Journal of Economics and Business

This study aims to get an overview of the Influence of training factors and locus of control on self-efficacy and work ability of employees in three-star hotels in East Java. The reputation in this study is the three-star hotel industry in East Java, which totals 20 hotels and all employees at the three-star hotel in East Java, with a sample size of 501 employees. The analysis technique used is to use path analysis. Based on the analysis of the results and the hypothesis proposed previously that the variable trainer power, training material, training methods, training facilities, and locus of control partially have a significant influence on the variable self-efficacy is acceptable/true. While the hypothesis which states that the trainer variable, training material, and training methods and locus of control partially influence the workability variable can also be accepted, only one hypothesis that cannot be accepted is the effect of training facility variables on workability.


Introduction
At present, the improvement of the quality of human resources (HR) has become a demand for all business and non-business organizations.The faster the flow of changes due to intense competition, both on a local, regional, and global scale requires organizations to continue to make changes so as not to lag behind with their competitors.The main factor that determines an organization still exists in the face of competition is quality human resources.For each organization, the most important asset that must be considered by management is the assets of humans and organizations (Simamora, 2012).The importance of the role of HR for an organization requires management to continue to carry out HR development activities.HR development is an effort to improve the quality and competence of human resources in the organization (Ruky, 2013).One effort to develop human resources includes training to improve skills in carrying out work.According to Zwick (2010), employee training conducted by companies is one and most important measure in improving and maintaining productivity levels.
Various studies show that effective training has a significant effect on improving work capacity.The study conducted by Endayani (2015) Training as a human resource development tool relates to improving employee skills and increasing the ability to meet the demands of an ever-changing work situation.The success of the training held by the company can be found through the opinions given by the training participants through a questionnaire at the end of the training (post-test).The question in the questionnaire concerns the participants' satisfaction with the training as a whole, namely the trainer, the material presented, the content of the material, the materials provided, the training environment (space, rest time, food and air temperature), training methods, and management commitment.
According to Mangkunegara (2013), new employees need orientation training, and they need to understand the objectives, rules, and work guidelines that exist in the organization of the company.Besides that, they need to understand their obligations, rights, and duties according to their work.So that training factors, such as: training materials, trainers (trainers), training methods, and training facilities and locus of control effect the effectiveness of training in determining employee self-efficacy.While self-efficacy, as indicated by individual beliefs about their ability to take part in the training, will determine the work ability of employees in three-star hotels in East Java-Indonesia.Through this research, it is expected to be able to know the training factors that effect the selfefficacy and work ability of employees in three-star hotels in East Java-Indonesia.

Training
Understanding training, according to Simamora (2012), is a systematic process of changing the behavior of employees in a direction to improve organizational goals.In training created an environment where employees can obtain or learn specific attitudes, abilities, skills, knowledge, and behaviors related to work.Increasing the career development of human resources can be done by training.Many experts expressed various opinions regarding training.Dessler (2010) stated, "training is a business through the process of teaching new employees or old employees, by providing knowledge of the basic skills needed to run a job."Mangkuprawira (2011) states, "a process teaches knowledge and skills that are assisted, as well as the attitude so that employees are skilled and able to carry out their responsibilities better, according to standards."Training can help employees acquire new knowledge and skills to meet the needs and achievement of the organization.According to Bangun (2012), training is a process to maintain or improve employee skills to produce effective work.
From the various opinions of experts above it can be concluded that the notion of training is a planned process that is used to change attitudes, knowledge or behavior, improve abilities and skills, which are needed not only new employees but old employees also need the training to complete the work in a way that effective.Training is a short-term educational process where non-managerial employees learn their knowledge and skills to improve their effectiveness and work productivity to help achieve organizational goals.Training as a tool for developing human resources is related to improving employee skills and increasing work capabilities to meet the demands of an everchanging work situation.Training must be able to increase employee effectiveness and improve employee satisfaction.

Locus of Control (LoC)
According to Robbins (2012: 139), Locus of Control is the rate at which individuals believe that they are determinants of their destiny.Likewise, Locus of Control in training participants is considered to influence the ability to transfer the skills they have just learned.Locus Of Control is one of the personality variables, which is defined as the reason for the self-confidence of each individual in being able to control destiny.Locus of control refers to the extent to which an individual connects his life events to external factors or other people (external or to their (internal) deposition.Locus Of Control is a term used to refer to the individual's perception of personal control, especially concerning control over important results.In the same quote, Benson defines Locus of Control as a person's belief in how the individual attempts to achieve the desired results.
According to Robbins (2012: 138), Internal Locus of Control are individuals who believe that they are the holders of control over whatever happens to them, said to have an Internal Locus of Control.Individuals with internal Locus of Control have a perception that the environment can be controlled by themselves so that they are able to make changes according to their wishes, including in applying the results of training obtained in their work.Because individuals feel they can control themselves, there is a tendency to have high confidence that they are able to absorb the content of the training program so that they can then apply the results of the training to work.Individual internal factors which include work ability, personality, work actions related to work success, selfconfidence, and failure of individual work are not caused due to relationships with work partners.

Self-Efficacy
Self-efficacy theory includes self-regulation and self-reflection.Self-reflection capability is that people reflect back on the actions/experiences of certain events and then process cognitively how much they believe in completing future tasks -acting as a theoretical basis for self-efficacy (Bandura: 2005).Bandura strongly emphasizes that self-efficacy is the most important psychological mechanism of self-influence.The formal definition of self-efficacy that is often used is the Bandura statement regarding personal judgment or beliefs about "how well someone can take the necessary actions to deal with a prospective situation" (Bandura: 2005).Capability assessment is very important for individuals, individuals who overestimate their abilities when carrying out activities that cannot be achieved as a result he has difficulty reducing his credibility and suffering failure, while individuals who overestimate their abilities will limit themselves from beneficial experiences, for which individuals must obtaining self-knowledge pleases with abilities, physical skills, and skills to deal with situations encountered daily.

Work Ability
Ability is intended as the ability of employees to carry out work.This ability contains various elements such as manual and intellectual abilities, even to the personal qualities possessed.These elements also reflect the education, training, and abilities demanded following work details (Zainun, 2013).Ability here refers to an individual's capacity to work on various tasks in a job (Gibson et al. 2012), namely ability is a trait (innate or learned) that allows a person to carry out an action or mental or physical work.
Ability consists of two main elements, namely intellectual and physical abilities.Intellectual ability is needed to do mental activities, namely activities that are complex and require thinking.While physical abilities lead to the abilities needed to perform tasks that require stamina, dexterity, strength, and similar skills.Psychologically, the ability of employees consists of potential abilities (IQ) and reality abilities (knowledge and skills).This means that employees who have an IQ above the average (110-120) with adequate education for their positions who are skilled in doing daily work, it will be easier to achieve expected performance (Mangkunegara, 2013).

Type of the Research
In social research, it is generally divided into three types, namely exploratory research, descriptive research, and explanatory research.This study aims to get an overview of the influence of training factors and locus of control on the self-efficacy and work ability of employees in three-star hotels in East Java, so referring to the three types of research, this research is included in explanatory research.While the method used in this study is the survey method.Research using survey methods is an activity to obtain factual information and facts or phenomenal explanations that exist in the object of research.
The reputation in this study is the three-star hotel industry in East Java, which totals 20 hotels and all employees at the three-star hotel in East Java.In this study, there were two sample groups, namely the sample of three-star hotels and samples of permanent employees of three-star hotels in East Java.For samples of three-star hotels using the sampling method, namely Total Sampling or Census.The researcher assumes that it is better to consider investigating all elements of the population if the population elements are relatively small and the variability of each element is relatively high (heterogeneous) (Sugiyono, 2012: 44) Thus, all 20 three-star hotels in East Java will be the subject of research.Then, determining the size of the employee sample (sample size) for star hotel employees in East Java, which numbered 3,338 people, based on the Issac and Michael formulas and found that the sample needed for N = 3,338 with a standard error of 5% obtained a total sample of 317 person.Seeing this fact, the researchers decided to take a sample of 15% of the number of three-star hotel employees in East Java, which amounted to 501 employees.

Variable of Coaching Staff
The Coaching Staff variable is to see the influence of the quality of the instructor/trainer in improving Work Ability and employee performance.This variable is measured through several indicators, namely: (1) relevance of educational background to the material provided, ( 2) mastery of Training materials, (3) ability to deliver Training materials, ( 4) appreciation of feedback and trainees, ( 5) attitude and how the trainer will deliver Training materials.

Variable of Training materials
The training materials variables are to improve the effectiveness of work skills and employee performance.This variable is measured through the following indicators: (1) the benefits of training materials on quality and quantity of work, ( 2) the relevance of the assignment of material to type of work, (3) the composition of training materials according to the field of work, ( 4) the ability to understand employee capabilities in the operational field.

Variable of Training Method
The Training Method variable is to see the effect of the Training Method in improving Work Ability and employee performance.This variable is measured through several indicators, namely: ( 1 4) the opportunity given by the trainer in developing problem analysis or alternative problem solving through case studies, and ( 5) the benefits of work practice simulations in supporting employee work.

Variable of Training Facilities
The Training Facilities variable is to see the effect of Training materials in improving Work Ability and employee performance.This variable is measured through the following indicators: (1) availability of equipment to support training activities, (2) the effect of availability of records in helping employees understand Training materials, (3) conditions of training buildings and surrounding environments, (4) conditions of lecture / training rooms in providing convenience for participants, and ( 5) complete library facilities in supporting the training process.

Variable of Locus of control
Locus of control (X1) is an individual's view of achieving conceptual success as an individual's level of confidence in perceiving that success is achieved because they have confidence, that there is a correlation between effort and success, they can manage their lives, and have self-confidence.
Instrument measurement for the locus of control used in this study used 5 statement instruments.Variable measurements using a Likert scale with a score of 1 (strongly disagree) up to 5 (strongly agree), grouping internal and external locus of control is based on the score of the answer.The higher the respondent's answer score, the more likely it is to have an internal locus of control, and vice versa if the respondent's total locus of control score is low then the respondents tend to have an external locus of control.

Variable of Self-Efficacy
Self-efficacy after training (Z) is the individual's belief in his ability to attend training and confidence in completing tasks assigned to him.The self-efficacy instrument used consists of 5 statements, using a scale of 1-5.Range of scale 1 (strongly disagree) up to 5 (strongly agree).If the total score of high respondents shows that the respondents have high self-efficacy and vice versa if the total score of respondents is low, the respondents have low selfefficacy.

Variable of Work Ability
Work Ability as an intermediate variable, namely as a dependent variable that is influenced by variables in training as well as independent variables that effect employee performance.Work Ability indicators in some question items are used to measure Work Ability variables.Indicators used to measure Work Ability variables, namely: (1) the ability to complete work on time, (2) the intensity of consultation with superiors if faced with difficulties in carrying out work, (3) physical condition of employees in supporting workloads that must be completed, ( 4) the ability to maintain attitude and behavior as a reflection of the characteristics / characteristics of the company, and ( 5) the intensity of following information and developments in the banking world.

Data Analysis
Based on the conceptual framework, the analytical technique used is to use path analysis, the steps according to Solimun (2010: 47-55) are as follows: The first step in path analysis is to design a model based on the conceptual framework of the research, the formulation as follows: The conversion of the Path model into a mathematical model becomes as follows: 2. Parameters γ (gamma), including γ1, ... .γ5 are the parameters of the influence of exogenous variables on endogenous variables self-efficacy (Z).
3. The influence of β (beta), among others, is β1, ... .β5 is the parameter of the influence of exogenous variables on endogenous variables Work Ability (Y), while β6 is the parameter for the influence of endogenous variables on other endogenous variables.4. Parameter of ε (epsilon), that is the parameter relating to errors in latent variables based on manifest variables.
The second step of path analysis is an examination of underlying assumptions.Assumptions that underlie path analysis can be divided into two, namely: 1).assumptions relating to the model, and 2).assumptions relating to testing parameter estimates and testing hypotheses.The third step in path analysis is parameter estimation and hypothesis testing based on the path coefficient, while the parameter estimation method used is a maximum likelihood, with AMOS 4.0 software.The fourth step in path analysis is checking the validity of the model through testing the Goodness of Fit for the overall model (overall model).The goodness of Fit testing for overall models aims to measure the correctness of the structure model and measurement model in an integrated manner so that it must be tested against several fitindexes.

Result Testing the Goodness of Fit Analysis Results Model
After the raw data obtained were tested for validation and reliability, then the processing and testing of path analysis models were carried out following the existing conceptual framework.In this study, the processing and testing of models were carried out using the AMOS 4.0 program.

Table 1. Results of Analysis of Direct Influence Between Variables of Research
Source: primary data In direct analysis between variables in this study, two things must be considered, namely: 1).path coefficient value (standarized value) that shows the magnitude of the influence of a variable on other variables, and 2). the level of significance of the influence of variables partially on other variables.AMOS 4.0 output can be seen from the value of the critical-ratio (c.r) which is analogous to the t-test in regression.
Value c.r. (critical ratio) compared with the value of t table with α = O.05 (i.e., ± 1.98).If the value -1.98> tvalue <1.98 then the variable partially does not have a significant effect, conversely if tv values <-1.98 or tvalue> 1.98, the variable partially has a significant effect.

Analysis of Direct Variable Training Facilities (X4) Influence on Self-efficacy (Z) and Work Ability (Y) variables
Based on Table 5, the output path analysis is obtained, which shows the direct Influence of Training Facilities (X4) variable on the variables self-efficacy (Z) and Work Ability (Y) as follows:

Analysis of Direct Variable Self-efficacy (Z) influence on Work Ability (Y)
Based on Table 7, output path analysis is obtained, which shows the direct effect of Variable Self-efficacy (Z) on Variable Work Ability (Y) as follows: Facilities (X4).Because there is no direct effect between Variable X1-X5 on Variable self-efficacy, the total value of the fifth Variable effect on Variable self-efficacy is the value of the lengsung effect itself, while the total effect of the five Variables on Variable Work Less ability is the accumulation of effects direct and indirect effects.

Overall Model Testing (Overall Mode)
Overall model testing aims to measure the correctness of structural models and measurement models in an integrated manner so that it must be tested against several fit-indexes.The model is said to be good (fit) if the development of a hypothetical model is conceptually and theoretically supported by empirical data.The following are some conformity indices and their cut-off values to test whether the model can be accepted or rejected, including: 1. Chi-Square (χ 2 ) Statistics, is the most fundamental test tool for measuring "overall fit" and is very sensitive to the size of the research sample.The greater the number of samples, the smaller the value of χ 2 , the better the model, with p-value 0.05.
2. RMSEA (Root Mean Square Error of Approximation), is an index that can be used to compensate for Chi Square Statistic in a large sample.RMSEA value <0.08 shows a "close fit" of the model based on the degree of freedom.
3. GFI (Goodness of Fit Index), is used to calculate the weighted proportion of variance in the sample covariance matrix described by the estimated covariance matrix.Vulnerable value is 0 <GFI <1, the closer to 1 the better.
4. AGFI (Adjusted Goodness of Fit Index), analogous to R2 in regression with acceptance level> 0.9 5. CMIN / DF (The Minimum Sample Discrepancy Function / Degree of Freedom), as an indicator, to measure the suitability of a model.The expected value is <2.00 6. TLI (Tucker Lewis Index), an index that compares a model tested against the baseline model.The expected value is> 0.90.
7. CFI (Comparative Fit Index), this index is not influenced by the size of the sample with the recommended value is> 0.90, the closer to 1 the better.
Table 10.Index of Goodness Of Fit for the Overall Model Source: Primary Data Based on the results of the overall model test, it appears that some of the test results show the Not good value, namely in Chi-Square (χ 2 ), RMSEA, and CMIN / DF.Not good's Chi-Square (χ 2 ) value indicates that the data differ significantly from the expected results based on the theoretical foundation used.However, this value is not very much considered in testing the correctness of a model because it is very sensitive to the number of samples, the larger the sample, the Chi-Square value (χ 2 ) will also be smaller (significant).Values from RMSEA and CMIN / DF are test values related to Chi-Square values (χ 2 )or those related to data compatibility with models/theories.So if the Chi-Square (χ 2 )value is Not good, then the RMSEA and CMIN / DF test results will also be Not good.This can be overcome by increasing the number of research samples so that the Chi-Square value (χ 2 )will be significant, as well as the values of RMSEA and CMIN / DF.From the results of the above tests it can be concluded that in general, the Goodness Of Fit test for the overall model has shown the corresponding value and the cut-of value set so that the path analysis model above can be said to be good.

Analysis of Direct Variable Coaching Staff Influence on Variable Self-efficacy and Work Ability
Based on the path analysis output that shows the direct effect of Variable Coaching Staff on Variable self-efficacy and Work Ability.The magnitude of the effect of Direct Variable Coaching Staff on self-efficacy is 0.264.From the significance test, it is obtained that the value of c.r. is equal to 2.661> 1.98 so that the Variable can partially be declared to have a significant effect.The magnitude of the influence of the Variable Coaching Staff directly on the employee's Ability is 0.199, with the value of c.r. amounting to 2.263> Based on the respondent's answers, it was shown that most respondents had internal locus of control, namely having confidence that success could be achieved and being active in responding to changes in the surrounding environment, while 40.4% had external locus of control, they were not sure they could respond to changes in the environment , so that in themselves they are more likely to be resigned to what happens, meaning that even though they have the initiative to try to apply what is obtained from the training, it seems that they still have confidence that fate can change the plans that have been made and success depends on luck.
The results of this study are relevant to the theory which states that basically locus of control is a dimension in the form of copper and internal to external or vice versa, so that an internal person can be external and an external person can be internal because of the conditions that accompany him where he lives and do activities.In line with this theory, in this study locus of control has a significant effect on Work Ability, it is possible because an internal locus of control believes that success is determined by oneself, so it is possible to apply training results to work, but when external factors influence it, for example: working patterns, existing facilities, people around him (working group) who tend to use old patterns, so the farmers may not be able to improve their abilities.

Analysis of the Direct Influence of Self-efficacy Variables on Variable Work Ability
Based on the results of the analysis Variable self-efficacy partially has a significant effect on Variable Work Ability.The amount of direct influence The variable self-efficacy on Work Ability is 0.482, with the value of c.r. amounting to 4.672> 1.98 so that this variable partially also has a significant effect on Variable Work Ability.
Seeing the results of the analysis means that the benefits of integrated pest control training support the work of employees in completing work difficulties, understanding the conditions of the work environment, giving influence to the ability of farmers.The results of this study are consistent with the opinion of Notoadmojo ( 2009) stating that among the factors that influence performance, only the ability factor can be developed which can be intervened or treated through training.

Conclusion
Based on the analysis of the results and the discussion that has been described previously, then some conclusions can be drawn as follows: ) the suitability of the Training Method used by the participant's condition, (2) the composition of the technique / Training Method, (3) the level of participant participation during the discussion program in the training program, ( efficacy) and Y (Work Ability) as endogenous variables, while X1 (Coaching Staff), X2 (Training materials), X3 (Training Methods), X4 (Training Facilities), and X5 (Locus of control) as exogenous variables.
Direct Variable Coaching Staff Effect (X1) on Self-efficacy Variables (Z) and Work Ability (Y) Based on the table above, output path analysis is obtained, which shows the direct effect of Coaching Staff variable (X1) on the variables self-efficacy (Z) and Work Ability (Y) as follows: Table 2. Direct Effect of Variable Coaching Staff (X1) on Work Ability (Y) Variables and Self-efficacy (Z) Source: primary data From the table above, it can be seen that the direct Influence of Coaching Staff variable (X1) on self-efficacy (Z) is 0.264.And the significance test is obtained that the value of c.r. is equal to 2.661> 1.98 so that the variable can be partially declared significantly.The direct Influence of the Coaching Staff variable (X1) on Work Ability (Y) is 0.199, with value c.r. Amounting to 2.263> 1.98 so that this variable partially also has a significant influence on the variable self-efficacy (Y).From the hypothesis proposed in this study which states that the Coaching Staff variable (X1) partially has a significant influence on the variables self-efficacy (Z) and Work Ability (Y) is acceptable/true.The direct relationship model between Coaching Staff variables (X1) on the variables self-efficacy (Z) and Work Ability (Y) is converted into a mathematical equation as follows: Variable Influence of Training Material (X2) on Self-efficacy (Z) Variables and Work Ability (Y) Based on table 3, the output path analysis is obtained which shows the direct Influence of the Training materials (X2) variable on the variables self-efficacy (Z) and Work Ability (Y) as follows: , for example, resulted in the conclusion that training covering Training Methods and Training Materials affected Work Ability.Training methods are influential in improving employee capabilities.The training methods used include instruction, internships, coaching, written modules that influence work ability.

Table 3 .
DirectFrom the above table, it can be seen that the magnitude of the Influence of the Direct Training materials (X2) variable on self-efficacy (Z) is 0.594.From the significance test, it is obtained that the value of c.r. is equal to 6,354> 1.98 so that these variables can partially be declared to have a significant Influence.The magnitude of the direct Influence of the Training materials (X2) variable on Work Ability (Y) is 0.243, with a value of c.r of 2.349> 1.98 so that this variable partially also has a significant influence on the Work Ability (Y) variable.From the hypothesis proposed in this study which states that the Training materials (X2) variable partially has a significant influence on the variables self-efficacy (Z) and Work Ability (Y) is acceptable/true.The model of the direct relationship between the Training materials (X2) variable on the variables self-efficacy (Z) and Work Ability (Y) is converted into a mathematical equation as follows: Influence of Variable Training Materials (X2) on Self-efficacy Variables (Z) and Work Ability (Y) Source: Primary Data

Influence Analysis of Direct Variable Training Method (X3) on Self-efficacy Variables (Z) and Work Ability (Y) Based
on Table4, the output path analysis is obtained, which shows the direct effect of Variable Training Method (X3) on Variable self-efficacy (Z) and Work Ability (Y) as follows: Table 4. Influence Analysis of Direct Variable Training Method (X3) on Self-efficacy Variables (Z) and Work Ability (Y) Source: Primary Data From the table above, it can be seen that the magnitude of the Influence of the Training Method (X3) direct variable on self-efficacy (Z) is 0.363.And the significance test is obtained that the value of c.r. is equal to 4,075> 1.98 so that the variable can be partially declared to have a significant Influence.The amount of direct Influence of Training Method (X3) variable on Work Ability (Y) is 0.279, with c.r value of 3.367> 1.98 so that this variable also partially has a significant Influence on Work Ability (Y) variable.

Table 5 .
Analysis of Direct Variable Training Facilities (X4) Influence on Self-efficacy (Z) and Work Ability (Y) variablesFrom the table above, it can be seen that the magnitude of the Influence of the Direct Variable Training Facilities (X4) on self-efficacy (Z) is 0.177.From the significance test, it is obtained that the value of c.r. is equal to 1,985> 1.98 so that the variable can be partially declared to have a significant Influence.The magnitude of the influence of the Variable Training Facilities (X4) on Work Ability (Y) is 0.152, with the value of c.r. amounting to 1979 <1.98 so that this Variable is partially not Influence Work Ability (Y).From the hypothesis proposed in this study, that Variable Training Facilities (X4) partially have a significant Influence on variable self-efficacy can be accepted, while the Variable Work Ability is not accepted.The direct relationship model between Variable Training Facilities (X4) to Variable self-efficacy (Z) and Work Ability (Y) is converted into a mathematical equation as followsVariable Direct Influence Analysis of Locus of control (X5) on Variable Self-efficacy(Z) and Work Ability(Y) Based on Table6, output path analysis is obtained, which shows the direct effect of Variable locus of control (X5) on Variable self-efficacy (Z) and Work Ability (Y) as follows: Table 6 Variable Direct Influence Analysis of Locus of control (X5) on Variable Self-efficacy (Z) and Work Ability (Y) Source: Primary Data From the table above, it can be seen that the magnitude of the direct Influence of the variable locus of control (X5) on self-efficacy (Z) is 0.397.From the significance test, it is obtained that the value of c.r. is equal to 6.315> 1.98 so that the Variable can be partially declared significantly.The magnitude of the direct Influence of the Variable locus of control (X5) on Work Ability (Y) is 0.269, with the value of c.r. amounting to 2.116 <1.98 so that this variable is partially not Influence Work Ability (Y).From the hypothesis proposed in this study, that Variable locus of control (X5) partially has a significant Influence on variable self-efficacy can be accepted, while the Variable Work Ability is not accepted.The model of the direct relationship between Variable locus of control (X4) to Variable self-efficacy (Z) and Work Ability (Y) is converted into a mathematical equation as follows:

Table 7 .
Analysis of Direct Variable Self-efficacy (Z) influence on Work Ability (Y) Source: Primary DataThe amount of direct effect of variable self-efficacy (Z) on Work Ability (Y) is 0.482, with the value of c.r. amounting to 4.672> 1.98 so that this variable partially also has a significant effect on Variable Work Ability (Y).From the hypothesis proposed in this study that Variable self-efficacy (Z) partially has a significant effect on the Variable Work Ability is true.The model of a direct relationship between Variable self-efficacy (Z) and Work Ability (Y) can be converted into a mathematical equation as follows: Y = 0.440 X 4 + 0.094 Analysis Exogenous Variable Indirect Influence on Endogenous Variables in Research Data regarding indirect effects between Variable Coaching Staff (X1), Training materials (X2), Training Method (X3), Training Facilities (X4), and locus of control (X5) on Variable Work Ability (Y) as follows:Table 8. Analysis Exogenous Variable Indirect Influence on Endogenous Variables in Research Source: Primery Data Based on the table above, it appears that Variable Training materials (X2) have the largest indirect effect on the Variable Work Ability (Y), which is equal to 0.284.Next are Variable Coaching Staff (X1), Training Method (X3), Training Facilities (X4), and locus of control (X5) with 0.121; 0.168; 0.079; and 0.259.

Table 9 .
Results of the Analysis of the Total Influence Between Variable Research Source: Primery Data From the data above, Training materials (X2) still provide the greatest total effect on Variable self-efficacy (Z) and Work Ability (Y), followed by locus of control (X5), Training Method (X3), Coaching Staff (X1), Training

of the Direct Influence of Variable Training materials on Variable Self-efficacy and Work Ability
Mangkunegara (2013)riable partially also has a significant effect on Variable Work Ability.From the hypothesis proposed in this study, which states that Variable Coaching Staff partially has a significant effect on Variable self-efficacy and Work Ability is acceptable/true.These results are following the opinion of As'ad (1998) and Notoadmojo (2009) that trainers / Coaching Staffan is one of the important components in training.Variable training materials partially have a significant effect on variable self-efficacy and work ability.The direct effect of Variable Training materials on self-efficacy is 0.594.From the significance test it was obtained that value c.r. is equal to 6,354> 1.98, so that the Variable can be partially declared significantly.The magnitude of the direct effect of Variable Training materials on Work Ability is 0.243, with a value of c.r of 2.349> 1.98 so that this Variable partially also has a significant effect on Variable Work Ability.Based on the results of the path analysis, it shows that Training materials, the composition of Training materials and the ease of training participants in understanding and understanding Training materials.Provide benefits to self-efficacy and work capabilities of three-star hotel employees in East Java.This is following the opinion ofMangkunegara (2013), which states that training and development material must be adjusted to the objectives achieved.

of the Direct Influence of Variable Training Methods on Variable Self-efficacy and Work Ability
Mangkunegara (2013) path analysis shows the influence of the Variable Training Method Directly on Variable selfefficacy and Work Ability Variable Training Method partially has a significant effect on Variable self-efficacy and Work Ability.The magnitude of the effect of the Direct Variable Training Method on self-efficacy is 0.363.From the significance test, it is obtained that the value of c.r ... is equal to 4.075> 1.98, so that the variable can be partially declared to have a significant effect.The influence of the Variable Training Method directly on Work Ability is equal to.0.279, with the value c.r. amounting to 3,367> 1.98 so that this Variable partially also has a significant effect on Variable Work Ability.So that the suitability of the Training Method, the composition of the technique / Training Method, the level of participant participation, the development of problem analysis through case studies, and the benefits of work practice simulations have a positive influence on self-efficacy and Work Ability of three-star hotel employees in East Java.The success of the training program is inseparable from the methods used in training.The variation of the Training Method will be able to eliminate participants' boredom.In employee education programs in three-star hotels in East Java, although in addition to using exposure methods in the classroom, group discussions, lab work with tools and field trips were also used.Mangkunegara (2013)states that the Training Method and development must be following the level of ability of employees who are participants.

of Variable Training Facilities Direct Influence on Self-efficacy Variables and Work Ability
Output path analysis shows the direct effect of variable training facilities on variable self-efficacy and work ability.Variable Training Facilities partially have a significant effect on Variable self-efficacy, while the Variable Work Ability has no effect.The effect of Direct Variable Training Facilities on self-efficacy is 0.177.From the significance test it is obtained that the value of c.r. is equal to 1,985> 1.98, so that the variable can be partially declared to have a significant effect.The magnitude of the effect of the Direct Variable Training Facilities on Work Ability is 0.152, with value c.r of 1.979 <1.98 so that this Variable is partially not Influence Variable Work Ability.Based on the results of the path analysis, it means the availability of equipment to support training activities, the benefits of equipment in helping to understand Training materials, the condition of training buildings and the surrounding environment, the conditions of training rooms to provide participants' convenience, and library facilities / reading materials in supporting training affect self-efficacy and Work Ability.

Direct Effect Analysis of Variable Locus of control on Self-efficacy Variables and Variable Work Abilities Based
on the path analysis output shows the direct effect of Variable locus of control on self-efficacy is 0.397.From the significance test, it is obtained that the value of c.r. amounting to 6.315> 1.98, so that the Variable can partially influence significantly.The amount of direct influence Variable locus of control on Work Ability is 0.243 with a value of c.r. amounting to 2.116> 1.98 so that this variable is partially Influence Variable Work Ability.
1. Variable Coaching Staff provides direct influence on Variable self-efficacy and Work Ability.The direct effect of Variable Coaching Staff on self-efficacy is 0.264.While the magnitude of the influence of the Direct Variable Coaching Staff on Work Abilities is 0.199.From the significance test obtained, the value of c.r is 2.661 and 2.263, which is greater and table 1.98 so that this variable partially has a significant effect on the two Variables.2. Variable training materials partially have a significant effect on variable self-efficacy and work ability.The amount of direct influence of variable training materials on self-efficacy is 0.594.The magnitude of the effect of the Direct Variable Training materials on Work Ability is 0.243.From the significance test, the value c.r. each of them is 6.354 and 2.349> 1.98 so that this variable partially also has a significant effect on Variable Work Ability.3. Variable training methods partially Influence self-efficacy and Variable Work Ability.The magnitude of the effect of the Direct Variable Training Method on self-efficacy is 0.363, while the effect on Work Ability is 0.279.From the significance test, it is obtained that the value of c.r. is equal to 4,075 and 3,367> 1.98, so that the variables can be partially stated to have a significant effect.4. The effect of the Direct Variable Training Facilities on self-efficacy is 0.177.From the significance test it is obtained that the value of c.r. is equal to 1,985> 1.98, so that the variable can be partially declared to have a significant effect.Whereas the magnitude of the effect of the Direct Variable Training Facilities on Work Ability is 0.152, with the value of c.r equal to 1.979 <1.98 so that this Variable partially does not significantly influence the Work Ability variable.5. Variable locus of control partially Influences self-efficacy and Variable Work Ability.The amount of direct influence Variable locus of control on self-efficacy is 0.397, while the effect on Work Ability is 0.269.From the significance test, it is obtained that the value of c.r. is equal to 6,135 and 2,116> 1.98 so that the Variable can be partially expressed significantly. 6. Variable Work Ability partially has a significant effect on variable employee performance.The amount of direct influence of variable work ability on employee performance is 0.482, with a value of c.r of 4.672> 1.98 so that this variable partially has a significant effect on the variable performance of employees.7. From the hypothesis proposed previously that Variable Coaching Staff (X1), Training materials (X2), Training Method (X3), Training Facilities (X4), and locus of control (X5) partially have a significant effect on variable self-efficacy is acceptable /right.While the hypothesis states that Variable Coaching Staff (X1), Training materials (X2), and Training Method (X3) and locus of control (X5) partially Influence Variable Work Ability can also be accepted, only one hypothesis that is unacceptable is influenced Variable Training Facilities (X4) against Work Ability.Suggestion 1. Management must always pay attention to training factors, including: Training materials, Coaching Staff (instructors), Training Methods, and Training Facilities so that the effectiveness of the training can have a positive influence on self-efficacy and Work Ability.The suitability of these training factors with established standards is an important factor so that the determined output can be achieved properly.Determination of Coaching Staff, the composition balance of training materials is more ideal as needed, a variety of Training Methods that emphasize more on practical approaches and Training Facilities that further support the absorption of Training materials must always be done.2. Management needs to constantly carry out continuous evaluation and improvement in the training process so that it is in line with the development of science and technology, as well as changes that occur in the world of plantations.