THE INFLUENCE OF MOTIVATION AND LEADERSHIP STYLE ON EMPLOYEE PERFORMANCE AT THE ENVIRONMENT AND FORESTRY SERVICE OF RIAU ISLANDS PROVINCE

This study aims to see the influence of motivation and leadership style on the performance of the Environment and Forestry Service of Riau Islands Province. The sample used was 35 respondents from 55 employees. The sampling technique used was census. The data analysis method used is multiple regression. Based on the results of the data analysis, the regression equation is Y = 36,192 + 0.132X1 + 0.602X2 + e. Based on the t test, motivation has no positive and insignificant effect on performance, with another explanation that each reduction per unit of the motivation variable (X1), the performance variable (Y) will decrease by 0.132 or the performance will decrease by 13.2%. Leadership style has a positive and significant effect on performance. This means that each addition per unit of the Leadership Style variable (X2), the performance variable (Y) will increase by 0.602 or the performance will increase by 60.2%. Based on the F test, motivation and leadership style together have an effect on performance. Based on the coefficient of determination, that 30.9% of employee performance is determined by the variable of motivation and leadership style collectively, while the remaining 69.1% is determined by other factors.


INTRODUCTION
In government organizations, good performance will be obtained from employees who work efficiently and effectively and have a high work ethic so that it will result in high productivity. Thus the quality aspect will be related to the achievement of performance which must be supported by other factors such as motivation and leadership style.
According to Kenneth N. Wexley and Gary A. Yuki (2005: 127), motivation is a process that requires energy behavior and direction. Motivation is one of the instruments in achieving organizational goals. Motivation is not only an encouragement in doing work, but how attitudes and behaviors arise as a form of responsibility in carrying out tasks or work based on feelings of enthusiasm. Motivation has a very important role in creating good performance in an organization.
Leadership factors also have a close relationship with employee performance. Leadership includes influencing other people to do more effort in a number of tasks or change their behavior (Kenneth N. Wexley and Gary A. Yuki (2005: 189). In the leadership factor, the leadership style of a leader also plays an important role in encouraging employee motivation..According to Veithzal Rivai (2003: 64), leadership style is a set of ways that leaders do to influence subordinates so that organizational goals are achieved or it can also be said that leadership style is a pattern of behavior and strategies that are liked and often applied by a leader.
According to Mangkunegara (2008: 9) in Arini Yulianita (2017: 41) states that "performance is the result of work in terms of quality and quantity achieved by employees in carrying out their duties in accordance with the dependents given to him ". Supardi (2013) in Okky Camilla Bianca (2017: 10) suggests that performance is the work that has been achieved by someone in an organization to achieve goals based on standardization or size and time adjusted to the type of work and in accordance with established norms and ethics. set.
From several studies, there is a strong relationship between motivation and leadership style on performance. Based on the research by Yohanes Susanto (2017); Adri Agustiningrum, C. Dyah S. Indrawati, Andre N. Rahmanto (2012); H.M Syarfi Hutauruk (2015) and Dicky Ari Vanjery MD (2016), motivation and leadership style together have a positive and significant effect on performance.
During its development, work motivation and leadership style are not only considered by private organizations, but also government organizations in improving employee performance. As for government organizations, one of them is the Riau Islands Province Environment and Forestry Service in improving the performance of employees who are faced with obstacles that arise, namely the inadequate understanding of the main duties and functions of employees in supporting task implementation, motivation, work quality, level of discipline, ability to complete work, lack attention and response and obedience to orders from leadership, and others. Therefore, to support the implementation of existing duties, every employee must always have a good understanding of what their main duties and duties have been and be carried out properly with full responsibility.
Based on this, shows a gap (gap) against what should or what the organization hopes to achieve. These things indicate that employee performance has not been achieved optimally. By looking at and paying attention to the descriptions above, it is deemed necessary to conduct research at the Riau Islands Province Environment and Forestry Service with the title "The Influence of Motivation and Leadership Style on Employee Performance of the Riau Islands Province Environment and Forestry Service. This study aims to see: 1. Does motivation affect the performance of the staff of the Riau Islands Province Environment and Forestry Service; 2. Does the leadership style affect the performance of the staff of the Riau Islands Province Environment and Forestry Service; 3. Do motivation and leadership style jointly affect the performance of the employees of the Environment and Forestry Service of Riau Islands Province.

IMPLEMENTATION METHOD a. Types of research
The type of research used in this research is descriptive and quantitative.

b. Location and Time of Research
The location of this research is located at the Environment and Forestry Service of Riau Islands Province. c. Population, Sample and Sampling Technique According to Arikunto (2010: 173) the population is the entire research subject. The population in this study was the Environment and Forestry Service of Riau Islands Province, amounting to 55 people. The sample according to Arikunto (2010: 109) is a part or representative of the population to be studied. Sampling for research according to Arikunto (2010: 112), if the subject is less than 100 people, all of them should be taken, if the subject is large or more than 100 people can be taken 10-15% or 20-25% or more. Based on this understanding, the sampling technique used is a census so that the sample in this study should be 55 respondents. However, when the research was carried out, as many as 20 employees were not in place because they were on duty outside the city and were carrying out studies outside the city so that the number of samples in this study was 35 respondents.

d. Operationalization of Variables
Variable operationalization is intended to explain each of the variables used which are closely related to the research to be carried out. More clearly, the operationalization of this research variable can be seen in the following description: 1. Motivation (X1) According to Hasibuan (2006: 163) in Anwar's book (2003: 34) in Aloysius Bau Kabelen (2013: 28), motivation theory has sub variables, namely: motives, expectations and incentives. The meaning is: -Motive is a stimulant of desire (want) and driving force of a person's willingness to work. Each motive has a specific goal to be achieved. -Expectancy is an opportunity given occurs because of behavior to achieve it.
-Incentives (incentives), namely motivating (stimulating) subordinates by giving gifts (rewards) to those who excel above standard achievements. -The indicators used for the motivation variables in this study are motives, expectations and incentives.

Leadership Style (X2)
According to Siagian (2002: 121) in Dicky Ari Vanjery MD (2016: 54), leadership style variables are: a climate of mutual trust, appreciation for the ideas of subordinates, taking into account the feelings of subordinates, attention to work comfort for subordinates, take into account the job satisfaction factor of subordinates in completing the tasks entrusted to him and appropriate and professional recognition of the status of subordinates.

Employee Performance (Y)
According to Handoko (1996: 10) in Dicky Ari Vanjery MD (2016: 54), the performance variables are: ability, accuracy and objectivity, scope of duties and on time. e. Types and Sources of Data Types According to Sugiyono (2010: 13-14), according to the type, data can be grouped into quantitative and qualitative data, where quantitative data is data in the form of numbers or qualitative data which is deemed to be predictive, while qualitative data is data in the form of words, sentences, schemes and pictures . f. Data source The data sources used in this study are as follows: 1. Primary data is data obtained directly from respondents through a list of questions in the form of a questionnaire. 2. Secondary data in this research is supporting data that comes from research objects, books (libraries), or other parties that provide data that is closely related to the objects and objectives of research that come from documents, books, magazines, the internet and journal. g. Data collection technique To collect data in this study, the data collection techniques used were observation, questionnaires, and documentation which are described as follows: observation, questionnaire and documentation h. Research Instruments The research instrument or questionnaire used in this study will be measured using a Likert scale, in which the respondent is given several alternative categories to choose one of the most appropriate answers. Each instrument item that uses a Likert scale has a gradient from very positive to very negative, which can be in the form of words. After the data were tabulated from the results of the questionnaire, then the instruments were tested. The instrument testing is as follows: 1). Validity Test Validity shows the extent to which a measuring instrument measures what will be measured. If the validity of a measuring instrument the higher the accuracy. So the validity test refers to the extent to which an instrument performs its function. The instrument is said to be valid if the instrument can be used to measure what you want to measure (According to Sugiyono (2008: 363). The criterion is that the instrument is valid if the correlation value (pearson correlation) is positive, and the correlation probability value [sig. (2-tailed)] <significant level (α) of 0.05. 2). Reliability Test Reliability refers to an understanding that an instrument can be trusted to be used as a data collection tool because the instrument is considered good. i. Quantitative Analysis In this study, the classical assumption test used was the normality test, multicollinearity test, and heteroscedasticity test.
1). Normality Test One of the easiest ways to see residual normality is to look at a histogram graph that compares the observed data with a distribution that is close to the normal distribution. However just looking at the histogram can be misleading especially for a small sample size.
A more reliable method is to look at a normal probability plot that compares the cumulative distribution of the normal distribution. The normal distribution will form a straight diagonal line, and the plot of residual data will be compared with the diagonal line. If the residual distribution data is normal, then the line representing the actual data will follow the diagonal line. The results of the assumption test are seen in the output of the SPSS program.

2). Multicollinearity Test
The multicollinearity test is used to determine whether or not there are deviations from the classic multicollinearity assumption, namely the linear relationship between the independent variables in the regression model.

3). Heteroscedasticity Test
In principle, the heteroscedasticity test with this method is to see the Scatterplot graph between the predicted value of the independent variable, namely ZPRED and its residue, SRESID. The basis for decision making in the Heteroscedasticity Test with the Scatterplot Graph, namely: Furthermore, the data analysis aims to test the proposed hypothesis, in which the data analysis used is quantitative analysis. In this study, simple regression testing used the SPSS program. The multiple regression model used according to Sunarto and Riduwan (2009: 108) is as follows: Y = a + b1 X1 + b2 X2 ……. + bn + Xn + e

RESULTS AND DISCUSSION a. Overview of the Environment and Forestry Service of Riau Islands Province
The Riau Islands Province Environment and Forestry Service is one of the Riau Islands Provincial government apparatus which has the main task of exercising the authority of the provincial government in the environmental and forestry sector. The tasks carried out refer to the Regional Medium Term Development Plan (RPJMD). To carry out its five-year duties and activities, the Riau Islands Provincial Forestry and Environment Agency makes a strategic plan and in carrying out its duties. The Environment and Forestry Service of Riau Islands Province is responsible to the Governor based on the vision, mission and main tasks and functions.
b. Description of Employees in the Environment and Forestry Service of Riau Islands Province.
In accordance with the data of the Head of the General and Civil Service sub-division of the Riau Islands Province Environment and Forestry Service, the number of Civil Servants (PNS) in the Riau Islands Province Environmental and Forestry Service is 55 people (not including employees of the Regional Technical Implementation Unit / UPTD). However, for this study, the number of respondents to whom the study was conducted was 35 people with a picture of education as follows Table 1. Educational Characteristics of Respondents The data above shows that the level of education is quite high. Of the number of employees, undergraduate education is 71.43% of the total respondents.

c. Analysis of Research Instrument Test Data
In research using quantitative methods, the quality of data collection is largely determined by the quality of the instruments or data collection tools used. A research instrument is said to be of quality and can be justified if its validity and reliability have been proven. The instrument test consisted of the validity test and the reliability test.
1. Validity Test Validity test is used to determine the validity or validity of an instrument. Based on the validity test on the motivational variable (X1) which was carried out using the SPSS V.25 program with the Pearson correlation method, it was calculated that out of the 12 questions, there were 2 questions that proved invalid, so that there were question item data that were excluded from the calculation, namely number 3, and 4. Further calculations show that all the questions are valid with rcount> rtable. For leadership style variables with 18 questions (X2) and performance (Y) 16 questions, all valid questions and none were issued after the calculation obtained rcount> rtable as well as positive correlation (pearson correlation) and correlation probability value (sig. (2-tailed)] <significant level (α) of 0.05 so that the instrument used can be said to be valid.

No
Type of Education Total (people) Percentage ( %)

Reliability Test
Reliability testing is to measure the constancy or constancy of an instrument in measuring what it measures. The benchmarks for interpreting the degree of reliability of the instrument obtained are in accordance with the following table: Table. 3. Reliability Test The estimation results of the reliability test conducted show the high reliability coefficient value, so the results of the questionnaire data have a good level of reliability, or in other words, the data from the questionnaire results can be trusted. The results of reliability testing for the motivation variable (X1), leadership style (X2) and performance (Y), the reliability coefficient (r) was greater than the criteria (rcriteria = 0.70).

d. Analysis Results
Data Data analysis was carried out after all data from field observations were collected which were then processed using the SPSS V.25 application which supports data analysis consisting of classical assumption tests and statistical tests. Classical assumptions that must be met are assumptions, multicollinearity, autocorrelation, heteroscedasticity, and normality.
1. Multicollinearity Testing According to Ghozali (2005) multicollinearity can be seen from the tolerance value and the opposite of Variance Inflation Factor (VIF). The cut off value commonly used to measure the absence of multicollinearity symptoms is a tolerance value> 0.10 or a VIF value <10.  Table 4. Multicollinearity Testing Based on the multicollinearity test results as shown in the table above, it is known that the tolerance value is 0.801> 0.10 and the VIF value is 1.249 <10, it is concluded that the variables of motivation and leadership style on performance do not have multicollinearity symptoms.

Autocorrelation Testing
The autocorrelation test aims to test whether in the linear regression model there is a correlation between confounding error in period T and confounding error in period T-1 (previous). If there is a correlation, it can be said that there is an autocorrelation problem. A good regression model should be free or not have autocorrelation. In this study using the Durbin Watson autocorrelation test. This test is part of the classic assumption test. The basis for the decision making of Durbin Watson's autocorrelation test is as follows:

Heteroscedasticity Testing
Heteroscedasticity test is part of the classic assumption test in the regression model. Where one of the requirements that must be met is a good regression model is that there is no symptom of heteroscedasticity. To determine whether there is heteroscedasticity in this study is to look at the scatter plot graph, if the results of the data are spread, namely above and below the zero value, the regression model is feasible because it is free of heteroscedasticity From the picture above. It is seen that there is no certain pattern in the scatter plot graph, this can be seen from the distribution of data (points) that occur randomly, both above and below the zero value on the Y axis. Then it can be concluded that the regression model used is declared good and feasible. to be used because there is no heteroscedasticity.

Normality Testing
The normality test aims to determine whether the residual results are normally distributed or not. A good regression model has a residual value that is normally distributed. The Kosmogorov Smirnov normality test is part of the classical assumption test. The basis for making decisions in the Smirnov kosmogorov test are: -If the significance value is> 0.05, the residual value is normally distributed -If the significance value <0.05, the residual value is not normally distributed Based on the results of the normality test, it is known that the significance value is 0.200> 0.05, it can be concluded that the residual value is normally distributed.

Statistical Testing
Statistically, it can be measured from the coefficient of determination, the value of the F statistic and the value of the t statistic.
a. Simultaneous Signification Test (Test F) The test is used to prove whether the independent variables (Motivation Variables and Leadership Style) together (simultaneously) have a significant effect both positive and negative on the dependent variable (performance).
From the results of the ANOVA test, it is known that the value of F count (7.147)> F table (3.27) and a significance value of 0.003 <0.05, it can be concluded that the variables of motivation (X1) and leadership style (X2) have a simultaneous effect on performance. (Y) b. Partial Significance Test (t test) The partial significance test (t test) was conducted to see the significance of the influence of the dependent variable individually and to consider other variables constant. The parameter used for the t test in this study was to compare the significance value with the 5% confidence level. From the results of data processing, it can be stated that: a. The significance value of the motivation variable is 0.427 where the result is greater than the significance value of 0.05 and is negative, meaning that the motivation variable partially has no positive and insignificant effect on the performance variable. The t value of the motivation variable (-805) <t table (1.68957), so that the motivation variable partially does not have a positive and insignificant effect on the performance variable b. The significance value of the leadership style variable is 0.001 where the result is smaller than the significance value of 0.05 and is positive, meaning that the leadership style variable has a positive and significant effect on the performance variable. The t value of the leadership style variable (3.665)> t table (1.68957), so that the leadership style variable partially has a positive and significant effect on the performance variable c. Multiple regression analysis Multiple regression analysis is a model to determine the effect of the independent variable, namely the motivation variable and the leadership style variable, on the dependent variable, namely performance. The statistical calculation in multiple linear regression analysis used in this study is to use the help of a computer program that supports data processing using the SPSS for Windows V.25 program. Based on the results of the calculations that have been done, the multiple linear regression equation is obtained as follows: Y = 36.192 + 0.132X1 + 0.602X2 + e.

Variable Motivation
The results of the analysis show that the variable regression coefficient of Motivation (X1) = -0.132, Leadership Style (X2) = 0.602 and produces a constant value of 36.192 so that the multiple linear regression equation Y = a + β1X1 + β2X2 + e or Y = 36.192 + 0.132X1 + 0.602X2 + e. This means that every reduction per unit of the Motivation variable (X1), then the Performance variable (Y) will decrease by 0.132 or the performance will decrease by 13.2%.

Leadership Style Variables
The regression coefficient value of 0.602 shows that if each additional unit of the Leadership Style variable (X2), then the performance variable (Y) will increase by 0.602 or the performance will increase by 60.2%. d). The coefficient of determination