THE EFFECT OF EDUCATION, TRAINING AND MOTIVATION ON PERFORMANCE THROUGH COMPETENCE IN MEMBERS OF THE DIRECTORATE OF DRUG INVESTIGATION OF THE RIAU ISLANDS REGIONAL POLICE

The purpose of this study is to determine and analyze the effect of education, training and motivation on performance through competence in members of the Directorate of Drug Investigations of the Riau Islands Regional Police. The method used in this study was a causal model survey method, data collection using questionnaires and distributed to 118 respondents. Statistical data analysis using SEM-PLS (Sructural Equation ModellingPartial east Square) and using path analysis to test relationship patterns that reveal the influence of variables on other variables, both direct and indirect influences assisted by Smart PLS Ver 4.0 software. The results in the study showed that education directly has a positive and significant effect on employee competence with a p-value of 0.000 < 0.05, training directly has a positive and significant effect on competence with a p-value of 0.000< 0.05, competence directly affects positive and significant performance with a p-value of 0.002 < 0.05, education directly affects positive and significant performance with p-value of 0.316 < 0.05, training directly has a positive and significant effect on performance with a p-value of 0.562 < 0.05, motivation directly affects performance with a pvalue of 0.000 < 0.05, performance directly affects positive and significant through educational mediation on employee kineja with a p-value of 0.000 < 0.05, performance of positive and significant influence through mediation of training media on competence with p-value 0.019 < 0.05, positive and non-signific influence performance through motivational mediation on competence with p-value 0.03326 <0.05.


INTRODUCTION
Human resources are the assets of an organization to achieve its mission and success. For this reason, quality human resources are needed in the group. Quality human resources want to ensure the success of the institution's mission. To achieve this mission requires education, training, and encouragement to improve quality human resources who are able to perform obligations in accordance with the mission of the organization.
One way that can be done in an effort to improve member performance is through education, training and motivation programs for members to achieve the expected performance. By providing this program, it is hoped that it can improve the quality or performance of members. One way to solve a case that ultimately increases people's sense of security is through education and training, although it is not the only factor that affects the performance of members.
Based on the pre-research conducted, researchers obtained a reflection of the increasing performance of members of the Ditresnarkoba Polda Kepri, this can be seen from the work capacity, work quality, insight, productivity, cooperation, work initiatives of members, and the influence of training attended by members which in this case improves the quality of work of Ditresnarkoba members in disclosure of drug cases in the jurisdiction of the Kepri Regional Police. The increase in timely settlement of cases rom the total number of cases per year is seen Z ( Kinerja) 0,340 0.814

Table 3 Cronbach's Alpha
Sourced from the information presented in table 6 above, it can be seen that the cronbach alpha number of all latent variables Above 0.7. These results present that each variable has met cronbach alpha as a result it can be concluded that the entire variable has a large degree of reliability. Data analysis using the SEM approach. The analysis tool used in analyzing SEM models and hypothesis testing using Partial Least Square (PLS) with SmartPLS 3.0 software. The data analysis used in this study is path analysis. Partial Least Squares (PLS) evaluation of models based on predictive measures with nonparametric properties (Ghozali 2010, p. 24).

.2 PLS Outer Mod
Check the structural model by looking at the number R-squared. Evaluation of the model utilizing PLS begins by looking at the R-squared for each dependent latent variable. The Rsquared number transformation can be used to account for whether some independent latent variable has a substantial impact on the dependent latent variable. In this research, the structural form is: Table 4 presents the R-square Adjusted figures presenting that the magnitude of the influence of X1, X2 and X3 on Y is worth 0.752 or 75.2% this means that the influence of X1, X2 and X3 on Y can be considered a strong influence. While the magnitude of the influence of X1, X2, X3 and Y on Z worth 0.548 or 54.8% this means that the influence of X1, X2, X3 and Y on Z on lecturer performance can be considered weak influence.

Table 4 Test Godness of Fit -Inner Model (Structur Model)
The proposed direct influence hypothesis can be tested from looking at the magnitude of the T-statistic number. Because PLS does not take into account normality and dissemination of information, so PLS uses a nonparametric test to ascertain the significance level of the path coefficient, where the t-statistical number obtained by carrying out the bootstraping algorithm on SmartPLS 3. 0. The benefit is to ensure whether or not the assumptions proposed are accepted. The hypothesis will be accepted if the statistical t-number exceeds 1. 96 Ghozali, ( 2012  1. The results of the first hypothesis test, namely the effect of X1 on Y, resulted in a T-count worth 4,477 with a significant p-value of 0.000. These results present that X1 has a positive and significant impact on Y., this is due to the t-count number (4,477) > the t-statistical number (1.96) and the significant figure 0.000 < 005. Then H1 In accepted. 2. The results of the second hypothesis test, namely the influence of X1 on Z, resulted in a Tcount worth 6,289 with a significant p-value of 0.000. These results present that X1 has a positive and significant impact on Z., this is due to the t-count number (6,289) > the tstatistical number (1.96) and the significant number 0.000 < 0.02. Then H2 In accepted. 3. The results of the third hypothesis test, namely the effect of X2 on Y, resulted in a T-count worth de 2016 with a significant p-value of 0.000. These results present that X2 has a positive and significant impact on Y., this is due to the T-count number (2,016) > the tstatistical number (1.96) and the significant number 0.002 < 0.05. Then H3 In accepted. 4. The results of the fourth hypothesis test, namely the effect of X2 on Z, resulted in a T-count worth 2,332 with a significant p-value of 0.000. These results present that X2 has a positive and significant impact on Z., this is due to the T-count number (6,377) > the t-statistical number (1.96) and the significant figure 0.002 < 0.05. Then H4 In accepted. 5. The results of the third hypothesis test, namely the influence of X3 on Y, resulted in a Tcount worth 0.722 with a significant p-value of 0.471. This result presents that X3 has no significant effect on Y., this is due to the T-count number (0.722) < the t-statistical number (1.96) and the significant number 0.011<0.05. Then H5 Received. positive and significant impact on Y., this is due to the t-count number (4,477) > the tstatistical number (1.96) and the significant figure of 0.000 < 0.05. Then H1 In accepted. 2. The results of the second hypothesis test, namely the effect of X1 on Z, resulted in a Tcount worth 6,289 with a significant p-value of 0.000. These results present that X1 has a positive and significant impact on Z., this is due to the t-count number (6,289) > the tstatistical number (1.96) and the significant figure 0.000 < 0.02. Then H2 In accepted. 3. The results of the third hypothesis test, namely the effect of X2 on Y, resulted in a T-count worth de2,016 with a significant p-value of 0.000. These results present that X2 has a positive and significant impact on Y., this is due to the t-count number (2,016) > the tstatistical number (1.96) and the significant figure of 0.002 < 0.05. Then H3 In accepted. 4. The results of the fourth hypothesis test, namely the effect of X2 on Z, resulted in a Tcount worth 2,332 with a significant p-value of 0.000. These results present that X2 has a positive and significant impact on Z., this is due to the t-count number (6,377) > the tstatistical number (1.96) and the significant figure 0.002 < 0.05. Then H4 In accepted. 5. The results of the third hypothesis test, namely the effect of X3 on Y, resulted in a T-count worth 0.722 with a significant p-value of 0.471. These results present that X3 had no significant effect on Y., it 6. This is due to the T-count number (0.722) < the T-Statistical number (1.96) and the significant figure 0.471> 0.05. Then H5 Received.

5.2.SUGGESTIONS
1. It is better for the Riau Islands regional police to pay more attention to the education of members through competence in order to improve the performance of police members in the Directorate of Investigation Department. 2. It is better for the Riau Islands regional police to pay more attention to training members through competence in order to improve the performance of police members in the Directorate of Investigation and Development. 3. It is better for the Riau Islands regional police to pay more attention to the motivation of members through competence in order to improve the performance of police members in the Directorate of Investigation Department.