## THE TESTING OF ENTREPRENEUR INTENTION MODEL OF SMK STUDENTS IN SPECIAL REGION OF YOGYAKARTA: RESULTS

With sample data as many as 202 respondents, validity and reliability testing were done. The tested validity was factor validity through convergent validity. The testing of convergent validity was obtained from model measurement data of each variable (measurement model), this testing was done to determine the validity of each estimated indicator, by measuring dimension from the concept that was tested in research. If each indicator had critical ratio that was bigger than two times standard error, indicated that indicator validly had measured what had to be measured in the provided model. Regression quality value indicated that critical ratio that was bigger than two times standard error meant that all items in research were valid toward each research variable. The reliability of the collected data was then tested used the method of Cronbach Alpha, helped by SPSS program. Each factor in this research was reliable because had Alpha that was bigger than 0.6. According to Nunnally, reliability coefficient value that was suggested was minimally 0.6.

Model compatibility testing result used chi-square, CMIN/DF, GFI, AGFI, RMSEA, TLI, and CFI was summarized as follows. Table 1 indicates that the planned model was wholly fit, because after the compatibility was tested, the result of the value of GFI, AGFI, TLI, and CFI compared with the value of structural model equality reference was good. The scoring criteria of the value of p-value, CMIN/df (relative x2), the value of RMSEA were good compared with cut off value, helped by chi-correct. Wholly, Chi square value with probability 0.119>0.05 indicated overall goodness-of-fit. It meant the model that described the tendency influence to take risk, entrepreneurship attitude, subjective norm, and self efficacy toward entrepreneurship intention was fit with the empirical data in the field.

The complete model causality testing result in this research was as follows. Regression quality result of latent inter-variable that was often called as estimation of loading factors or lambda value could be used to analyze variable causality testing. Based on the significance of CR (Critical Ratio) value with probability value (p)=0.05. Causality testing regression quality result was as follows. The more explanation of that regression quality evaluation analysis could be analyzed and explained that tendency variable to take risk influenced self efficacy significantly with significance level as much as 0.000, smaller than probability value <0.05. Entrepreneurship attitude variable influenced entrepreneurship intention significantly with significance level as much as 0.000, smaller than probability value <0.05. Subjective norm variable influenced entrepreneurship intention significantly with significance level as much as 0.000, smaller than probability value <0.05 that meant that hypothesis 1 was accepted. Self efficacy variable influenced entrepreneurship intention significantly with significance level as much as 0.000, smaller than probability value <0.05. So, hypothesis 2 was accepted. The amount of the influence of each latent variable, standardized direct effect, standardized indirect effect, and standardized total effect, was summarized in table 3 as follows. The amount of the influence of each latent variable: standardized direct effect, standardized indirect effect, and standardized total effect, indicated that the risk tolerance had influence as much as 0.163 toward entrepreneurship intention. Variable that had direct influence that was bigger than the other variable toward entrepreneurship intention was subjective norm; that was 0.492. The amount of variable contribution simultaneously toward the other variable was summarized in table 4 as follows. Based on table 4 above, it was known that entrepreneurship attitude variable, subjective norm, and self efficacy had role as much as 41.2% toward entrepreneurship intention. Variable risk tolerance had role as much as 32.2% toward self efficacy. Based on survey result using open questionnaire, it was known that students’ willing to be entrepreneurs was caused by several reasons; those were continuing or developing parents’ business, free or did not want to be commanded by the bosses (business owners), not time-bound if working with others. Meanwhile students did not want to be entrepreneurs because the changed income risk, needed capital, inexperienced, prestige/position in career. itat on

**Table 1. Goodness of Fit**

Index | Cut off Value | Result | Model evaluation |

Chi square | Approx 0 | 187,674 | Good |

Probability | > 0,05 | 0,119 | Good |

CMIN/DF | < 2,00 | 1,131 | Good |

GFI | > 0,90 | 0,916 | Good |

RMSEA | < 0,08 | 0,025 | Good |

AGFI | > 0,90 | 0,894 | Marginal |

TLI | > 0,90 | 0,992 | Good |

CFI | > 0,90 | 0,993 | Good |

**Table 2. Regression Weight Evaluation**

Variabel | Estimate S.E. C.R. P |

Self efficacy <— Risk tolerance Entrepreneurship Intention <— Entrepreneurship Attitude Entrepreneurship Intention <— Subjective norm Entrepreneurship Intention <— Self efficacy | ,648,076 8,503 ***,252,057 4,442 ***,693,112 6,195 ***,169,038 4,439 *** |

**Table 3. Direct Effect, Indirect Effect and Total Effect**

Variabel | Direct | Indirect | Total |

Risk Tolerance ^ Self efficacy | 0,567 | – | 0,567 |

Risk Tolerance ^ Entrepreneurship Intention | 0,163 | 0,163 | |

Entrepreneurship Attitude^ Entrepreneurship Intention | 0,295 | 0,295 | |

Subjective norm ^ Entrepreneurship Intention | 0,492 | 0,492 | |

Self efficacy ^ Entrepreneurship Intention | 0,287 | ‘ | 0,287 |

**Table 4. Square Multiple Correlation**

Independent Variable | Dependent Variable | Effective Contribution |

Entrepreneurship Attitude Subjective norm Self efficacy | Entrepreneurship Intention | 41,2% |

Risk Tolerance | Self efficacy | 32,2% |

*Figure 2. Path Model*