The effect of action-based entrepreneurship education on intention to become an entrepreneur

The kingdom of Morocco has launched over the last decade major reform projects in order to strengthen youth entrepreneurship. Therefore, it is important to identify factors contributing to enhanced youth entrepreneurship activity. Hence, this method article examines the determinants of public university students’ entrepreneurial intention, by focusing on the importance of action-based entrepreneurship education. Data were collected using a face-to-face questionnaire from management students who had completed a program in action-based entrepreneurship. The data analysis design incorporates both exploratory (PCA using IBM SPSS Statistics 26) and confirmatory factor analysis (PLS-SEM using SmartPLS 3). Findings showed that action-based entrepreneurship education positively and significantly affects attitude towards entrepreneurship, and perceived entrepreneurial capacity. In addition, social norms positively influence attitude towards entrepreneurship and perceived entrepreneurial capacity, which turns to enhance students' entrepreneurial intention. Managers of Moroccan higher schools of technology may use this method article to pinpoint critical factors for enhancing students' entrepreneurial intention.• This method article proposes a practical approach to teaching entrepreneurship based on the learning-by-doing approach.• This method article can be used as a reference for researchers interested in studying the role of entrepreneurship education in promoting entrepreneurship in universities.• This method article can be used in order to identify the determinants of entrepreneurial intent among engineering students.

Environmental Science More specific subject area: Entrepreneurship Method name: Evaluate the effect of entrepreneurship education on intention to become an entrepreneur using the exploratory factor (PCA) and confirmatory factor analysis (PLS-SEM). Name and reference of original method: Exploratory factor analysis [1] and partial least squares structural equation modeling [2] .

Method details
With the aim of evaluating the effect of action-based entrepreneurship education on students' attitude towards entrepreneurship, perceived entrepreneurial capacity, and students' intentions to start up a business, this study mobilizes the exploratory factor analysis and the confirmatory factor analysis as two complementary approaches [1] . Fig. 1 outputs the different steps of the method implementation. Table 1 synthesizes the different steps of setting up the exploratory and confirmatory factorial analysis. We performed principal component analysis (PCA) to purify the measurement scales. Further, we performed a structural equation modeling (SEM) to test hypotheses and the research model.

Conceptual model
The conceptual model of this study was built on the expansion of the theory of planned behavior, by adding entrepreneurship education. Fig. 2 outlines the conceptual model, which supposes the direct and positive effect of entrepreneurial education on attitude towards entrepreneurship (H 1 ), and perceived entrepreneurial capacity (H 2 ). This model also indicates that social norms influence attitude towards entrepreneurship (H 3 ), perceived entrepreneurial capacity (H 4 ), and students' entrepreneurial intentions (H 5 ). In addition, attitude towards entrepreneurship (H 6 ), and perceived entrepreneurial capacity (H 7 ) positively influence students' entrepreneurial intentions.

Constructs operationalization
For operationalization of the constructs, we used measurement scales selected from existing studies. Therefore, attitude towards entrepreneurship (ATE) was measured with five items [3] . We selected four items to measure social norms (SON) [4] . The measurement scale for perception of entrepreneurial capacity (ENC) comprised 14 items [5] . The students' entrepreneurial intentions were measured using six items [6] . Entrepreneurship education (ENE) was measured using eight items [7] . As well, a 7-item Likert-type scale ranging from 1 (total disagreement) to 7 (total agreement) was employed to measure the questions related to these variables.
The sampling frame consisted of final-year management students of Laayoune Higher School of Technology, including professional bachelor and university diploma of technology students. These students underwent 50 hours of entrepreneurship and project management education. At this level, the pedagogical program adopted was designed around the learning by doing approach, which was conducted in three steps. The first step provided students with a theoretical background of entrepreneurship, by focusing on project management methods, entrepreneurial approach, entrepreneurial culture, entrepreneurs' typology, idea and business opportunity, and business model canvas. The second step consisted of in-group workshops composed of five students, working together on a business idea, market study, and the elaboration of the financial plan. After this second step, the last step consisted in organizing a business plan competition in order to conduct an individual and collective evaluation. This training program is designed to build a positive attitude among management students in terms of self-efficacy and tolerance for ambiguity, as well as to improve their knowledge and skills, particularly in marketing, finance, problem-solving and critical thinking [8] .

Data collection technique
The questionnaire was conducted face-to-face among students who validated this training program, during a week-long period in April 2019. At this stage, 98 eligible responses have been obtained. As Marvelous Determining the number of factors to be considered Kaiser criterion % total variance explained > 60% Examination of eigenvalues Selection the factors before inflection point Factorial solution interpretation Varimax rotation -Orthogonal rotation: in order to streamline interpretation of the factors by reducing the number of variables with strong correlations on each factorial axis. Communalities value must be higher than 0.4 Factor loading value must be higher than 0.5 Reliability analysis Crombach alpha

Second stage. CFA -Partial least squares structural equation modeling (PLS-SEM)
A Outer model evaluation (reflective model)

Convergent validity assessment
Cronbach's alpha α value must be higher than 0.7 Reliability ρ A value must be higher than 0.7 Composite reliability ρc value must be higher than 0.7

Loadings
Loadings must be higher than 0.7 Average variance extracted AVE must be higher than 0.5 Discriminant Validity assessment

Cross-loadings
The loading of an indicator on its assigned latent variable should be higher than its loadings on all other variables.

Heterotrait-Monotrait Ratio
The HTMT ratio values must be lower than 0.9 Fornell-Larcker criterion The square root of the AVEs for each structure should be greater than the construct's correlations with all other constructs A Significant at p-value < 0.01 * * t-value = 3.29 Significant at p-value < 0.001 * * * . illustrated in Fig. 3 , the sample included more females (65.3%) than males (34.7%), with the majority of them are aged between 19 and 23 years (68.4%). More than 54 percent of participants in this survey were students of the professional bachelor's degree in human resources management, whereas 45.9 percent of them were studying for a university diploma of technology in management techniques. Further, 58 percent of the interviewed students prefer entrepreneurship training based on the learning by doing approach. The largest proportion of surveyed students (84.7%) had no family background  in entrepreneurial activities. Lastly, only 39.8 percent of them have previously been volunteers with associations.

PCA results and discussion
The implementation of the principal component analysis (PCA) procedure allowed the purification of the different measurement scales. Using the IBM SPSS Statistics 26, this technique allowed us to remove ten items serving to measure the perceived entrepreneurial capacity, including ENC2, ENC3, ENC4, ENC6, ENC8, ENC9, ENC11, ENC12, ENC13, and ENC14. These items showed low scores regarding commonality ( < 0.4) and loading ( < 0.5). In addition, the PCA indicated that for each of the measurement scales only a single factor was retained ( Table 2 ). Table 3 presents the evaluation of the reflective measurement models. The average variance extracted, the Cronbach's alpha, the reliability ( ρ A ), and the composite reliability ( ρ c ) values are higher than 0.5, 0.7, 0.7, and 0.7, respectively. Moreover, discriminant validity is checked using the Fornell-Larcker criterion [9] , and the Heterotrait-Monotrait (HTMT) ratio [10] . Likewise, the discriminant validity was assessed according to the cross-loading ( Table 4 ). Table 5 shows the results of inner model assessment based on the coefficient of determination (R 2 ), and the predictive relevance (Q 2 ). The R 2 value of students' entrepreneurial intentions, attitude towards entrepreneurship, and perceived entrepreneurial capacity are 0.46; 0.44 and 0.40, respectively. Also, the data analysis indicates that the Q square values of all endogenous constructs are above 0, which demonstrates an acceptable predictive relevance [11] . As shown in Table 6 , all effect size values of exogenous construct on endogenous construct are acceptable, except the f 2 value of social norms on students' entrepreneurial intentions, which is 0.001.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Writing -original draft, draft preparation. Abdelaziz Ouajdouni : Review and Editing, translation and reviewing.

Funding resources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.