Willingness to Pay Household-based PDAM Clean Water Service in Purwosuman Village Sidoharjo District Sragen Regency

. Purwosuman Village is one of the villages in Sidoharjo District, which has several textile industries. The textile industry's existence provides various impacts on the environment, one of which is a decrease in water quality and quantity. Problems related to water resources trigger high costs to obtain water. This study examines the Willingness To Pay of the PDAM's desired clean water service by households and the factors influencing the value of Willingness To Pay . The sampling technique used is simple random sampling . Data analysis used the Contingent Valuation Method (CVM) approach to determine the value of Willingness To Pay. M ultiple linear regression was performed to determine the factors that affect Willingness To Pay. This research shows that the average value of Willingness To Pay PDAM clean water tariff is IDR3.656,00/10m³; this value is more significant than the PDAM clean water rate in Sragen Regency IDR3.400,00/10m³. The total value of Willingness To Pay is IDR292.480,00. Factors that affect Willingness To Pay are income, education, and distance to water sources, while the number of family members does not affect Willingness To Pay (WTP).


Introduction
Water resources are an essential element for living things that are indispensable for the survival of life.The different potential of water resources in each region causes the issue of water resource problems to be studied by international and national organizations [1].Population growth and increasing demand for water while the availability of water is fixed can cause various water resource problems [2].In Indonesia, the problem of water resources that is difficult to solve is related to the availability of clean water and sanitation.The consumption of humans with the increasing demand for water causes the availability of water to decrease so that it can trigger high costs that need to be spent to obtain water [2].The value of Willingness To Pay (WTP) can be used as a measuring tool for costs incurred and knowing the size of the community's desire to obtain a change towards environmental improvement [3].
The problem of meeting the need for clean water or drinking water occurs not only in urban areas with dense populations but also in rural areas.For example, Purwosuman Village is one of the villages in the Sidoharjo District which has several textile industries.Industrial liquid waste resulted in a decrease in water quality in Purwosuman Village, especially around the Winong river that the physical quality of the river water had a brownish colour, there was oil and dirt from * Corresponding author: rikaharini@ugm.ac.id factory waste were still visible [4].The decreased water availability, especially during the dry season, is also a problem in Purwosuman Village.Some people do not use healthy water and prefer water from PDAM (Regional Drinking Water Company) to meet their water needs.The condition of water in Purwosuman Village will trigger high prices to obtain clean water with good PDAM water services so that it can be used to meet community water needs.
The need for clean water in Purwosuman Village gets a supply of clean water managed by Perumda Tirto Negoro Drinking Water, Sragen Regency.PDAM has made efforts to build clean water networks to meet water needs.The construction of the clean water network is expected to meet the water needs of both urban and rural residential areas.Figure .1 illustrates that, in general, the people of Sragen Regency have quite good access to clean water.
The condition of water quality and PDAM services based on the perception of PDAM user households still requires improvement due to complaints such as slightly cloudy water, low water pressure, and abnormal water meter.The problem of water resources cannot be separated from the existence of economic analysis, one of which is a study conducted using Willingness To Pay.Willingness To Pay can be used as a measure of the size of the costs received and issued as compensation for the impact of resource utilization [5].This study focuses on the perceived values associated with PDAM services from a household perspective.The value given by the household is the choice expressed by the household when faced with a hypothetical scenario.The survey-based method raises respondents' Willingness to pay (WTP) with the Contingent Valuation Method (CVM) approach.CVM is the correct method to assess the improvement of water quality and PDAM services because water resources include both use value and non-use [3].Some examples of studies that have elicited respondents' preferences for improving water quality and clean water services by determining the value of Willingness To Pay are provided by [6]

Methods
The primary method used in this research is descriptive using a quantitative approach.Sources of data used include primary data and secondary data.Primary data collection was carried out by direct measurement in the field through structured interviews with questionnaire research instruments to respondents on a household scale and simple random sampling techniques.The consideration of using simple random sampling in this study is that members of the population are considered homogeneous, the unit of sample selection only consists of one type, accuracy, time, acceptance of results and generalizability.The population of this research is PDAM user households in Purwosuman Village.Based on the Krejcie and Morgan formula calculation results, the number of samples taken in Purwosuman Village was 80 respondents'.The secondary data collection used in the study was sourced from village profile data, literature studies, and previous research.

Location
The research location was conducted in Purwosuman Village, Sidoharjo District, Sragen Regency (can be seen in figure .2).The area owned by Purwosuman Village is 450 Ha or 4,50 km², which is included in the Solo Zone.Purwosuman Village was chosen as the research location because Purwosuman Village has several textile industries, which indicates water pollution due to production waste, the availability of water is not always the same, especially in the dry season, and the distribution of water provided by PDAM that is not by the wishes of households with complaints.From the people of Purwosuman Village, such as cloudy water, abnormal water meter, and water pressure that is not always the same.

Willingness to Pay Value Analysis
Analysis of the Willingness To Pay value for householdbased PDAM clean water services using the Contingent Valuation Method (CVM) approach.The Contingent Valuation Method (CVM) is a dichotomous choice question method (Close-Ended Referendum).Stages in conducting a study using the Contingent Valuation Method (CVM) according to [10] are: 1. Determining the Market Hypothesis 2. Getting Auction Value (Bids) 3.Estimated Average Willingness To Pay, with the following formula [11]:  3 Result and Discussion

Characteristics of Respondents
The characteristics of PDAM users in Purwosuman Village are explained based on criteria such as gender, age, education, employment, income, PDAM expenditures per month, number of family members, water source, and water source distance.

Gender.
The results of data collection of as many as 80 respondents showed that 57 respondents (71%) were male and 23 respondents (29%) were female (can be seen in figure .3).The male dominates respondents because men are the head of the family (decision-makers) in a household.Therefore, the answering questions will better represent the actual situation of perceptions and economic conditions of PDAM services.

Age
The characteristics of respondents according to age are quite varied, divided into seven age groups ranging from 5 years.Respondents were dominated by the 47-52 year age group with 26 respondents (33%) and the 41-46 year age group with 17 respondents (21%).At least three respondents have an age between 65-70 years, or 4% (can be seen in figure . 4).This explains that adults dominate this research.Adult respondents can provide more mature thoughts to be more responsive and actual in answering questions.

Education
The characteristics of respondents according to education level are divided into four groups: elementary school graduates, junior high school graduates, high school graduates, and university graduates.Based on the data in figure.5, it is known that the highest level of education is high school graduates/equivalent to as many as 27 respondents (34%).The data can explain that the answers from respondents are quite reliable and represent actual conditions because the answers given are rational.The education of respondents with a high literacy rate has a better chance of maximizing utility and welfare in drinking water and access to clean water [6].

Employment
Characteristics of respondents by occupation are quite varied.Private employees dominate the type of work of PDAM users in Purwosuman Village, as many as 30 respondents (38%), and entrepreneurs/traders as many as 24 respondents (30%) (can be seen in figure . 6).
Respondents have jobs with high incomes will give high appreciation in terms of their Willingness to pay clean water tariffs [12].Respondents are dominated by having an income of IDR 3.000.000-IDR4.499.999as many as 33 respondents (41%).Most respondents have an income level above the minimum wage in Sragen Regency, which is set at IDR 1.839.000so that the standard of living can be fulfilled.Minimums such as for health, efficiency, and well-being.Most of the respondents using PDAM in Purwosuman Village belong to the middle class.

PDAM Expenditures per Month
Respondents predominantly have PDAM expenditures per month of IDR35.000-IDR75.000, as many as 36 respondents (45%) (can be seen in figure.8).Households with higher monthly expenditures for PDAMs are more likely to be willing to pay for PDAM clean water service.In addition, households charged with higher fees will be more willing to pay for better drinking water services [6].

Water Sources Distance
The distance of the water source is related to the ease of access to clean water.Based on figure.11, the distance of the water source is dominated by a distance of 5-6 km for 39 respondents (49%).Therefore, the distance to water sources can affect Willingness's value to pay for clean water rates [13].1. Determining the Market Hypothesis The market hypothesis built in this research is based on environmental issues related to water resources.All respondents were told that using PDAM water to meet current water needs can still provide an adequate water supply.However, the water quality and quantity may decrease in the future.Improving the quality of PDAM services with better conditions if carried out by PDAMs requires the role of the community to bear the costs of implementing the plan to improve the quality of PDAM services.Submission of this market hypothesis to respondents resulted in information about the public's Willingness to pay for better PDAM clean water services.

Obtaining Auction Value (Bids)
The bidding of the value of Willingness To Pay for PDAM clean water services desired by households is carried out using a dichotomous choice question method (Close-Ended Referendum).Offers to the public are made with a starting point submitted to customers of IDR3.500.The starting point can be based on the water tariff paid per the first 10 m³.Willingness To Pay is IDR4.500/10m³ with a frequency of 6 respondents (8%), while the lowest value that PDAM users are willing to pay is IDR3.500/10m³.The first is 61 respondents (76%).Therefore, the value of Willingness To Pay that the respondents most chose in this study was IDR3.500.

Estimating the Bids Curve Willingness To Pay
Curve Willingness To Pay illustrates the relationship between the Willingness To Pay and the number of respondents willing and able to pay any Willingness To Pay value.Figure 12 shows that the higher the Willingness To Pay, the fewer respondents are willing to pay for PDAM clean water services.Following the law of demand that the price is inversely proportional to the quantity demanded.The higher the price, the less quantity of goods or services demanded.

Factors Affecting the Value of Willingness
To Pay

Classic Assumtion Test
The classical assumption test proves that multiple linear equations are BLUE (Best Linear Unbiased Estimator), where decision-making through the F significance test and t-test cannot be biased [14].Multiple linear regression models must meet data normally distributed, linear, and homogeneous variation criteria.The classical assumption test can be done with five tests: normality, linearity, multicollinearity, heteroscedasticity, and autocorrelation [15].This study did not use the autocorrelation test because the data used were not in the form of time series.The normality test is used to determine whether the residual value in the data is normally distributed or not [15].Data is reasonable and feasible if it has a normal distribution [14].The normality test in this study used the Kolmogorov Smirnov test.Based on the normality test results listed in table 3, it is known that the value of Asymp.Sig (2-tailed) is 0,061, where the value is more significant than 0,05 (α=5%) so that the data is normally distributed.A linearity test determines whether the model's specification is correct.Good data and feasible to use is linear data [15].The linearity test in this study uses deviation from linearity.The decision-making criteria if the significance value of deviation from linearity is more significant than 0,05, then there is a linear relationship between the independent variable and the dependent variable.Based on the results of the linearity test listed in table 4, the significance value of deviation from linearity of income, education, number of family members, and distance to water sources is 0,451; 0,054; 0,500, and 0,446, respectively.This value is more significant than 0,05, indicating a linear relationship between income and WTP, education and WTP, number of family members with PAP, and the distance between water sources and WTP.The multicollinearity test determines whether there is a correlation between independent variables in the regression model.A good regression model is a regression model that does not correlate with the independent variables [14].Multicollinearity test can be seen from the Tolerance and Variance Inflation Factor (VIF) value.If the Tolerance is more than 0.1 and the Variance Inflation Factor (VIF) value is not more than 10, then the model is said to be free from multicollinearity [15].The results of the multicollinearity test were obtained; the Tolerance value of all independent variables is close to 1, range 0,8 to 0,9.Variance Inflation Factor (VIF) value on all independent variables is around 1, which is in the range of 1-1,2 (seen in table 5).This value indicates that the regression model does not occur multicollinearity.The heteroscedasticity test aims to test whether, in a regression model, there is an inequality of variance from residuals in one observation to another [15].A good regression model is a regression model that does not show heteroscedasticity symptoms.The heteroskedasticity test in this study used Spearman's rho test.The basis for decision-making in Spearman's rho is that if the significance value is more significant than 0,05, there is no heteroscedasticity.The heteroscedasticity test in table 6, the significance value for the variables of income, education, number of family members, and distance to water sources is 0,167; 0,310; 0,998, and 0,865.Therefore, the significance value is more significant than 0,05; so it can be concluded that the regression model does not occur heteroscedasticity.The F statistical test determines whether the independent variables simultaneously or together can affect the dependent variable [15].The basis for decision-making on the F statistical test is if the significance value is less than 0,05 or the calculated F value is greater than the F table value, the independent variables simultaneously (together) affect the dependent variable.Based on the results of the F statistical test listed in table 8. it is known that the calculated F value is 21,679 with a significance value of 0,000.The calculated F value is greater than the F table (2,494), and the significance value is less than 0,05; so it is known that the independent variables, namely income, education, number of family members, and the distance of water sources together (simultaneously) affect the Willingness To Pay.
The t-statistical test determines whether the independent variable gives an individual (partial) effect on the dependent variable [15].The basis for decision making is on the t statistical test; if the significance value is less than 0,05, it indicates that there is an influence of the independent variable individually (partial) on the dependent variable.Based on the t statistical test results in table 9, the income variable has a significance value of 0,000, so the individual income variable (partial) affects Willingness To Pay.The education variable has a significance value of 0,000; so the individual education variable (partial) affects Willingness To Pay.The variable number of family https://doi.org/10.1051/e3sconf/202346807002ICST UGM 2023 members has a significance value of 0,568, which is more remarkable than 0,05, so the number of individual family members (partial) does not affect Willingness To Pay.The water source distance variable has a significance value of 0,004; the individual water source distance variable (partial) affects Willingness To Pay.

Multiple Linear Regression Analysis
Regression analysis measures the strength of the relationship between two or more variables and shows the direction of the relationship between the dependent and independent variables [15].The regression test results listed in Based on equation ( 5) and the results of hypothesis testing, it can be interpreted that the constant value in the equation is 46,459.If the independent variables, namely income, education, number of family members, and the distance of water sources, are constant or not changing, the amount of WTP has a value of 46,459.The income regression coefficient has a positive value of 0,003, indicating that if the income variable increases by one unit, the WTP variable will increase by 0,003.Based on the hypothesis testing that the income variable has a significance value of 0,000, which indicates that income has a significant effect on Willingness To Pay, there is a positive and significant effect of the income variable on Willingness To Pay.
This study is in line with research conducted to explain that the income variable has a positive and significant effect on Willingness to pay for better drinking water services [6][16] [8].The higher the income of the respondent, the respondent is willing to spend more money for access to better and quality water.This high Willingness to pay is driven by an awareness of the importance of environmental and health sustainability and the expectation of an increase in water quality and services that PDAM will provide.On average, respondents with a higher income are willing to pay more to improve drinking water quality [17].The income variable significantly influences WTP in improving drinking water quality, where higher family income can increase the WTP value by US$0.162 in Mexico City [18].
The education regression coefficient has a positive value of 2.122, indicating that an increase of one unit in the education variable will increase the WTP variable by 2.122.The results of the hypothesis test that has been carried out are known that the education variable has a significance value of 0.000 or less than 0.05 (α=5 %), which indicates that education has a significant effect on Willingness To Pay.Therefore, the education variable positively and significantly affects Willingness To Pay.This research is in line with previous research that explains that education positively and significantly influences the Willingness To Pay for better water services [19][8].The higher a person's education, the higher the understanding of water resources, concern for environmental sustainability and health, and the greater the Willingness to pay to obtain better and quality PDAM clean water services.Furthermore, the education E3S Web of Conferences 468, 07002 (2023) https://doi.org/10.1051/e3sconf/202346807002ICST UGM 2023 level of the head of the household was positive, indicating that the head with higher education showed a higher willingness to pay compared to the head of the household with low education [6].
The regression coefficient of the number of family members has a positive value of 0.378, indicating that an increase of one unit in the variable number of family members means that the WTP variable will increase by 0.378.Based on the results of hypothesis testing, the number of family members has a significance value of 0.568, which indicates that the variable number of family members has no significant effect on Willingness To Pay.Previous research by [20] stated that the number of family members significantly affected Willingness To Pay.The more the number of family members, the higher the WTP.However, in this study, the number of family members does not affect the value of Willingness To Pay.The results of this study are the same as research by [21] that the number of family members has no significant effect even though it has a positive value on the value of Willingness To Pay.Many family members are not necessarily Willingness To Pay (WTP) for clean water to be higher and vice versa; there may be other factors that affect the value of Willingness To Pay besides the number of family members.The value of Willingness To Pay depends on the respondent's perception and not on market behaviour.Respondents do not consider the number of family members in answering the question of how much rupiah they are willing to pay for better PDAM clean water services.The more number of family members does not cause the higher the value of Willingness To Pay for clean water PDAM paying for clean water PDAM is not a priority [9].
The regression coefficient for the distance of water sources has a positive value of 2.284, indicating that an increase of one unit in the education variable means the WTP variable will increase by 2,284.Based on the results of hypothesis testing, it shows that the water source distance variable has a significance value of 0.004 which means that the water source distance variable has a significant effect on Willingness To Pay, there is a positive and significant effect of water source distance on Willingness To Pay.This study is in line with previous research that explains the distance of water sources has a positive and significant effect on the value of Willingness To Pay for clean water [6] [13].The distance of the water source is related to the ease of access to clean water; where the water source is far, a person's efforts to obtain water include a willingness to spend more money compared to a short distance.

Conclution
Willingness To Pay (WTP) PDAM clean water service in Purwosuman Village uses the Contingent Valuation Method (CVM) approach with the highest value of IDR4.500/10m³ and the maximum WTP value of IDR3.500/10m³, namely 61 respondents (76%).The average value of WTP is IDR3.656/10m³, which is greater than the first PDAM water tariff in Sragen Regency is IDR3.400/10m³.The Total Willingness To Pay (TWTP) value is done by multiplying the average Willingness To Pay sample by the number of respondents of 80 respondents resulting in a TWTP value of IDR292.480/10m³.
Based on multiple linear regression analysis, it is explained that the influence of the independent variables, namely income, education, number of family members, and the distance of water sources simultaneously (together) on Willingness To Pay is 51,1%.The remaining 48,9% is influenced by the variable others not investigated.The variables of income, education, and distance to water sources positively and significantly affect the value of Willingness To Pay.In contrast, the variable number of family members has a positive regression coefficient and does not affect the value of Willingness To Pay.

Fig. 1 .
Fig. 1.Graph of Percentage of Population Who Get Access to Clean Water from PDAM Sragen Regency in 2015-2018 [7][8][9].Although the CVM method has been used to estimate the value of Willingness To Pay in developed countries, its application in developing countries remains very limited.Research to estimate the value of Willingness To Pay water quality and PDAM services has never been done in Purwosuman Village.Willingness To Pay for clean water in each area is different and can be influenced by several factors.PDAM clean water services used by the community in Purwosuman Village indicate that households are willing to pay more for water quality and better PDAM services.This study examines the Willingness To Pay of PDAM clean water services that households want and the factors that influence the value of Willingness To Pay.The results of this study can be used as consideration for improving PDAM clean water services and deciding on determining the standard price of clean water services for the people.

Fig. 12 .
Fig. 12.The Bids Curve Willingness To Pay

table 1 .
The value of Willingness To Pay in detail is presented in

Table 1 .
Value of Willingness To Pay.

Table 2 .
Average Willingness To Pay.

Table 7 .
Coefficients of Determination.