Employee Green Behavior on generation X and Y Millennial

Employee Green Behavior (EGB), which is an important component of organizational sustainability. This study aims to analyze the effect of Green Recruitment, Green Training and Employee Green Behavior on Environment Performance in two groups of generation X and generation Y employees. The paradigm in this study is Positivism with the type quantitative research and using survey methods. The study population was employees at PT PT Wira Cipta Perkasa with a total sample of 100 people. The data analysis used was the Structural Equation Model with Smart PLS software. The results of the study prove that Green Recruitment and Green Training have an effect on Environment Performance through Employee Green Behavior in Generation X. Whereas in Generation Y Millennia it is proven that Green Recruitment has an effect on Environment Performance through Employee Green Behavior but Green Training has no effect.


INTRODUCTION
Environmental performance is the company's performance to create a green environment (Suratno et al., 2006). Environmental performance is one of the important steps a company takes in achieving business success. The more the company contributes to the environment, the better the company's image in the eyes of the community. Based on field data, it was proven that there was an increase in the amount of organic and non-organic waste from 2016 to 2018 at PT Wira Cipta Perkasa/WCP, namely from 57,528 in 2016, 60,295 in 2017 and 62,946 in 2018. Interviews were conducted with related company managers. with factors that affect the company's environmental performance and obtained 3 variables, namely Green Recruitment, Green Training, Employee Green Behavior.
Employee Green Behavior is an important component of organizational sustainability. In improving environmental performance the company instills an environmentally friendly attitude towards employees.
Based on company data, there was a decline in Green Behavior, as in the following Table 1.
Researchers also conducted a pre-survey on 20 people related to Green Recruitment, Green Training and the survey results can be seen in the Table 2. Green recruitment is a paper-free recruitment process that minimizes environmental impact. For the recruitment process, applications can be used through online media such as email, online application forms, websites, or telephone or video-based interviews.
The pre-survey results show that the results of Green Recruitment are still not optimal regarding the recruitment process. Green Training is one part of Green Human Resources management that is important in companies to build employee awareness of the environment by implementing behaviors such as reducing waste, saving energy, reducing paper use and other things that aim to familiarize employees with environmental awareness. Implementation of environmentally friendly training (Training), carried out through training programs conducted through online media by means of e-learning training where employees can gain knowledge and materials with easy access through online methods. It also aims to apply environmentally friendly behavior to employees through training programs specifically designed by the company for the needs of improving the quality of employee performance. The results of the pre-survey showed that there were still problems related to the training method which made the training still not optimal. One of the challenges for companies in this globalization era is that companies must manage workers across generations (generation X and generation Y). Each generation has its own characteristics in behaving environmentally conscious in its work environment. Based on the above phenomena, a research was carried out to analyze the effect of Green Recruitment, Green Training and Employee Green Behavior on Environment Performance in the group X and Generation Y employees.

Green Recruitment
Green recruitment is a process of recruiting new talents who are aware of the ongoing process, environmental systems and roots of environmental conservation. Green recruitment ensures that new talents are familiar with green practices and environmental systems that will support effective environmental management within the organization (Stringer, 2009). Meanwhile, according to Prachi (2011), Green recruitment is a paper-free recruitment process that minimizes environmental impact. Recruitment is usually done through online media such as websites, e-mails, online application forms and other applications. The dimensions of Green Recruitment (Astuti and Wahyuni, 2018) are Job Design and Recruitment Process.

Green Training
The definition of Green Training according to Roy and Therin (2008) is training that is carried out with the aim of instilling environmentally friendly behavior in employees in the work environment. The aim of Green training is to train employees to collect waste data and increase knowledge of green environmental management. The dimensions of green training (Jabbour, 2015) are Implementation and Evaluation. Ones and Dilchert (2012) states that Employee Green Behavior is a measurable individual behavior that contributes to or reduces environmental impact in the context of work. Thomas (2016) emphasizes that Employee Green Behavior includes activities to save energy, use resources efficiently, reduce waste, recycle and care for the environment. The dimensions of Employee Green Behavior (Thomas et al., 2015) are Environmental Concern, Task Control and Efficiency.

Environment Performance
Environmental performance is a measurable result of the environmental management system, in an organization that is related to the control of its environmental aspects. Khsan (2008) explains that environmental performance assessment is based on environmental policies, environmental targets and environmental targets. According to Sukirman (2012), the dimensions of environmental performance consist of: Anticipation, Measurement of environmental performance and concern for the environment.
Several previous studies related to Employee Green Behavior and Environmental Performance are research conducted by Janaka et al. (2017) which states that practices that Human Resources Management can integrate with environmental management activities to improve environmental performance. Norton et al. (2015) also examined the factors influenced by Employee Green Behavior for organizational sustainability. The importance of Green Recruitment and Green Training in companies that will affect the performance environment in the company was studied by Deepak and Prakash (2015). Several studies (Sakher et al., 2017;Lenny and Ahmad, 2018) prove the connection of Green Recruitment, Green Training on Sustainable Corporate Performance. Other research also discusses the effect of Employee green behavior on Environmental sustainability (Qaisar et al., 2018) and Green Recruitment and Green Training on Employee green behavior (Shweta and Shruyi, 2018;Brenton et al., 2016).
The hypotheses in this study are: H1a: Green Recruitment affects the Employee Green Behavior of generation X employees H1b: Green Recruitment affects the Employee Green Behavior of Generation Y employees H2a: Green Training affects Employee Green Behavior for X generation employees H2b: Green Training affects Employee Green Behavior in Y generation employees H3a: Green Recruitment affects Environment Performance for X generation employees H3b: Green Recruitment has an effect on Environment Performance in Generation Y employees H4a: Green Training affects the Environment Performance of generation X employees

RESEARCH METHODS
This study uses a quantitative approach with survey methods conducted on employees of PT Wira Cipta Perkasa with a sample size of 100 people.
The statistical analysis in this study used SEM-Smart PLS.
All variables in the study were measured using a Likert scale of 1-5.

Evaluation of Measurement Model (Outer Model)
The evaluation of the measurement model (outer model) is carried out to determine the validity and reliability of the indicator and its latent variables. The measurement model has been analyzed based on PLS-SEM with the help of Smart PLS 3.0. For assessment of measurement models, factor loading, composite reliability, Cronbach's alpha, average extracted variance (AVE), and Discriminant validity. Figures 1 and 2 and Table 1 show the results of the measurement model.
The loading factor value used in this study is >0.6, so if there is a loading factor value <0.6 in the calculation result of the measurement model (outer model), it will be excluded from the model.
The results of the calculation of the measurement model using.
SEM PLS version 3.0, then look at the loading factor value, there are several indicators with the loading factor in each research variable that has met the value >0.6 as can be seen in Figures 1  and 2 and Table 3.
Tables 4 and 5, show the loading factor value, Cronbach's alpha value, composite value and AVE. mentioned that Cronbach alpha more than 0.7 (α > 0.9) is very good. In the current study, more than 0.9 were excellent. In addition, AVE must be equal to or more than 0.5 and the composite reliability value must be 0.7 or higher (Fornell and Larcker, 1981;Hair et al., 2014). In this study, both AVE and composite were more than acceptable ranges for both X and Y generation. So, the measurement of the structural model was continued.

Evaluation of the Structural Model (Inner Model) or Hypothesis Testing
Assessment of the structural model After the assessment of the measurement model, the structural model is analyzed with the help of Smart PLS 3.
Evaluation of the structural model (inner model) or testing the hypothesis in this study through the steps of evaluating the path coefficient value, evaluating the R2 value, measuring the effect size f2, validates the overall structural model with the Goodness of Fit Index (GoF), and performs predictive relevance (Q2) testing. Testing the structural model of this research, obtained the path coefficient results through the calculate SmartPLS version 3.0 bootstrapping shown in Figures 3 and 4 and Table 6.
Evaluating the value of the path coefficient, based on the results of calculations using calculate SmartPLS version 3.0, the bootstrapping results are obtained by the path coefficient that describes the strength of the relationship or influence between constructs/variables as shown in Table 5. Test results of the X generation Path Coefficient and Generation Y.
Assessment of effect size levels using Cohen's f2. According to the defined criteria, the f2 values were equal to 0.0, 0.15, and 0.35 representing weak, moderate, and strong effect sizes. Table 6 provides a summary of the effect sizes. The results given in Table 7 show that in generation X, the Employee Green Behavior (0.928) construct has a strong effect size, the Green Recruitment construct (0.159) has a moderate effect size, while Green Training has a weak effect (0.032). In contrast, in generation Y, the Employee     Based on the calculation of predictive relevance (Q2), it was obtained 0.957 for Generation X, while for Generation Y it was 0.926. In this research model, endogenous latent variables have a predictive relevance (Q2) value greater than 0 (zero) so that the exogenous latent variables as explanatory variables are able to predict the endogenous variables, namely Environment Performance, or in other words prove that this model is considered to have predictive relevance. well, for generation X and generation Y.

Hypothesis Test
The results of statistical calculations of the effect of the independent variable on the dependent variable are presented in Table 5, presenting the results of the significance test of the structural model. The simultaneous effect of the Green Recruitment (X1), Green Training (X2), Employee Green Behavior (Y1), and Environment Performance (Y2) variables can be done by calculating the f/f statistic using the formula below. using the formula as below.

Explanation of Each Hypothesis in Group 1 (Generation X)
Hypothesis 1a: Green Recruitment has a positive and significant effect on Employee Green Behavior Hypothesis testing with the PLS approach produces a path coefficient of 0.596 with a t statistics of 4.461 which is greater than the value of t table = 1.96, and the value of P = 4.4E-06 which is smaller than α = 0.05. The coefficient value is positive, meaning that the Green Recruitment variable (X1) has a positive effect on the Employee Green Behavior (Y1) variable by 59.6%. Thus the hypothesis H1 in this study which states that "Green Recritment has a positive and significant effect on Employee Green Behavior" is accepted. According to employees, the most dominant thing in green recruitment is that environmental criteria have been included so that this way the company can get employee candidates who are aware of green in their work environment.

Hypothesis 2a: Green Training has a positive and significant effect on Employee Green Behavior
Hypothesis testing with the PLS approach produces a path coefficient of 0.344 with t statistics of 2.496, which is greater than the value of t table = 1.96, and the value of P = 0.013 which is smaller than α = 0.05. The coefficient value is Positive, which means that the Green Training variable has a positive and significant effect on Employee Green Behavior by 34.4%. Thus the H2 hypothesis in this study which states that "Green Training has a positive and significant effect on Employee Green Behavior" is accepted. The results showed that in the implementation of Green Training, generation X employees felt that every employee had the same opportunity to get training about the environment. This is what ultimately encourages employees to behave green.
Hypothesis 3a: Green Recruitment has a positive and significant effect on Environment Performance.
Hypothesis testing with the PLS approach produces a path coefficient of 0.242 with a t statistics of 2.907 which is greater than the value of t table = 1.96, and the value of P = 0.004 which is smaller than α = 0.05.
The coefficient value is positive, which means that the Green Recruitment (X1) variable has a positive effect on the Environment Performance (Y2) variable by 24.2%. Thus the hypothesis H3 in this study which states that "Green Recritment has a positive and significant effect on Environment Performance" is accepted.

It has been proven that what needs to be improved in Green
Recruitment is the dimension of the recruitment process, namely that the management should include the criteria for environmental knowledge in job vacancies.

Hypothesis 4a: Green Training has a positive and significant effect on Environment Performance
Hypothesis testing with the PLS approach produces a path coefficient of 0.09 with t statistics of 1.226 smaller than the value of t table = 1.96, and the value of P = 0.221 which is greater than α = 0.05. The coefficient value is Positive and insignificant for Environment Performance. Thus the hypothesis H4 in this study which states that "Green Training has a positive and significant effect on Environment Performance" is rejected. According to generation X employees, it turns out that Green Training does not affect Environment Performance. The thing that needs to be improved related to Green Training is the evaluation process, namely the use of environmental training results by employees so that it will lead to an increase in Environment Performance.
Hypothesis 5a: Employee Green Behavior has a positive and significant effect on Environment Performance Hypothesis testing with the PLS approach produces a path coefficient of 0.661 with t statistics of 6.425 greater than the value of t table = 1.96, and the value of P = 3.1E-10 which is smaller than α = 0.05. The coefficient value is positive, meaning that the Employee Green Behavior (Y1) variable has a positive effect on the Environment Performance (Y2) variable by 66.1%. Thus, the hypothesis H1 in this study which states that "Employee Green Behavior has a positive and significant effect on Environment Performance" is accepted. Based on the research results, it was found that the most influential dimension on Environment Performance is anticipation in which employees understand the environmental impact caused by the company so that various programs can be carried out for environmental conservation.
Hypothesis 6a: Green Recruitment has a positive and significant effect on Environment Performance through Employee Green Behavior Hypothesis testing using the PLS approach produces a path coefficient of the effect of Green Recruitment on Environment Performance through Employee Green Behavior with a significant effect with a path coefficient of 0.376. Since the two direct effects that form are significant, there is sufficient empirical evidence to accept.
H 1 : which states that Green Recruitment has a positive and significant effect on Environment Performance through Employee Green Behavior.
The positive coefficient indicates that the higher the Green Recruitment, the higher the Environment Performance by increasing the Employee Green Behavior path. The Employee Green Behavior variable is a partially mediating variable in the relationship between Green Recruitment and Environment Performance, because the coefficient of indirect effect of Green Recruitment on Environment Performance through Employee Green Behavior is greater than the coefficient of direct influence between Green Recruitment on Environment Performance and both have significant effects.
Partially mediation means that the Employee Green Behavior variable plays a role in strengthening the relationship between the Green Recruitment Variable and the Environment Performance. Thus the hypothesis H1 in this study which states that "Green Recruitment has a positive and significant effect on Environment Performance through Employee Green Behavior" is accepted. Generation X employees stated that the most influential dimension of Employee Green Behavior is motivation in which employees understand the actions of safety needs, security at work.
Hypothesis 7a: Green Training has a positive and significant effect on Environment Performance through Employee Green Behavior.
Hypothesis testing with the PLS approach produces a path coefficient of the effect of Green Training on Environment Performance through Employee Green Behavior with a significant effect with a path coefficient of 0.227. Since the two direct effects that form are significant, there is sufficient empirical evidence to accept H 7 : which states that Green Training has a positive and significant effect on Environment Performance through Employee Green Behavior. The positive coefficient indicates that the higher the Green Training, the higher the Environment Performance by increasing the Employee Green Behavior path. The Employee Green Behavior variable is a perfect mediation variable in the relationship between Green Training and Environment Performance, because the coefficient of direct influence between Green Training on Environment Performance is insignificant.Perfect mediation means that the Green Training variable does not explain the diversity of Environment Performance Variables, but the Employee Green Behavior variable which explains the diversity of Environment Performance. Variables in the relationship between Green Training and Environment Performance. Thus the hypothesis H7 in this study which states that "Green Training has a positive and significant effect on Environment Performance through Employee Green Behavior" is accepted.
The results of the study state that the dimensions that must be improved regarding Employee Green Behavior are control related to tasks, namely minimizing the impact of environmental damage.

Explanation of Each Hypothesis in Group 2 (Generation Y (Millennial))
Hypothesis 1b: Green Recruitment has a positive and significant effect on Employee Green Behavior Hypothesis testing with the PLS approach produces a path coefficient of 0.699 with t statistics of 8,292, which is greater than the value of t table = 1.96, and the value of P = <0.000 which is smaller than α = 0.05. The coefficient value is positive, meaning that the Green Recruitment variable (X1) has a positive effect on the Employee Green Behavior (Y1) variable by 59.6%. Thus, the H1 hypothesis in this study which states that "Green Recruitment has a positive and significant effect on Employee Green Behavior" is accepted.
According to Y generation employees regarding Green Recruitment, the dimensions of the work design have been carried out well. That the company has entered Environmental Criteria and the company's commitment to the environment during the employee recruitment process.
Hypothesis 2b: Green Training has a positive and significant effect on Employee Green Behavior.
Hypothesis testing with the PLS approach produces a path coefficient of 0.166 with t statistics of 1.547 smaller than the value of t table = 1.96, and the value of P = 0.122 which is greater than α = 0.05. Thus the H2 hypothesis in this study which states that "Green Training has a positive and significant effect on Employee Green Behavior" is rejected.
It is proven in this study that the dimension that needs to be improved in relation to green training is evaluation. Training evaluation is very important to see the effectiveness of the training implementation.
Hypothesis 3b: Green Recruitment has a positive and significant effect on Environment Performance.
Hypothesis testing with the PLS approach produces a path coefficient of 0.057 with t statistics of 0.813 smaller than the value of t table = 1.96, and the value of P = 0.417 which is greater than α = 0.05. Thus the H1 hypothesis in this study which states that "Green Recruitment has a positive and significant effect on Environment Performance" is rejected.
Employees stated that the recruitment process must be improved, namely in finding candidates who have competence in the field of environmental management.
Hypothesis 4b: Green Training has a positive and significant effect on Environment Performance.
Hypothesis testing with the PLS approach produces a path coefficient of 0.187 with t statistics of 2.655, which is greater than the value of t table = 1.96, and the value of P = 0.008 which is smaller than α = 0.05. The coefficient value is positive, meaning that the Green Training (X2) variable has a positive effect on the Environment Performance (Y2) variable by 18.7%. Thus the H1 hypothesis in this study which states that "Green Training has a positive and significant effect on Environment Performance" is accepted.
The results of the study stated that the implementation of the Green Training had gone quite well. All employees are given equal opportunities in environmental training.
Hypothesis 5b: Employee Green Behavior has a positive and significant effect on Environment Performance.
Hypothesis testing with the PLS approach produces a path coefficient of 0.829 with t statistics of 11.506 greater than the value of t table = 1.96, and the value of P ≤ 0.000 which is smaller than α = 0.05. The coefficient value is positive, meaning that the Employee Green Behavior (Y1) variable has a positive effect on the Environment Performance (Y2) variable by 82.9%. Thus, the hypothesis H1 in this study which states that "Employee Green Behavior has a positive and significant effect on Environment Performance" is accepted.
Related to Environment Performance that needs to be improved is concern for the environment. Employees Increase employee awareness of the applied employee management system. The dominant dimension of Employee Green Behavior, according to the opinion of generation Y employees, is that there is a good intention of positive behavior for employees to behave in an environmentally conscious manner so that this encourages behavior control to behave in green.
Hypothesis 7b: Green Training has a positive and significant effect on Environment Performance through Employee Green Behavior.
Hypothesis testing with the PLS approach produces a path coefficient of the effect of Green Training on Environment Performance through Employee Green Behavior which has no significant effect with a path coefficient of 0.138. Since the two direct effects that form an indirect effect are one of which is insignificant, namely the effect of Green Training on Employee Green Behavior, there is sufficient empirical evidence to accept H 0 : which states that Green Training has an effect on Environment Performance without going through Employee Green Behavior.
So that the Employee Green Behavior variable is not a mediating variable in the relationship between Green Training and Environment Performance. Thus the H 1 hypothesis in this study which states that "Green Training has a positive and significant effect on Environment Performance through Employee Green Behavior" is rejected.
Based on the research results, it is found that the waste management is not optimal so that it needs attention from the company.

Conclusion
On Generation X Employees • Green Recruitment and Green Training have an effect on Employee Green Behavior • Green Recruitment and Employee Green Behavior affect the Environment. Performance while Green Training has no effect on Environment Performance • Green Recruitment and Green Training affect Environment Performance through Employee Green Behavior.

• On Generation Y Employees
• Green Recruitment has an effect on Employee Green Behavior, while Green Training has no effect • Green Training and Employee Green Behavior have an effect on Environment Performance while Green Recruitment has no effect on Environment Performance • Green Recruitment affects Environment Performance through Employee Green Behavior, while Green Training has no effect.

Suggestion
For companies • Improve the Green Recruitment system by including criteria for candidate employees who are aware of environmental management and behave in green. • Improve the training system related to training evaluation so that training effectiveness can be achieved • Improve the Employee Green Behavior system related to task control to increase employees' sense of concern so that it will have an impact on employee behavior in environmental management • Improve the company's existing environmental performance appraisal system so that it will increase employee awareness of the employee management system.
For future researchers, it is hoped that the existing research framework will be developed and add other variables that are thought to affect Employee Green Behavior and environmental performance.