0, 1], the log odds ln(odds) can be taken to any real number, and it is required that the function z can be taken to any number in the real number range. In equation (11), function z is a linear function of the explanatory variables, so function z can take any number in the real number range, which can also be seen in Figure 1(a).

Figure 1.

Logistic regression model function

With the logistic regression model in place, the next task to be accomplished is to use mathematical methods to solve for the vector of regression parameters β in function z.

In the actual modeling and analysis process, vector x usually contains more than one explanatory variable x₁, x₂…x_n, so the parameter vector β has more than one parameter β₀, β₁, β₂…β_n to be regressed, and the function z is expressed as follows: 12z=βTx=β0+β1x1+β2x2+…+βnxn

In the above equation: z is a function of the explanatory variable x₁, x₂…x_n and β₀, β₁, β₂…β_n is the regression parameter to be solved. Each database used for event analysis is a sample of data, and the overall database is not available, so it is necessary to

To use the sample data to estimate the overall parameters, in statistics, the maximum likelihood method is an efficient way to solve multiple regression parameters, multiple regression parameters can be obtained by solving the system of equations, which will not be repeated here.

After obtaining the regression parameters, a function of the probability of occurrence of the event can be obtained in the form of the following equation: 13π(x)=11+exp[−(β0+β1x1+β2x2+…+βnxn)]

In the above equation π(x) is the probability of occurrence of the event, x₁, x₂…x_n is the explanatory variable and β₀, β₁, β₂…β_n is the regression parameter.

The study launched an anonymous whole group survey on full-time academic graduate students enrolled in the School of Public Administration of a university in Province H in January 2024, and the information was collected by filling out the electronic questionnaire online, and 96 valid questionnaires were finally obtained.

The questionnaire was designed by ourselves, and the Cronbach’s alpha coefficient of the questionnaire was 0.938, which was greater than 0.9 and had a good reliability, and the KMO value of the questionnaire was 0.801, which was greater than 0.7 and had a good validity. The survey content of the questionnaire includes the basic information, ability cultivation, government support, enterprise cooperation, university support, and value leadership of the current graduate students, which is measured and assigned by Likert five-level scoring method.

SPSS24.0 was used to describe and analyze the data statistics. Firstly, the chi-square test was used to carry out single-factor and multi-factor analysis, and then the variables with statistical significance (P<0.05) in the ANOVA were included in the logistic regression model to start the analysis. Among them, the spirit of biculturalism and the performance of Civics and Politics were used as dependent variables, and ability cultivation, government support, enterprise cooperation, university support, and value leadership were used as independent variables, respectively.

The results of the univariate analysis are shown in Table 1. Overall, the median and the plural of the self-rating number of graduate students’ innovation and entrepreneurship aspirations are both 13, and about 47.6% of the enrolled graduate students have high innovation and entrepreneurship aspirations after dichotomous treatment, and the results of the chi-square test show that the innovation and entrepreneurship aspirations of graduate students are related to the cultivation of ability, government support, cooperation with enterprises, support from colleges and universities, and value leadership, which indicates that graduate students who are adequately high in innovation and entrepreneurship ability, receive government and university support, maintaining cooperation with enterprises, and receiving value leadership are more likely to show high innovation and entrepreneurship aspirations of graduate students.

The students’ learning data were selected and used to perform a multifactor ANOVA. In the general linear model, univariate commands (i.e., a single dependent variable) were executed to interpret the results in terms of competence development and value leadership. Table 2 shows the test for effects between subjects.

It can be seen that the significance tests of the overall model are less than the significance level, which shows that this ANOVA model is significant. Calculation shows that R2 = 0.927, so the conclusion is that ability cultivation and value leading are effective in improving the performance of Civics, and these two independent variables are also one of the main variables explaining the spirit of bicentennialism, with an explanatory part of 92.7%. In the two-by-two comparison of competence cultivation and value leadership, these two indicators have a significant effect on the effectiveness of Civics achievement.

The results of the multivariate tests are displayed in Table 3. As can be seen from Table 3, the significance test concludes that the significance value of each test is less than 0.01, and the effect of variables is significant. It can also be seen through the data that the gap between Pillai’s tracking data and Hotelling’s tracking data in the table is relatively large, with values of 0.465 and 0.907, respectively, which is a relatively large gap, indicating that the groups have a significant impact on the total model of the multivariate ANOVA.

Table 4 shows the test of between-subjects effect. From the data fed back in Table 4, it can be seen that the effect of each statistical variable on the validity of the Civics scores was highly significant (Sig. = 0 < 0.01), using the significance level (P = 0.01) as a measure.

From the analysis of the data, it can be seen that the ANOVA performed by these five independent variables verified the main and interaction effects of the multifactor ANOVA, and the independent effects on the observed variables were presented, which had a more significant effect on the distribution of the observed variables, and it was clear that among these five variables, all of them were statistically significant. Therefore, it can be determined that all five statistical variables are indicators of variables that have a significant impact on the performance of Civics.

SPSS 24.0 was run, the data in the Excel table was imported and the data was weighted by setting fixed class variables in the numerical and variable views of SPSS 24.0, then descriptive statistics were performed, the viewer gave cross-tabulations and corresponding chi-square tests to ensure the accuracy of the statistics, and the Fisher correction was used when the data were not satisfied.

The effectiveness analysis of competency development is depicted in Table 5. From the data in Table 5, it can be seen that the Pearson value in the total = 686, which is a large value to reflect the difference between the two variables. p = 0.001 < 0.01, the difference is relevant and statistically significant, so there is this conclusion that competence development has an impact on the performance of Civics.

Table 6 shows the validity analysis of government support. Combining the data in Table 6, it is easy to find that the p critical value is nearly 0 and Sig. < 0.01, which indicates that the government support factors that need to be validated are statistically significant in enhancing students’ performance in Civics.

The results of the analysis of the effectiveness of business cooperation are shown in Table 7. From the data feedback in Table 7, it can be seen that the Pearson value in the total is equal to 646.123, which is a large value and can reflect the difference between the two variables. The calculated minimum expected count = 0.04 < 5, and the exact Sig. approximates zero < 0.01, which shows that the statistical significance of the effectiveness of corporate cooperation on the performance of Civics exists.

Table 8 shows the validity analysis of college support. From the calculation of the results in Table 8, it can be seen that the minimum expected count = 0.05, which is significantly less than 5, which meets the requirements of the cardinality, so the value of the Pearson cardinality exact significance (two-sided) can be chosen. The asymptotic significance (two-sided) is close to zero, which shows the difference, so there is a statistical significance between the two variables.

The validity test of value leadership is shown in Table 9. As the results in Table 9 can be calculated, the minimum expected count = 0.06 < 5, the exact Sig. The chosen value is approximately zero and less than 0.01, which can be concluded that there is a statistically significant difference between the two variables.

Table 10 shows the logistic regression analysis of the level of dual entrepreneurial aspirations. The inclusion of variables with statistically significant chi-square tests in the logistic regression model of innovation and entrepreneurship aspirations reveals that value leadership is the most important statistically significant influencing factor, followed by government support and capacity development. Specifically, graduate students who received value leadership had approximately 14 times more innovative entrepreneurial aspirations than those who did not (OR=14.237).

Graduate students with high cultural training were more likely to have high innovation and entrepreneurship aspirations than those with low cultural training (OR=4.341). The degree of innovation and entrepreneurship aspirations of graduate students with high government support was about 3.7 times higher than that of students with low government support (OR=3.696). In addition, there was a statistically significant impact of corporate cooperation and university support on innovation and entrepreneurial aspirations.