PREDICTING ENTREPRENEURIAL AND CROWDFUNDING INTENTIONS ‒ A STUDY OF ROMANIA AND SOUTH KOREA

Crowdfunding has recently gained ground as an alternative financing means, particularly for start-ups and risky enterprises. We hereby analyse Romanian and South Korean students’ intentions to use crowdfunding in their future ventures. Using PLS-SEM (N=441), we first investigated the influence of the Theory of Planned Behaviour variables- attitudes, social norms, perceived behavioural control along with entrepreneurship education on entrepreneurial intentions. In addition, we looked into the influence of perceived trust and risk, performance expectancy, effort expectancy and facilitating conditions on crowdfunding intentions. Finally, we checked entrepreneurial intentions with crowdfunding intentions. We find that attitudes towards entrepreneurship, social norms, and entrepreneurial education have a positive influence and perceived behavioural control have a negative influence on entrepreneurial intentions. In terms of crowdfunding intentions, they were found to be positively influenced by performance expectancy, perceived trust and risk and entrepreneurial intentions. The results are useful both for educators designing higher education programmes in the field of entrepreneurship, as well as for crowdfunding platform owners in designing platform functionalities and image.


Introduction
Entrepreneurship is a creative and innovative process that helps to add economic value, create employment prospects, increase industry production, revive and expand markets, improve social welfare, and overall boost the economy (Guerrero et al., 2008). To be a successful entrepreneur, one must have the positive and strong intention to be one. Entrepreneurial intentions are defined as "the conscious state of mind that precedes action and directs attention toward entrepreneurial behaviours, such as starting a new business and becoming an entrepreneur" (Moriano et al., 2012, p. 165). As the job market has not seen much growth in the past two years given the contraction during the COVID-19 pandemic, many students may choose entrepreneurship as a career. Universities are increasingly backing courses and activities that develop entrepreneurial intentions, which is a proxy variable for actual entrepreneur behaviour in students (Meoli et al., 2020). Moreover the European Union has recommended that entrepreneurship should be one of the fundamental aptitudes for lifelong learning (EC, 2016).
To examine entrepreneurship behaviour, intention-based models, which include both theoryoriented and process-oriented approaches, are recommended. Of the existing intention models, the theory of planned behaviour (TPB) proposed by Ajzen (1991), is a prominent theory from social psychology that has been extensively used to predict intentions and predominantly in the entrepreneurial intentions domain (Esfandiar et al., 2019;Zaremohzzabieh et al., 2019). This theory postulates that behavioural intentions depend on the impact of three components: (a) attitudes towards the behaviour, (b) social norms (SN) (referring to family or society values), and (c) perceived behavioural control (PBC) (how difficult enacting a behaviour is perceived).
Mere intentions cannot guarantee the success in starting a business. There are many factors that make it hard for aspiring entrepreneurs to pursue their dream. Small and medium sized enterprises (SMEs) often find it difficult to obtain finance (Lee et al., 2015). There has been a significant decline in lending towards SMEs since the financial crisis of 2008 (Cowling et al., 2012). Tighter loan and lending regulations gave rise to alternative finance channels, and crowdfunding was at the forefront to provide access to finance for SMEs (Antonenko et al., 2014). Crowdfunding is an open call to the general public to contribute funds, usually small amounts, in a business idea, start-up or a cause in return for nothing (donation-based), nonfinancial (reward-based) and financial returns (equity and lending-based) (Baber, 2020). In the recent past, crowdfunding has not only helped entrepreneurs to raise funds, but also provided the means to promote their product to potential customers (Brem et al., 2019). The intention to raise funds through crowdfunding depends on many factors that enhance the perceived trust in the platform and reduce the risk for the entrepreneurs. Islam and Khan (2019) suggested that perceived performance expectancy, effort expectancy, social influence and facilitating conditions influence the adoption of crowdfunding to raise funds by the entrepreneurs.
The study will examine the factors influencing the entrepreneurial intentions of university students of Romania and South Korea. Roibu and Roibu (2016) contrasted and compared the two countries in terms of entrepreneurship. The study revealed that the positive factors which affect the entrepreneurial environment in South Korea are the decent physical infrastructure, market dynamics and efficient government policies, such as refining entrepreneurial education and supporting entrepreneurs in financing, infrastructure, etc. In Romania's case, the positive factors are the market dynamics and the legal and physical infrastructure. Also, Romania is one of the Central and Eastern European countries with one of the largest increases in entrepreneurial activity. These countries differ on the levels of entrepreneurship focus: South Korea is an innovation-driven economy and Romania is opportunity-driven. Therefore, it will be interesting to investigate the entrepreneurship intentions of students and further examine the intention to raise finance for their enterprise through crowdfunding. The study will predict entrepreneurial intentions through the TPB and entrepreneurship education. Entrepreneurial intentions will be further checked with crowdfunding intentions. Crowdfunding intentions will be observed through variables such as perceived trust, perceived risk, performance expectancy, effort expectancy and facilitating conditions. The study will contribute to practical implications by helping crowdfunding platforms to understand the points that are important from the perspective of an entrepreneur when choosing to raise funds through this method and then selecting a platform for this purpose.

TPB and entrepreneurial intentions
Personal propensity to get into business, the feeling of being capable to pursue such an activity and the support from those around the entrepreneur are factors that may have a large influence on entrepreneurial intentions, especially among students. These factors are grouped in the TPB, whichhas been previously used to explain entrepreneurial intentions among students (Nguyen et al., 2019). Esfandiar et al. (2019) have shown that entrepreneurial intentions are moderately influenced by attitude and self-efficacy (similar concept to PBC and feasibility), and that SNs have a relatively low influence on entrepreneurial intentions. However, the latter holds true in societies with low social legitimacy of entrepreneurship, as is Iran. A study on university students in Malaysia by Al-Jubari et al. (2019) reveals that attitudinal antecedents (attitude, SNs and PBC) can explain entrepreneurial intentions, with SNs having the weakest influence, although previous studies indicate that attitude fails to do that in collectivist contexts for cultural differences reasons (Siu and Lo, 2011). In line with the above findings, personal attitudes and SNs were found to be positively correlated with entrepreneurial intentions; however, PBC was not found to be correlated with the same variable (Gieure et al., 2019) or the correlation was very weak (Nguyen et al., 2019). SNs as seen through the lens of social context, i.e. supportive relevant others, influence entrepreneurial intentions and their enactment (Meoli et al., 2020). We formulate the following three hypotheses related to TPB: H1 -Attitude towards entrepreneurship has positive influence on entrepreneurial intentions.

H2 -SNs have positive influence on entrepreneurial intentions.
H3 -PBC has positive influence on entrepreneurial intentions.

Entrepreneurship education and entrepreneurial intentions
Besides the personal propensity for business or social influence, business knowledge and skills may also be determining factors of entrepreneurial intentions. The extant literature explains the role of entrepreneurship education in the formation of entrepreneurial intentions (Ndofirepi, 2020;Nguyen et al., 2019). Gieure et al. (2019) found that entrepreneurial skills

Predicting Entrepreneurial and Crowdfunding Intentions -A Study of Romania and South Korea
gained through effective education and training are conducive to entrepreneurial intentions, and further increase propensity to start a business. Therefore, we formulate the hypothesis: H4 -Entrepreneurship education has positive influence on Entrepreneurial intentions.

Performance expectancy, effort expectancy, facilitating conditions and crowdfunding intentions
Crowdfunding involves new technologies, and in considering crowdfunding intentions, one has to look into the factors related to acceptance and use of such techonologies. Performance expectancy, effort expectancy and facilitating conditions positively influence entrepreneurs' crowdfunding intentions as the decision to use crowdfunding is made on a previous costbenefit (achievements vs. efforts/easiness) analysis of Islam and Khan (2021). Performance expectancy is "the degree to which an individual believes that using the system will help him or her attain gains in job performance" (Venkatesh et al., 2003, p. 447). In crowdfunding, performance expectancy covers the base of backers, the reputation of the platform in fulfilling the promise, the time to raise funds and the usefulness of the platform. Effort expectancy refers to the "degree of ease associated with the use of the system" (Venkatesh et al., 2003, p. 450) and it signifies the perceived easiness of handling a new information system or technology (Islam and Khan, 2019). Effort expectancy in crowdfunding implies clarity and easiness of user interface, understanding the process of crowdfunding and perceived easiness in raising funds through crowdfunding. Facilitating conditions are viewed as the perceived technological and administrative support of infrastructure to provide hassle-free use of crowdfunding platforms (Baber and Fanea-Ivanovici, 2021).  found that facilitating conditions explain investors' intention, followed by performance expectancy and effort expectancy. Moon and Hwang (2018) checked that effort expectancy is a determinant of crowdfunding intentions, while performance expectancy is not. We formulate the following three hypotheses: H5 -Performance expectancy has positive influence on crowdfunding intentions.
H6 -Effort expectancy has positive influence on crowdfunding intentions.
H7 -Facilitating conditions have positive influence on crowdfunding intentions.

Perceived trust, perceived risk and crowdfunding intentions
Adopting new technologies in various fields of activity depends on the assessment of benefits and trust, on one hand, and associated risks, on the other hand. The studies of Moon and Hwang (2018) and Kim et al. (2020aKim et al. ( , 2020b identified that perceived trust has positive influence on crowdfunding intentions. On the other hand, Kim et al. (2020a) show that perceived risk has a weak negative influence on crowdfunding intentions. Baber and Fanea-Ivanovici (2021) found that backers' perceived trust and risk associated with platform influence positively on the intention to participate in crowdfunding campaigns.
H8 -Perceived trust has positive influence on crowdfunding intentions.
H9 -Perceived risk has negative influence on crowdfunding intentions.

Entrepreneurial intentions and crowdfunding intentions
In starting business, entrepreneurs evaluate the existing financing sources, and crowdfunding has emerged as a viable alternative. Baber (2022) has looked into the relationship between entrepreneurial intentions and crowdfunding intentions of international students and has found a positive correlation. In order to extend the scope of the previous study, we propose the following hypothesis: H10 -Entrepreneurial intentions have positive influence on crowdfunding intentions.

Research methods
The study is based on a sample of data collected from two countries, Romania and South Korea. The sample consisted in students of universities studying management, economics and business programmes. The reasons for the selection of the countries are given in the introduction section and the data was collected through a snowball sampling approach from each country. The survey link was shared with students in online zoom meetings, by mail and learning management system (LMS). The data was collected through an English administrated questionnaire as students were studying in an international setup. The items of constructs were taken from past studies, as follows: Attitude towards entrepreneurship and PBC from Nguyen et al. (2019) (2018), Effort expectancy from , Facilitating conditions and Perceived risk from Islam and Khan (2021), and crowdfunding intentions from Baber (2020). A total sample size of 441 was collected from both countries -Romania (224) and South Korea (217). Most of the respondents were from the age group 18-22 (59%), followed by the 23-26 age group (32%) and the rest (9%) were above 27 years of age. Interestingly, females were in majority (51%), 48% were males and 1% preferred not to disclose their gender.
Students were asked if they have any family or personal business experience. Around 34% of the students said they had, and out of those, 20% had less than 1 year of experience in handling the business and 12% had between 1-5 years. The sample (16%) had some experience in raising funds or backing a project in crowdfunding.
The data was analyzed using the Partial Least Square structural equation modelling (PLS-SEM) approach through SmartPLS 3.2 software. PLS-SEM is more useful in the earlier phases of theory development, it helps in exploration and theory development, it is convenient for testing a research framework where it is important to test the dependencies of the variable, predict the dependent variable, and where the structure is complex and data may lack normality (Hair Jr et al., 2020). Table no. 1 reports the factor loading of each item in the construct and reliability of constructs. Factor loadings for all items exceed the minimum 0.70 thresholds. The loading value of PBC3 is a little less than 0.7, hence it can be accepted. Cronbach's alpha and composite reliability criteria were used to assess the reliability of the data and confirmed reliability as the values of both assessing criteria were above 0.7 (Hair et al., 2019). To measure the convergent and divergent validity of the items, AVE (average variance extracted) and Fornell-Larcker criteria were used. AVE values exceed 0.5 minimum acceptable level (Hair et al., 2019), thus confirming that each item measures its corresponding construct; moreover, the square root of all values are higher than the correlation between constructs; hence, divergent validity is established (Fornell and Larcker, 1981).

Results
The path relationships of the overall sample (441) are shown in Table no Most of the coefficients are statistically significant at 95% (p-value < 0.05) and 90% (p-value < 0.10). There is no significant relationship evidence of effort expectancy and facilitating conditions on crowdfunding intentions. Interestingly, there is a negative but significant relationship between PBC and entrepreneurial intentions. Therefore, we accept all hypotheses except H6 and H7.
In PLS-SEM, the goodness-of-fit of the estimated model is evaluated by assessing goodnessof-fit indices through beta (β), P-values, effect size (f2), and R2, as suggested by Hair et al. (2019). The effect size (f2) shows the change in the R2 value of an endogenous construct due to eliminating an exogenous construct from the structural model. The effects established by path coefficients can be small (0.02), medium (0.15), or large (0.35) (Kock, 2020). Among the exogenous constructs, entrepreneurship education on entrepreneurial intentions (0.301) has a large effect and others, except H6 and H7, which fall between low and medium effect. The adjusted R2 values of entrepreneurial intentions and crowdfunding intentions are 0.372 and 0.377, respectively, which implies the variance in these constructs is reasonably explained by the predicting constructs, as shown in Figure no. 1.
To further confirm the model fit, the standardized root mean square residual (SRMR) is estimated, which is 0.05 for this proposed model, and the value of ≤ 0.1 is treated suitable for PLS path models (Kock, 2020).  Notes: ** Significance level at 95% (p-value < 0.05). * Significance level at 90% (p-value < 0.10). ***negative instead of proposed positive hypothesis.

Figure no. 1. Results of Hypothesis Testing
To further confirm the findings, the Partial Least Squares Multi-Group Analysis (PLS-MGA) was employed to differentiate the estimate of the paths for the two regions separately: Romania versus South Korea. The results of the MGA were evaluated using the Henseler-MGA nonparametric technique as shown in Table no. 3. This technique estimates the differences between the path coefficients of the two regions and then assesses group differences in PLS-SEM (Hair et al., 2019). All the p-value of the differences in path

Predicting Entrepreneurial and Crowdfunding Intentions -A Study of Romania and South Korea
coefficients between Romania and South Korea are above 0.05 (5% significance level), which implies there is no significant difference between Romania and South Korea students in the structural relationships.

Discussion
The results of the study confirm our hypothesis that students' attitudes towards entrepreneurship are positively related with the entrepreneurial intentions, which is in line with previous studies (Al-Jubari et al., 2019;Esfandiar et al., 2019;Gieure et al., 2019), and also expected due to the choice of the respondents' studies, i.e. in the field of business and economics. A positive attitude towards entrepreneurship is necessary for having the intention, and as students belong to management and economics programs, it is certain they will have a good attitude towards it. For South Korea and Romania, social norms are yet another explanatory positive factor of entrepreneurial intentions, which is in line with previous research (Gieure et al., 2019;Meoli et al., 2020), but contrary to other studies, such as those of Esfandiar et al. (2019) and Siu and Lo (2011). Our conclusions would rather place the two countries in the area of entrepreneurship-fostering societies from this perspective. Both countries have collectivist societies and, therefore, it is obvious that social influence will be evident on the career decisions of the students. The last item of the TPB we included in the study, perceived behavioural control, is surprisingly negatively correlated with entrepreneurial intentions. This is in opposition with previous findings, which indicate a positive correlation between the variables (Al-Jubari et al., 2019;Esfandiar et al., 2019), weak correlation (Nguyen et al., 2019). or no correlation at all (Gieure et al., 2019). The reason for such negative relationship can be that students who believe they cannot handle the affairs of the business are more in favour of being an entrepreneur. Favourable government policies, economic conditions, education and training and ease of doing business can bring confidence among those students to pursue their dream.
Entrepreneurial education plays an important role in forming entrepreneurial intentions, and our findings confirm previous studies, which indicate a direct and positive correlation between the two variables (Gieure et al., 2019;Nguyen et al., 2019;Ndofirepi, 2020). This conclusion complements that of the first hypothesis, as attitudes and skills prepare the stage for intention, and later implementation. Education about entrepreneurship imparts knowledge about the business, economics, organisation, and other functions of business, which brings confidence and improves the skills of students to aspire to be an entrepreneur.
Crowdfunding intentions were found to be positively correlated with performance expectancy, in line with previous studies (Venkatesh et al., 2003;Islam and Khan, 2021), but not influenced by effort expectancy and facilitating conditions, thus contradicting other studies (Moon and Hwang, 2018;Islam and Khan, 2021). Therefore, students believe the system proposed by crowdfunding platforms will help them achieve their objectives, without considering the attached diminished efforts to raise funds or the crowdfunding platform support in their choice. This conclusion complements and supports the positive correlation between perceived trust and crowdfunding intention, as performance expectancy and trust are kindred factors through platform reputation. This provides useful cues to platform owners as to how build their image and reputation in an attempt to attract financiers and fundraisers. Surprisingly and contrary to other studies (Kim et al., 2020a), perceived risk is a positive determinant of crowdfunding intentions. This may be because students do not have the necessary previous business experience to assess risk from various standpoints and because risk associated with crowdfunding platforms would, in the end, be borne from a financial standpoint, by backers, and the only menace for the start-up would be a delay in obtaining funds.
Last but not least, entrepreneurial intentions are an explanatory factor for crowdfunding intentions, as previously shown by Baber (2022). This is an indicator of good business skills in young entrepreneurs, as it reveals their knowledge of alternative financing tools and also of intention to use them.

Conclusions
The digital transformations in society have started to cover all areas of activity, including entrepreneurial activities and finance. In this context, crowdfunding has brought along more opportunities to implement projects and start new businesses, which can represent a source of increase for the SMEs and start-ups and of resilience for established companies facing difficulties as we are navigating turbulent times. Crowdfunding may not only be seen as a source of funding channel, but also as a platform to showcase the prototype of the product to be produced in the enterprise. Failure to achieve the desired funding goal on crowdfunding platform implies that backers do not believe in the project, and hence need more polishing of the idea or even dropping it. Either case, the crowdfunding platform can act as a marketing testing lab for budding entrepreneurs at a minimal a risk. The entrepreneurial intentions of these budding entrepreneurs are shaped by the attitude, social influence, perceived ability to be one and proper entrepreneurial education. They also often face the obstacle of raising funds. However, an innovative idea is appreciated by the financers, and the crowdfunding platform provides a podium to not only showcase the idea or product, but also convince the backers to invest in the project. There are various factors which influence an entrepreneur to forgo the traditional means of raising funds and opt for this novel method. These factors include performance expectancy, perceived trust and perceived risk related to the platform. Therefore, the findings of the present study will benefit multiple stakeholders, such as entrepreneurs, product buyers, financiers, crowdfunding platform owners, but also legislators when designing crowdfunding regulations.