Winter Journal: IMWI Student Research Journal e-ISSN: 2723-8709 Vol.
No.
January - July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable Intan Masito.
Siti Safaria* Institut Keuangan Perbankan dan Informatika Asia Perbanas Jakarta.
Indonesia Email: siti.
safaria@perbanas.
Abstrak This study aims to determine the effect of work-life balance and leadership style on turnover intention through job satisfaction as an intervening variable.
The case study for this research was conducted at PT.
Permodalan Nasional Madani.
Jember Branch.
The basis for this study is the issue of turnover intention, or the desire of employees to leave the company, which is one of the challenges currently faced by PT.
Permodalan Nasional Madani.
Jember Branch.
As we know, a high number of turnover intentions greatly affects the development of the company's business.
The data used in this study were obtained from 391 employees who filled out a questionnaire provided by the researcher, with a deadline of 5 days.
The results of this study indicate that a good work-life balance improves job satisfaction.
However, even with a good work-life balance, turnover intention cannot be reduced.
High job satisfaction can reduce turnover intention.
A good leadership style improves job satisfaction.
However, even with a good leadership style, turnover intention cannot be reduced.
Job satisfaction mediates work-life balance in reducing turnover intention.
Job satisfaction also mediates leadership style in reducing turnover intention.
Kata kunci : worklife balance, leadership style, job satisfaction, turnover intention INTRODUCTION PT Permodalan Nasional Madani, abbreviated as PT PNM, was established based on Government Regulation of the Republic of Indonesia No.
38 of 1999, dated May 25, 1999, concerning the participation of State Capital of the Republic of Indonesia for the establishment of Company Companies (Perser.
in the context of the development of cooperatives, small & medium enterprises.
The purpose of the establishment of PT PNM (Perser.
is outlined in Article 2 of the Government Regulation.
According to the Deed of Establishment of PT PNM No.
1, dated July 1, 1999, made before Notary Ida Sofia.
SH, the company's primary purpose is to implement and support government policies and programs in the field of economy and national development, particularly in the empowerment and development of small and medium-sized business cooperatives, applying the principles of Limited Liability Companies.
This aligns with the company's vision to become a leading financial institution by offering services, empowerment, and added value creation for ultra-micro, micro, and small business actors in a sustainable manner (Abror & Muharam, 2024.
Avien et al.
, 2025.
Kurniawati.
Febrianti.
Suryani.
Rizky, & Dewi, 2.
In accordance with the Decree of the Minister of Finance No.
487/KMK.
017/1999,
dated October 13, 1999.
PT PNM was appointed as the Coordinating State-Owned Enterprise (SOE) to distribute and manage 12 credit schemes from the former Bank Indonesia Liquidity Credit (KLBI) program until 2025.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable Throughout its business journey.
PT PNM has encountered numerous challenges, both internal and external.
According to Virgo .
, the internal environment consists of factors within the company that describe its strengths and weaknesses.
The internal environment can be divided into competencies, core competencies, and resources (Nilasari, 2.
On the other hand, the external environment consists of conditions or events outside the company that affect its growth, which is divided into the general environment and the industrial environment (Kadar et al.
, 2.
In the book Strategic Management and Leadership by Kadar et al.
, it is explained that a company can maintain its competitive advantage when it is able to focus on its resources.
Good internal environmental management helps the company achieve its goals more effectively and efficiently.
Therefore, understanding the internal environment and its impact on company performance is crucial (Alfiani et al.
, 2.
Furthermore, organizational structure, organizational culture, human resources, information systems, and business processes are also integral parts of the internal environment (Alfiani et al.
, 2.
One of the essential resources discussed in this research is Human Resources (HR).
According to Amelia et al.
HR is considered a driving force for an organization and is regarded as an asset to the company, requiring proper management and Human resources are critical components of a company, alongside capital and technology.
Effective Human Resource Management (HRM) is necessary to maintain productivity and ensure high performance within the organization.
HRM includes various processes such as recruitment, training, development, compensation, and other HRrelated activities to create productive employees.
HRM is supported by several elements, including entrepreneurs, employees, and leaders (Amelia et al.
, 2.
A significant obstacle in managing human resources is Turnover Intention.
According to Ardan et al.
, turnover intention is the tendency or intention of an employee to voluntarily leave their job.
Ideally, the turnover intention rate should not exceed 10% per year.
However.
PT PNM is experiencing a turnover intention rate exceeding 10%, particularly at the Jember Branch, which significantly impacts the companyAos business performance and target achievements.
The turnover intention rate spikes from July to December, consistently surpassing the 10% mark in every second semester from 2022 to 2024.
This surge occurs because the gap in fulfilling the yearly target is concentrated in the second semester, thus leading to a higher turnover intention rate.
Several studies have been conducted regarding turnover intention.
Shujat et al.
conducted research on private banking sector companies in Karachi.
Pakistan, which found that work-life balance has a small effect on job satisfaction and turnover Mazarei et al.
conducted research on physical education teachers in Bushehr.
Iran, and found a positive and significant relationship between servant leadership style and positive commitment, attitude, and work behavior.
Liden et al.
studied 71 restaurants in Chicago, demonstrating that servant leadership can enhance employees' positive behavior, improve performance, and generate positive followers for Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* the company.
Suifan et al.
also found that work-life balance does not negatively affect turnover intention.
Hafid et al.
conducted a study in the Food & Beverage division of Hotel Indonesia in Jakarta, which found that work-life balance negatively affects turnover intention.
Setiyanto et al.
examined the relationship between job satisfaction and turnover intention in an industrial estate in Batam City and concluded that job satisfaction does not affect turnover intention.
Further studies such as those by Amir .
and Prayogi et al.
emphasized the role of servant leadership in encouraging prosocial behavior and reducing turnover intention, while Ramma et al.
showed that servant leadership significantly positively affects job satisfaction.
Research by Anggoro et al.
revealed that servant leadership does not have a negative or significant effect on turnover intention at PT Bayyas Bio Fuels.
From the research above, it can be concluded that various factors influence an individualAos decision to leave or change jobs.
This study will delve deeper into the factors that influence the high turnover intention rate at PT PNM, particularly at the Jember Branch.
Interviews with several regional managers, branch human capital managers, heads of the Mekaar unit, and Account Officers (AO.
revealed that the most significant factors influencing the decision to resign include overtime working hours, lack of autonomy in performing tasks, inability to express opinions, strict superiors, and a gap between AOs and the head of the Mekaar unit.
This research will focus on these factors, specifically as a case study at PT PNM Jember Branch.
Another factor supporting this research is the fact that PT PNM is a financing company with a majority of female employees, mostly between the ages of 18Ae 25 years.
This demographic has not been adequately explored in previous studies.
The problem formulation in this study examines whether the implementation of work-life balance influences employee job satisfaction and turnover intention at PT Permodalan Nasional Madani Jember Branch, and whether work-life balance affects turnover intention through job satisfaction.
Additionally, the study investigates whether servant leadership style influences job satisfaction and turnover intention, including its indirect effect on turnover intention through job satisfaction.
Lastly, it explores whether job satisfaction itself has a direct effect on employee turnover intention.
The objectives of this study are to analyze the effect of work-life balance on job satisfaction and turnover intention, including its indirect effect through job satisfaction, to examine the influence of servant leadership style on job satisfaction and turnover intention, both directly and indirectly through job satisfaction, and to analyze the effect of job satisfaction on turnover intention at PT Permodalan Nasional Madani Jember Branch.
RESEARCH METHOD
Research is an art of systematic investment to seek information and knowledge about a particular topic or subject.
In general, research is carried out to find out the reality and facts that have not been revealed (Rosyidah.
, et al.
, 2.
The most difficult thing Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable to do in a study is to make a research design.
According to Khotari in Sudarmanto.
, et al .
, in making a research design, several things must contain, including: .
What things will be researched, .
Why the research will be carried out, .
Where will the research be conducted, .
What data is needed to conduct research, .
Where the data is obtained from, .
How long it takes to conduct research, .
What is the research sample, .
What techniques are used in data collection, .
How to analyze the data, .
What kind of research report will be presented.
Sari.
, et al .
also explained that research is carried out to solve a problem based on science using a scientific approach (Scientific Approac.
which in turn will result in scientific methods .
cientific metho.
There are two methods in conducting scientific research, namely the qualitative method, the quantitative method and the combination method (Rosyidah.
, et al.
, 2.
Quantitative method means a research approach based on the quantity of data so that analysis can be carried out formally and rigidly or statistically.
Based on the relationships between the variables, quantitative methods can be descriptive, correlated and associative (Ali.
, 2.
The descriptive quantitative method measures the level of a single variable in a population or its sample.
The quantitative method of correlation and associative measures the relationship between two or more variables, so that a causal relationship between related variables will be found.
Qualitative method means research based on subjective assessments of attitudes, opinions and behaviors.
Qualitative methods are carried out to see or find out the underlying reasons for human behavior.
In this study, an in-depth analysis is needed so that we can see people's motivation to do something.
Combination method means a mixed method between quantitative and qualitative.
this method model, there are 2 main models, namely the sequential .
and Competitor .
Method sequential itself consists of 2 methods, namely Sequential Explanatory .
he first stage is quantitative then continued qualitativel.
and vice versa Sequential Explanatory (The first stage is qualitative and then continued quantitativ.
Method Competitor .
also has 2 methods, namely Competitor triangulation .
ixed evenl.
and Concurrent Embedded .
ixed out of This research uses a quantitative approach method that is carried out systematically and accurately.
Research data can be proven and tested scientifically for truthfulness.
this study, 2 variables were used Independent .
is Worklife Balance (X.
and Leadership Style (X.
, variable Stuart T .
Turnover Intention .
esire or intention to move from the workplac.
, and variable Intervening namely Job Satisfaction.
Data Collection Methods The data collection method used in this study was by filling out questionnaires by 391 respondents.
A questionnaire is a question or statement prepared by the researcher and presented in writing to find out the opinions/perceptions of the research respondents about a variable being studied (Sudarmanto.
, 2.
The data collection technique with this questionnaire is very helpful for researchers in conducting research with a large number of samples.
In addition, the researcher also used Google Form as a medium for Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* disseminating questionnaires.
Then the results of the collected questionnaire were analyzed using the Smart PLS application.
Research Instruments The questionnaire distributed to the respondents was equipped with a choice of answers according to the likert scale with the following details:
Table 1 Likert Scale Remarks Score Strongly Agree (SS) Agree (S) Disagree (KS) Disagree (TS) Strongly Disagree (STS) Skala likert is one of the popular scales used in research.
However, many misinterpret the use of likert scales because not all scales whose answer choices are itemized .
temized scale.
are likert scales (Simamora.
, 2.
Measurement Scale The measurement scale of a variable can be seen in terms of its mathematical According to Sudarmanto.
, .
there are several types of measurement scales, namely: .
Nominal Scale, .
Ordinal Scale, .
Interval Scale, .
Ratio Scale.
What is used in this study is the Interval Scale, this is in accordance with the theory explained by Budiaji.
that the Likert scale that measures individual attitudes by using scores on each question is the interval measurement scale.
Instrument Validity Testing According to Anggraini.
, .
, a validity and reliability test must be carried out in a study.
The validity test is to measure how valid the data to be measured More simply, a validity test is carried out to measure what you want to measure.
The indicators in a questionnaire can be said to be valid if they count > r table.
On the other hand, if the value of r is calculated < r table, then the questionnaire indicator is declared Validity testing techniques in the research using SPSS software.
Instrument Reliability Testing The reliability test was carried out to measure the extent to which the data used in the study provided consistency over time (Anggraini.
, 2.
The reliability calculation of the questionnaire can be done if it has passed the validity test.
The size used in this reliability test is Cronbach Alpha or also known as alpha coeficient.
The range of alpha coefficient values ranges from 0 .
o reliabilit.
to 1 .
erfect reliabilit.
Data Management Analysis The data in this study uses path analysis (Path Analysi.
with a structural equation model or Structural Equation Modelling (SEM) software Smart Partial Least Squares (Smart-PLS) version 3.
0 The purpose of this study is to examine the influence of Worklife Balance and leadership style towards turnover intention employees with job Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable satisfaction as an intervening variable.
Definition of SEM according to Harahap.
& Pd.
is a set of statistical techniques used to assess a relatively complex relationship between variables compared to using linear regression analysis.
SEM
analysis consists of 3 activities, namely: .
checking the validity and reliability of the instrument .
onfirmatory factor analysi.
, .
testing the inter-variable relationship model .
ath analysi.
, .
get a model suitable for prediction .
tructural model and regression Furthermore.
Harahap.
, & Pd.
explained the advantages and disadvantages of using Smart-PLS software, including:
Disadvantages of Smart-PLS : Advantages of Smart-PLS:
Functions to test the relationship between variables in the same way as other statistical Powerful because it is not based on various assumptions The sample used was relatively small even with a fairly complex research model.
This cannot be done by other statistical software Using a random duplication method so that it does not affect the normality of the data and minimal sample use.
Able to test SEM models formatively and reflectively even with different indicator measurement scales.
Smart-PLS is only capable of reading data Excel in the form of CSV.
In this study, 2 data analysis methods were used, namely descriptive data analysis and inferential analysis.
Descriptive analysis serves to simplify information from the data that has been collected to be easier to understand, it can be in the form of graphs or tables (Hartinah.
, 2.
Meanwhile, inferential analysis is used to generalize the sample data that has been successfully collected in a limited amount (Fridayani.
, 2.
Descriptive Analysis Descriptive statistical analysis is the first step in the data analysis process before more in-depth inferential statistical analysis is carried out (Hartinah.
, 2.
Some commonly used techniques in statistical analysis include: mean, median and mode as well as the size of the spread such as range, variance and standard deviation (Hartinah.
The results of descriptive analysis can provide useful insights related to distribution and tendencies.
However, the results of this analysis must be used with caution because more in-depth analysis still needs to be done.
Variable measurement score statistics The variable measurement score functions to provide a value to each variable using an instrument in the form of questions that must be answered by the respondent.
The variable measurement in this study uses the interval method, which is a variable measurement method that has a clear distance on each data (Dahri.
Inferential Analysis According to the theory explained by Yulianto.
, inferential analysis is a conclusion based on data processing using a sample of a population where the conclusion can be representative of the population as a whole.
Statistical analysis methods used in inferential statistics include t-test.
Anova.
Anacova.
Regression Analysis.
Path Analysis.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* Structural Equation Model (SEM) and other analysis methods depending on the purpose of the research.
In this study, the analysis method used is the Partial Least Square (PLS) method version 3.
PLS is a data processing method in the Structural Equation Model (SEM).
The consideration is based on the decision to use PLS because in this study a latent variable is measured with a reflective variable.
Furthermore .
Yulianto.
explained thatinferential statistics are divided into 2, namely:
Parametric statistics, which are measures about parameters.
In this case, it is to measure the population that will be used in the research.
The measurement scale used in parametric statistics is an interval scale or ratio.
Non-Parametric Statistics, which are statistics that do not care about the value of a parameter in the research population.
These non-parametric statistics are used when the parametric statistics are no longer consistent so that they are not bound by the distribution model and the relatively small number of samples.
The data used in this statistical method is ordinal scale or nominal size scale.
RESULTS AND DISCUSSION
Research Analysis This study uses the help of SmartPLS to analyze the relationship between latent variables measured using several indicators, as well as to test the causal relationship between variables in a research model.
According to Hair et al.
, the PLS-SEM analysis consists of two main stages, namely the evaluation of the measurement model (Outer mode.
and structural model evaluation (Inner mode.
Outer Model Outer model or measurement models are used to evaluate the relationship between latent variables .
and indicators that measure them.
Main objectives of the evaluation Outer model is to ensure that the indicators used in the research are valid and reliable in representing the constructs being studied.
According to Hair et al.
, the evaluation Outer model aims to ensure that the research instrument is able to measure latent variables appropriately so that the results of the analysis obtained can be trusted.
Evaluation Outer model is done through the following tests.
Convergent Validity Convergent validity indicates the extent to which the indicators used in a construct have a high correlation with each other.
Convergent validity can be measured through two main indicators, namely:
Outer Loading Outer loading indicates the degree of relationship between the indicator and the measured construct.
Criteria according to Hair et al.
Loading value Ou 0.
The higher the value Outer Loading So the greater the contribution of these indicators in explaining latent variables.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026
The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention
Through Job Satisfaction as An Intervening Variable Table 2 Outer Loading Leadership
Style
job satisfaction Worklife Balance 0,753 0,759 0,782 0,847 0,870 0,864 0,613 0,921 0,885 0,849 0,930 0,894 0,918 0,929 0,931 0,900 0,870 0,873 0,911 0,912 0,885 0,834 0,902 0,878 0,890 0,847 0,816 0,905 0,863 0,913
0,883
0,920
0,808
0,889
0,899
0,861
0,858
0,833
0,763
0,878
0,869
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* 0,894 0,812 0,891 Source : SmartPLS Data Processing In the statement the x2.
1 indicator is still below 0.
7, so it needs to be removed so that it can be declared valid in a convergent manner Figure 1 Outer Loading Source: SmartPLS Data Processing Table 3 Outer Loading Phase 2 Leadership Style job satisfaction Worklife Balance 0,753 0,759 0,782 0,847 0,870 0,864 0,923 0,885 0,851 0,931 0,896 0,920 Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026
The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention
Through Job Satisfaction as An Intervening Variable 0,930 0,933 0,902 0,871 0,870 0,912 0,911 0,885 0,832 0,902 0,877 0,890 0,845 0,815 0,905 0,863 0,913 0,883 0,920 0,808 0,889 0,899 0,861 0,858 0,833 0,763 0,878 0,869 0,894 0,812
0,891
Source : SmartPLS Data Processing All variables have a value Outer Loading above 0.
70 so that all statements of these indicators are declared valid convergently.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* Figure 2 Outer Loading Phase 2 Source: SmartPLS Data Processing .
Average Variance Extracted (AVE) AVE shows the ability of constructs to explain the variance of its indicators.
The criteria according to Hair et al.
are Ou 0.
50 AVE value, meaning that the construct is able to explain more than 50% of the variance of the indicators used.
Table 4 AVE
Average Variance Extracted (AVE) Leadership Style
0,798
job satisfaction 0,732 turnover intention 0,781 Worklife Balance 0,662 Source : SmartPLS Data Processing In this study, all variables had an AVE value above 0.
50 so that the construct used met the requirements for convergent validity.
Discriminatory Validity Discriminant validity aims to ensure that a construct is completely different and does not overlap with other constructs in the research model.
According to Hair et al.
, the validity of discrimination can be tested through two main methods.
Fornell-Larcker Criterion Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable This method compares the square root of the AVE of a construct with the correlation between other constructs.
The root of AVE must be greater than the correlation between variables, if the value of diagonal (OoAVE) is higher than the correlation between constructs then it can be concluded that the construct has good discriminant validity.
Table 5 Fornell Larcker Criterion
Leadership Style
job satisfaction turnover intention Worklife Balance Leadership
Style
0,893
0,780
0,855
-0,164
-0,232
0,732
0,825
Worklife Balance 0,884 -0,175 0,834 Source : SmartPLS Data Processing The results of the study show that all constructs meet this criterion so that the validity of the discriminant is met.
Heterotrait-Monotrait Ratio (HTMT) HTMT is a newer method and is considered more sensitive in detecting discriminant validity issues.
Criteria according to Hair et al.
, a good HTMT value < 0.
The results of the analysis showed that the entire HTMT value between variables was below 0.
90 so that it can be concluded that the model has good discriminant validity.
Reliability Test Reliability is used to determine the level of consistency of indicators in measuring a construct.
According to Hair et al.
, reliability can be measured using:
Cronbach's Alpha Cronbach's Alpha is used to measure the internal consistency of an indicator in a construct with the value criterion Cronbach Alpha Ou 0.
Composite Reliability Composite Reliability is considered more accurate in the PLS analysis because it takes into account the contribution of each indicator with the criteria Composite Reliability Ou 0.
The results of the study showed that all variables had a value Cronbach Alpha and Composite Reliability above 0.
70 so that it can be concluded that the research instrument has excellent reliability.
Inner Model Inner model or structural models are used to analyze the relationships between latent variables in the research model.
Inner model Answer whether the relationship between variables is significant or not.
According to Hair et al.
, the evaluation Inner model is done through the following tests.
Coefficient of Determination (R-Squar.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* The R-Square value on the job satisfaction variable is 0.
This means 8% of the variation in job satisfaction can be explained by the variables of worklife balance and leadership style, the remaining 25.
2% are influenced by other variables that were not studied in the research model.
R-Square value on variable turnover intention by 055 .
ery wea.
That means 5.
5% variation turnover intention explained by the Worklife Balance, leadership style, and job satisfaction, the remaining 94.
5% were influenced by other variables outside the research model.
f-Square F-Square .
A) or effect size is a measure used to determine how much influence each independent variable has on the dependent variable.
Based on the results of the analysis effect size .
A), it can be concluded that Worklife Balance has the greatest influence on job satisfaction, while leadership style has a moderate influence on job satisfaction.
Meanwhile, the direct influence of the variable Worklife Balance and the leadership style has no effect on turnover intention.
This shows that job satisfaction plays an important intervening variable in the relationship between Worklife Balance, leadership style, and turnover intention.
Q-Square Q-Square (QA) is a measure used to determine the predictive ability of the research model against endogenous variables.
If the value of QA is positive, then the model has predictive relevance.
Based on the results of the analysis Q-Square, it can be concluded that the research model has predictive relevance, because the QA value on the job satisfaction variable and turnover intention greater than zero.
Hypothesis Test The hypothesis test aims to find out whether the relationship between variables in the research model has a significant influence or not.
According to Hair et al.
hypothesis testing in PLS is carried out using the Bootstrapping, which is a resampling technique used to estimate the statistical distribution of the research sample.
Bootstrapping It allows researchers to obtain t-statistical and p-value values so that they can determine whether the relationship between variables is statistically significant.
The t-statistical value (>1.
and p-value (<0.
according to the criteria of Hair et al.
showed that the research hypothesis was accepted .
Table 6 Direct Impact Test Original Sample (O) Sample Mean (M) Standard Deviation (STDEV) T Statistics (|O/STDEV|) P Values 0,380 0,385 0,064 5,982 0,000 0,036 0,031 0,076 0,465 0,642 Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Leadership Style -> Job Satisfaction Leadership Style -> The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable Turnover Intention Job Satisfaction -> Turnover Intention Worklife Balance -> Job Satisfaction balance -> -0,294 -0,296 0,082 3,561 0,000 0,547 0,542 0,059 9,332 0,000 0,041 0,044 0,085 0,479 0,632 Source : SmartPLS Data Processing Table 7 Indirect Influence Test Leadership Style -> Job Satisfaction -> Turnover Intention Worklife Balance -> Job Satisfaction -> Turnover Intention Original Sample (O) Sample Mean (M) Standard Deviation (STDEV) T Statistics (|O/STDEV|) P Values -0,112 -0,114 0,038 2,962 0,003 -0,161 -0,160 0,048 3,328 0,001 Source : SmartPLS Data Processing Influence Worklife Balance on Job Satisfaction The results of the analysis showed that the path coefficient value was 0.
547, the t-statistic was 9.
332, and the p-value was 0.
A path coefficient of 0.
547 indicates that it has a positive influence on job satisfaction, which means that it is getting better Worklife Balance The balance between work life and personal life of employees is the higher the level of employee job satisfaction.
The t-statistic value of 9.
332 is greater than 1.
96 and the p-value of 0.
000 is smaller than 0.
05, so the relationship between Worklife Balance and statistically significant job satisfaction.
Thus it can be concluded that Worklife Balance have a positive and significant effect on job satisfaction, so the research hypothesis is accepted.
(H1 accepte.
Influence Worklife Balance against Turnover Intention The results of the analysis showed that the path coefficient value was 0.
041, the t-statistic was 0.
479, and the p-value was 0.
The t-statistic value of 0.
479 is smaller 96 and the p-value of 0.
632 is greater than 0.
05, so the relationship between Worklife Balance and turnover intention is not statistically significant.
Thus it can be concluded that Worklife Balance has no significant effect on turnover intention, so the research hypothesis was rejected.
(H2 rejecte.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 Intan Masito.
Siti Safaria* The Effect of Job Satisfaction on Turnover Intention The results of the analysis showed that the path coefficient value was -0.
294, the t-statistic was 3.
561, and the p-value was 0.
The path coefficient of -0.
294 shows that job satisfaction has a negative influence on turnover intention, which means that the higher the employee's job satisfaction level, the lower the employee's desire to leave the The t-statistic value of 3.
561 is greater than 1.
96 and the p-value of 0.
is less than 0.
05, so that the relationship between job satisfaction and turnover intention statistically significant.
Thus, it can be concluded that job satisfaction has a negative and significant effect on turnover intention, so that the research hypothesis is accepted.
(H3
The Influence of Leadership Style on Job Satisfaction The results of the analysis showed that the path coefficient value was 0.
380, the t-statistic was 5.
982, and the p-value was 0.
The path coefficient of 0.
380 indicates that leadership style has a positive influence on job satisfaction, which means that the better the leadership style applied, the higher the level of employee job satisfaction.
The t-statistic value of 5.
982 is greater than 1.
96 and the p-value of 0.
000 is less than 0.
05, so the relationship between leadership style and job satisfaction is statistically significant.
Thus, it can be concluded that leadership style has a positive and significant effect on job satisfaction, so the research hypothesis is accepted.
(H4 accepte.
The Influence of Leadership Style on Turnover Intention The results of the analysis showed that the path coefficient value was 0.
036, the t-statistic was 0.
465, and the p-value was 0.
The t-statistic value of 0.
465 is smaller 96 and the p-value of 0.
642 is greater than 0.
05, so the relationship between leadership style and turnover intention is not statistically significant.
Thus, it can be concluded that leadership style does not have a significant effect on turnover intention, so the research hypothesis is rejected.
(H5 rejecte.
Mediation of Job Satisfaction on Influence Worklife Balance against Turnover Intention The coefficient value is -0.
161, t-statistic is 3.
328, and p-value is 0.
negative coefficient indicates that Worklife Balance can lower turnover intention through increased job satisfaction.
Since the t-statistic > 1.
96 and the p-value < 0.
05, this indirect influence is significant, so job satisfaction mediates the relationship between Worklife Balance and turnover intention.
(H6 accepte.
Mediation of Job Satisfaction on the Influence of Leadership Style on Turnover Intention The coefficient value is -0.
112, t-statistic is 2.
962, and p-value is 0.
Negative coefficients indicate that leadership style can decrease turnover intention through increased job satisfaction.
Since the t-statistic > 1.
96 and the p-value < 0.
05, this indirect influence is significant, so job satisfaction mediates the relationship between leadership style and turnover intention.
(H7 accepte.
Winter Journal: IMWI Student Research Journal.
Vol.
No.
January Ae July 2026 The Effect of Work-Life Balance and Leadership Style on Employee Turnover Intention Through Job Satisfaction as An Intervening Variable CONCLUSION The study concluded that work-life balance and a good leadership style have been shown to increase employee job satisfaction.
However, neither factor directly decreases turnover intention, as employees' decisions to leave the company are more influenced by other factors such as career opportunities, compensation, and work pressure.
Job satisfaction has been shown to be an effective mediator in the relationship between worklife balance and leadership style in reducing turnover intention.
Satisfied employees tend to be more committed to staying with the company.
The recommendations provided include improving the performance reward system, offering more flexible work scheduling, enhancing persuasive leadership communication, and evaluating the compensation system to make it more transparent and performance-based.
This study has limitations, such as homogeneous respondent characteristics, reliance on subjective perceptions for variable measurement, and a focus on only one branch unit of the company, meaning that the results cannot be widely generalized.
REFERENCE