We use multivariate data analysis to recognize the association between different motivation va The family structure riables. The analysis has shown the variables in a scatter plot and quantifying the strength of association using correlation analysis. An association is established, both empirically and theoretically, therefore we pursued to obtain a regression model. This model, used to predict the value on entrepreneurial intention (outcome), given the values  The family structure on the motivations to switch job if employed variables (predictors). Table 6, the p-value for the Levene’s test for equality of variance is 0.000, which is less than 0.05. Thus, equality of variances is not assumed. Table 7 depicts that the F-value is 3.933 and the degrees of freedoms are 7 and 188. The p-value of the test is 0.000, which is less than 0.05 and the eta-squared value of 0.128, which is less than 0.15,hence, at least one pair of means differ significantly. Therefore there is a need to identify the pair that differs significantly. In this case, the researcher has  The family structure performed posthoc tests. Based on Kruskal Wallis Test, the p-value of this test is 0.000, which is less than 0.05. Thus, at least one reason differs in terms of motivation towards self-employment rather employment. Based on Kruskal Wallis Test (Table 5) for the mean ranks, obviously, the degreed students are looking for a more challenging career in terms of expanded scope of work, more responsibilities and change in career path rank much higher compared to the students who looking to join a more established and stable organisations. On the other hand, 82.20% students feel that they need to escape from unfavourable office environment such as not happy with peers, bosses, office politics and etc. Based on Table 3.1, this supports students interest, whereby 97.4 cumulative percent and mean rank shown in table below of probability of considering the new business start-up.>