https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
DOI: https://doi.
org/10.
38035/dijefa.
https://creativecommons.
org/licenses/by/4.
Customer Preference Analysis: Why are R1 450VA Electricity Customers not Interested in Increasing Power? (A Case Study of PT PLN Batam Customer.
Asihon Siallagan1*.
Anita Maharani2.
Maya Maria3 Universitas Terbuka.
Indonesia, yaziidbatam@gmail.
Universitas Bina Nusantara.
Jakarta.
Indonesia, anita.
maharani@binus.
Universitas Terbuka.
Indonesia, maya@ecampus.
Corresponding Author: yaziidbatam@gmail.
Abstract: This study analyzes the preferences of R1 450VA electricity customers at PT PLN Batam who are not interested in increasing their electricity capacity, even though a free program is available.
The study aims to identify the factors influencing customers' decisions, including preferences, economic factors, the effectiveness of promotions, perceptions of benefits and risks, and long-term uncertainties.
This research adopts SEM with PLS approach using SmartPLS 4 Software.
Data were collected through questionnaires and interviews with active customers and internal documents from PT PLN Batam.
Findings confirmed that service quality has a positive and significant effect on customer satisfaction and subsequently the decision to enhance their capacity.
Additionally, the decision is also directly influenced by the economic capability.
Furthermore, trust and promotion do not affect customer satisfaction directly in relation to the capacity increase program.
Customer satisfaction is shown to mediate the relationship between several factors and the decision to enhance capacity.
In summary, to improve customer participation.
PT PLN Batam should prioritize enhancing the service quality, education and promotion strategies, customer profiling, uncertainty resolution, and trust Keywords: Customer Preferences.
Service Quality.
Electricity Upgrade.
Satisfaction.
Economic Capability.
INTRODUCTION
The analysis in this case is directed to understanding the reasons behind the reluctance of the R1 category electricity customers with 450 VA power at PT PLN Batam to upgrade their PT PLN Batam's struggle is to improve participation in the program, especially for customers who are the lowest power users and basic need only users.
This issue is important because greater electricity power can help improve customer comfort and quality of life, but in reality most customers do not take it up.
The purpose of this study is to analyze the factors that lead to reluctance towards increasing electricity power among some customers.
3313 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
As pointed out by Kotler and Keller .
, internal and external influences such as value perception and service quality impact consumer choices regarding products and services.
According to consumer preference theory, products are chosen by consumers based on important attributes such as their quality, price, and the benefits expected (Schiffman & Kanuk.
This study is also pertinent to the theory of customer satisfaction since it frequently acts as a pivotal element in influencing a purchase or decision to use a service over the long term (Oliver, 1.
Moreover, rational behavior theory (Becker, 1.
also posits that consumers act to maximize personal utility, which can be observed in the context of decisions to increase electricity power in view of associated benefits.
A decision of this kind is also influenced by economic factors.
As indicated by Muthmainnah .
, a consumerAos economic capability drives their decision-making.
Customers become more inclined to spend on power upgrades with a rise in income or improved economic conditions.
However, a lack of clarity around prolonged costs and their effect on monthly spend creates a need for further understanding (Tamba, 2.
Earlier studies conducted by Firdaus and Purnama .
highlight that insufficient marketing or education about a program will often lead to a misunderstanding which results in apathy towards As noted in the problem statement, the purpose of the study include finding out the marketing mix elements as well as the other factors that greatly attract customerAos interest to the power upgrade program.
It also aims to pose and answer the central question: why do R1 customers with 450 VA power at PT PLN Batam not want to increase their power to a higher one even when the upgrade is free? For this study, the analysis will emphasize the role of economic factors, service quality, marketing promotions, and customer trust in influencing the decision to utilize the power upgrade program.
In studying these aspects, it is hoped that the findings help formulate solutions and practical recommendations for PT PLN Batam in improving customer enrollment in the power upgrade program by enhancing program specific advertising along with educational campaigns about the benefits of increasing power.
METHOD
Research Type To determine the preferences of customers of R1 category 450 VA electricity service at PT PLN Batam who do not want to upgrade their power even with a free upgrade program, this research applies a quantitative approach with a survey.
This study particularly seeks to determine the reason behind the customer's decision not to participate in the power upgrade program and analyses the impact of service quality, promotions, trust, economic capability on those decisions.
Population and Sample The subjects for this study are all R1 category 450 VA registered customers of PT PLN Batam.
The sample consists of 254 purposively sampled respondents who meet the criteria of customers who do not want to upgrade their power service, even with a free upgrade program The sample was taken while considering the respondentsAo place of residence, duration of stay in Batam, and employment status to ensure that the respondents provide the required information regarding the reasons behind the customersAo decisions.
Research Time and Place This research was done from January to March in the year 2023.
The research was done at PT PLN Batam, and the participantsAo data was sourced from customers in Batam.
The study 3314 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
centers around customers with a long history of using the electricity services and have no interest to partake in the power upgrade program.
Research Instrument In this study, the data collection instrument is a designed questionnaire with the purpose of gathering data on the factors influencing customersAo decisions to increase their electricity Regarding service quality, promotions, trust, economic capability, and customer satisfaction with PT PLN BatamAos services, this questionnaire has relevant questions.
Moreover, a few interviews were conducted to understand better customersAo perceptions and Research Procedure In the research procedure, the first step is the design of the questionnaire which has the following items: service quality, promotions, trust, economic capability and customer It was administered to identified customers.
Subsequently, the obtained data were collected and analyzed by Structural Equation Modeling (SEM) using SmartPLS 4 Software.
The analysis was conducted to validate the impact of the independent factors .
ervice quality, promotions, trust, and economic capabilit.
on the decision to upgrade power, and also to assess customer satisfactionAos role as a mediating variable.
Data Analysis Technique The analysis of data was executed by means of Structural Equation Modeling or SEM with a Partial Least Squares PLS approach.
This approach enables testing of relationships between latent variables such as service quality, promotions, trust, economic capability, and customer satisfaction and dependent variables which include customers' decisions to increase SmartPLS 4 software was used to analyze the data for checking model validity and reliability, and for significant influence assessment among the factors under study.
RESULTS AND DISCUSSION
Research Object This research is conducted to analyze the preference of R1 customers electricity 450 VA power in PT PLN Batam that have no will to upgrade the power, although free program has been available.
This study seeks to examine the reasons for customer decisions .
, high costs, lack of knowledge on the benefits of power upgrade and the discrepancy between the amount of power demand and available power capacit.
Respondent Data The respondentsAo characteristics in the study, such as status of residency, length of stay in Batam, and employment status are presented in this study.
Most of the respondents lived in rented houses .
3%) and their length of stay in Batam was over 6 years .
6%), suggesting that the respondents were stable and knew their living environment well.
With respect to labor condition, the majority of respondents .
9%) has a stable job, indicating a higher expenditure propensity and an inclination to consume more power from electricity.
This gives valuable insight for analysing motivations for their preference on electricity services particularly when making the decision to upgrade their electricity power.
Theoretical Review The respondentsAo characteristics in the study, such as status of residency, length of stay in Batam, and employment status are presented in this study.
Most of the respondents lived in rented houses .
3%) and their length of stay in Batam was over 6 years .
6%), suggesting 3315 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
that the respondents were stable and knew their living environment well.
With respect to labor condition, the majority of respondents .
9%) has a stable job, indicating a higher expenditure propensity and an inclination to consume more power from electricity.
This gives valuable insight for analysing motivations for their preference on electricity services particularly when making the decision to upgrade their electricity power (Cialdini, 2.
ConsumersAo decision to continue their electricity power despite the free program offer can be driven by current conditions customer satisfaction and the fear of extra costs from the power upgrade.
Consumer behavior theory examines the way people select, buy, use, and dispose of goods to satisfy their needs.
This behavior is influenced by motivations, norms, and social pressure (Engel et al.
, 1.
The process of consumer decision making have some steps that the process begins with problem recognition, continuing with information search and evaluation of alternatives and completed with a purchase decision (Kotler & Armstrong, 2019.
Meanwhile, the uncertainty theory (Knight, 1.
explains that uncertainty in decision-making can lead to bias, where consumers are more likely to avoid losses rather than pursue equivalent gains, as seen in the decision not to upgrade power despite the availability of a beneficial Uncertainty can also be addressed by providing clearer information or safer alternatives to reduce perceived risks by consumers.
Discussion Outer Model Convergent Validity The majority of indicators meet the convergent validity criteria with an outer loading value > 0.
70 and AVE > 0.
However, there are several indicators (X4.
Z1.
3, and other.
with low outer loading values that need further evaluation, especially X4.
3, which is below the threshold of 0.
50, as shown in Table 1 below.
Tabel 1.
Outer Model Discriminant Validity (AVE) Promotion (X1.
0,625
Service Quality (X2.
0,645
Service Quality (X2.
0,672
Trust (X3.
0,681
Economic Capability (X4.
0,431
Customer Satisfaction (Z1.
0,522
Source: Processed data using SmartPLS 4.
Variable Reliability of Constructs In the analysis of construct reliability, although the values of Composite Reliability and CronbachAos Alpha are not shown, most indicators have an outer loading > 0.
indicating that the construct reliability is generally met.
Based on the criteria used, which are CronbachAos Alpha Ou 0.
Composite Reliability Ou 0.
70, and AVE Ou 0.
50, it can be concluded that the indicators in this model exhibit good internal consistency and adequate convergent validity.
Nevertheless, some signs such as X4.
3 and Z1.
3 present low loadings, which might also present problems of content-relevance, and the test for how poorly constructed or weakly related indicators affect reliability If these indicators need to be improved or removed from the model.
Discriminant Validity The Fornell-Larcker criterion is a widely used approach to assess discriminant validity in the measurement model.
The present procedure guarantees that each construct in the 3316 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
model can be easily differentiated from other constructs.
In this research, all constructs fulfill the discriminatoryvalidity criterion since the square root of the AVE for each construct is greater than the correlation between latent variables, showing that the constructs in the model are adequately explained by their indicators without overlapping heavily with other constructs.
This indicates that the measurement model has acceptable discriminant validity.
For the cross-loading examination, most measures indicate that the highest loadings are on the constructs they are expected to measure, which suggests that most of these measures are relevant measurements of the intended constructs.
But: There are two traces.
X4.
3 and Z1.
3, which are not highly loaded on their intended constructs, suggesting redundancy with other constructs.
Hence, both cues need more testing to verify they do indeed measure themselves properly, and do not induce bias in the measurement.
The HTMT Ratio is another test of discriminant validity, it is much more effective and accurate measure to conventional alternate of discriminant validity such as Fornelarcker Criterion.
The line of results from this work conclude that the HTMT estimates for Service Quality vs.
Trust, and Service Quality vs.
Customer Satisfaction surpass the criterion value .
, which indicates multicollinearity existence between the two This evidence suggests the necessity to evaluate and enhance the model in such a way that the distinction between these constructs is made more explicit, and in this way strengthening the validity and the quality of the results of analysis.
Finally, multicollinearity analysis of the model parameters was evaluated with VIF (Variance Inflation Facto.
, to assess how much the correlations between the indicators affect the estimation of the parameters in the model.
In our study, for all of the variables, the VIF < 5, suggesting that there is no serious multicollinearity.
The low VIF values ensure that we can make parameter estimations for the model more precise, with no involvement from the connection between indicators, which strengthens the reliability and validity of the model adopted in this research.
Inner Model R-Square (Coefficient of Determinatio.
As for the RA test, the RA value of 0.
531 for Customer Satisfaction shows that the modelAos endogenous construct explained approximately 53.
1% of the variance in order to generate Customer Satisfaction.
This value is in moderate category, which suggests that the model has a moderate ability to account for the variance of this case.
Nevertheless, there exist other variables, which have not been sufficiently addressed by the external constructs integrated in the model.
Overall, an RA above 0.
5 suggests the association between the constructs in the model is moderate but can be better explained.
Meanwhile, the RA value for Power Upgrade of 0.
356 indicates that the exogenous constructs can only explain 35.
6% of the variability in Power Upgrade, which is lower compared to Customer Satisfaction.
Although this also falls into the moderate category, the modelAos predictive power for Power Upgrade is lower.
In the context of social or business research, where many external factors influence the results, this RA value is still considered fairly good.
Overall, both constructs analyzed (Customer Satisfaction and Power Upgrad.
fall into the moderate category, indicating that this model can explain most of the variability, although there is still room for the addition or improvement of other factors that may significantly influence the results.
Effect Size .
A) Effect Size .
A) is used to measure the extent to which an independent variable influences a dependent variable in the measurement model.
Based on the given fA values, 3317 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
most of the relationships between constructs show a small effect.
For instance, the effect of Trust on Customer Satisfaction .
A = 0.
Promotion on Customer Satisfaction .
A = .
, and Trust on Power Upgrade .
A = 0.
all demonstrate a small effect.
Similarly, the relationship between Customer Satisfaction and Power Upgrade .
A = 0.
also falls into the small category, indicating that these factors contribute minimally to explaining the variability in the dependent construct.
However, the effect of Service Quality on Customer Satisfaction shows a value of fA = 0.
328, which is in the category close to large.
This indicates that Service Quality has a fairly significant effect on Customer Satisfaction, greater than the influence of other Thus.
Service Quality can be considered the main factor influencing Customer Satisfaction in this model, and it provides valuable insight that companies or organizations need to pay attention to service quality to improve overall customer satisfaction.
Path Coefficients and Significance (Bootstrappin.
Table 2 shows the results of the Path Coefficients and Significance (Bootstrappin.
analysis for the research model.
According to the data, there are two significant paths with p-values < 0.
05: the relationship between Service Quality (X.
and Customer Satisfaction (Y) with a coefficient of 0.
332, and the relationship between Economic Capability (X.
and Power Upgrade (Z) with a coefficient of 0.
Both of these paths have T-statistics 96, indicating that they have a significant influence in the model.
This confirms that Service Quality influences Customer Satisfaction and Economic Capability influences the decision to upgrade power.
Meanwhile, most other relationships are not significant because the p-value is greater 05 and the T-statistic is below 1.
For instance, the relationship between Trust (X.
and Power Upgrade (Z) (P = 0.
and between Trust and Customer Satisfaction (P = 0.
show that there is no significant influence between these constructs.
Some other paths also approach significance, such as Z Ie Y (P = 0.
and X1 Ie Y (P = 0.
, but they do not meet the set threshold.
This shows that, although there are relationships between the variables, their influence is not strong enough to be considered significant.
Hubungan X4 Ie Y X3 Ie Z X3 Ie Y ZIeY X2 Ie Z X2 Ie Y X1 Ie Z X1 Ie Y Tabel 2.
Path Coefficients dan Significanci (Bootstrappin.
Koefisien Keterangan 0,332 0,081 0,116 0,238 0,625 -0,099 0,078 0,162 2,787 0,005 Significant 0,537 0,591 Not Significant 0,771 0,441 Not Significant 1,857 0,063 Not significant .
4,54 Significant 0,688 0,492 Not Significant 0,931 0,352 Not Significant 1,721 0,085 Not significant .
Source: Data processed with SmartPLS 4.
Total Effect The total effect in the measurement model reflects the overall influence of one construct on another, both through direct and indirect paths.
In this framework, the direct effect in the path between the major constructs can be considered as the primary factor.
That is, an increase in one construct actually causes a change in another construct and this relationship is powerful enough to account for the relationships amongst constructs in the Hence the direct paths were mainly accounting for the variance of the dependent 3318 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
Moreover, the indirect effects are also on whole not significantly proven in this model, which implicates, thought the possibility of influence of these mediating paths are available, but is so small that may not affect the results.
It suggests that the direct effect between constructs is the stronger of the two and is instrumental in explaining the relationship in the model.
Therefore, direct connections among constructs should be more focused and are more emphasize than the ones that mediators or indirect ways.
Goodness of Fit (SRMR) The SRMR outcome for the Saturated Model and Estimated Model, which are 0.
097, respectively, are both above the 0.
08 threshold which means the model has not been tailored to fit the data.
More refinement is required in terms of model fitting.
Hypothesis Testing In Table 3, the results of hypothesis testing are presented alongside the respective coefficients.
T-statistics.
P-values along with decisions for every hypothesis.
For hypothesis H1 which examines the impact of promotion on customer satisfaction, the results yield a coefficient of 0.
Tstat of 1.
723 and P value of 0.
085 which indicates that this relationship is not significant as the P value is greater than 0.
With regards to Hypothesis H2 which looks into the impact of service quality on customer satisfaction, it has a coefficient of 0.
332, with a Tstat of 2.
787 and P value of 0.
005, meaning service quality does have a positive significant impact on customer satisfaction.
Next.
Hypothesis H4 tests the relationship between economic capability and the likelihood of increasing power, with a coefficient of 0.
625, a T-statistic of 4.
54, and a P-value 000, indicating a significant effect.
Meanwhile, other hypotheses such as H3 .
rust Ie customer satisfactio.
and H5 .
ustomer satisfaction Ie increase in powe.
show coefficients 095 and 0.
174, respectively, but with P-values of 0.
441 and 0.
063, meaning both are not statistically significant.
This analysis highlights which factors are significant and which need further analysis or improvement in the context of this model.
Hypothesis Tabel 3.
Hypothesis Testing Relationship Coefficient Promotion Ie Customer Satisfaction Service Quality Ie Customer Satisfaction Trust Ie Customer Satisfaction Economic Capability Ie Power Increase Description 0,14 0,332 1,723 2,787 0,085 0,005 Not Significant Significant 0,095 0,625 0,772 4,54 0,441 Not Significant Significant 0,174 Customer Satisfaction Ie Power Increase 1,862 0,063 Not Significant -0,06 0,54 Trust Ie Power Increase Source: Data processed using SmartPLS 4.
0,591
Not Significant Mediation Hypothesis Testing Mediation hypothesis testing in PLS-SEM aims to determine whether customer satisfaction acts as a mediating variable that links the effects of independent variables .
romotion, service quality, trust, and economic capabilit.
to the customer's decision to increase their electricity capacity.
The mediation testing process involves testing the path coefficients between the independent, mediating, and dependent variables.
In this case.
Baron & Kenny .
outlines three steps for proving the mediation effect.
First, the independent variable must significantly impact the mediating variable.
Second, the mediating variable must 3319 | P a g e https://dinastipub.
org/DIJEFA Vol.
No.
4, 2025
impact the dependent variable significantly.
Third, the direct impact of the independent variable on the dependent variable should diminish or vanish when the mediating variable is Moreover, testing for partial mediation assesses whether customer satisfaction strengthens or weakens the independent-dependent variable relationship, or whether it entirely negates the direct effect through mediation.
This testing is vital for understanding the relationships between these variables, which assists PT PLN Batam in developing more appropriate marketing initiatives.
PT PLN Batam will be able to craft tailored customer satisfaction programs that drive higher engagement in the electricity capacity increase program.
CONCLUSION
This study tries to determine the effects of certain factors relating to service quality and customer satisfaction on the decision of R1 450VA users at PT PLN Batam to increase their electricity capacity.
Based on the hypothesis test results, it was found that Service Quality has a significant impact on Customer Satisfaction which subsequently impacts the decision to increase electricity capacity.
This relationship is bolstered by a significant path coefficient between Service Quality and Customer Satisfaction with p-value < 0.
This result strongly confirms the role of service quality on customer satisfaction decisions in regard to the capacity increase program.
The Promotion to Customer Satisfaction (H.
and Trust to Customer Satisfaction (H.
hypotheses were not significant.
While promotion and customer trust may contribute to an overall impression of the service, they have no bearing on customer satisfaction concerning the decision to raise electricity capacity.
This suggests that more tangible factors, such as service quality, are more dominant in shaping customer satisfaction in this program.
Additionally.
Economic Capability was found to have a significant impact on the Decision to Increase Capacity, as revealed in hypothesis H5.
Customers with better economic capability tend to be more inclined to increase their capacity, despite concerns about additional Customer Satisfaction plays an important role as a mediating variable, strengthening or weakening the relationship between independent factors .
uch as promotion, service quality, and economic capabilit.
and the decision to increase capacity.
The mediation test results indicate that maintaining customer satisfaction is key to influencing their decision to participate in the electricity capacity increase program.
REFERENCES