i
Evaluating the Feasibility of Village-Based Pharmacies 
in Karakalpakstan, Uzbekistan
Prepared by 
Counterpart International 
Umir Nuri (Rainbow of Life) Child Survival Program 
Funded by U.S. Agency for International Development 
Karakalpakstan, Uzbekistan 
July 15, 2003 
ii
Contents 
Tables ................................................................................................................................ iv
Figures............................................................................................................................... iv
Part One: Introduction..................................................................................................... 5
Section 1.1: Background......................................................................................... 5 
Section 1.2: Willingness-to-Pay Surveys................................................................ 5 
Part Two: Survey Implementation.................................................................................. 6
Section 2.1: Survey Design..................................................................................... 6 
Section 2.2: Field Work .......................................................................................... 6 
Section 2.3: Data Preparation ................................................................................. 7 
Part Three: Data Analysis................................................................................................ 8
Section 3.1: Reliability and Validity of the Survey ................................................ 8 
Section 3.2: Overall Level of Willingness to Pay................................................... 9 
Part Four: Analysis of Pneumonia Survey ................................................................... 10
Section 4.1: Analysis of the Whole Group ........................................................... 10 
4.1.1 Willingness to Pay, by Experience of Illness, Distance to 
Source, and Household Income .................................................... 10 
4.1.2 Determinants of Willingness to Pay ............................................. 12 
Section 4.2: Analysis of the Subgroup with Illness Experience ........................... 12 
4.2.1 Willingness to Pay, by Distance, Income, and Payment............... 12 
4.2.2 Determinants of Willingness to Pay ............................................. 14 
Part Five: Analysis of Diarrhea Survey........................................................................ 14
Section 5.1: Analysis of the Whole Group ........................................................... 14 
5.1.1 Willingness to Pay, by Experience of Illness, Distance to 
Source, and Household Income .................................................... 14 
5.1.2 Determinants of Willingness to Pay ............................................. 16 
Section 5.2: Analysis of the Subgroup with Illness Experience ........................... 16 
5.2.1 Willingness to Pay, by Distance, Income, and Payment............... 16 
5.2.2 Determinants of Willingness to Pay ............................................. 18 
Part Six: Summary and Conclusions ............................................................................ 18
Section 6.1: Clear and Strong Willingness to Pay ................................................ 18 
Section 6.2: Factors Influencing Willingness to Pay............................................ 18 
Section 6.3: Financial Sustainability of the Village Pharmacy............................. 19 
Section 6.4: Improving the Welfare of the People................................................ 19 
Section 6.5: Policy Recommendations ................................................................. 19 
References........................................................................................................................ 20
iii
 
Appendices 
A Statistics on Maximum Willingness to Pay 
B Statistical Analysis of Pneumonia Sample 
C Statistical Analysis of Pneumonia Sample, with Households with Experience of 
Illness 
D Statistical Analysis of Diarrhea Sample 
E Statistical Analysis of Diarrhea Sample, with Households with Experience of 
Illness 
iv
Tables 
Table 1. Categories of Survey Respondents ....................................................................... 7 
Table 2. Maximum Willingness to Pay (in Sum) ............................................................... 9 
Table 3. Pneumonia Maximum Willingness to Pay, by Illness Experience, 
Distance, and Income................................................................................ 10 
Table 4. Regression on the Determinants of WTP for Pneumonia Antibiotics................ 12 
Table 5. Pneumonia Maximum Willingness to Pay, by Distance, Income, and 
Payment..................................................................................................... 12 
Table 6. Regression on the Determinants of WTP for Pneumonia Antibiotics................ 14 
Table 7. Diarrhea Maximum Willingness to Pay, by Illness Experience, Distance, 
and Income................................................................................................ 14 
Table 8. Regression on the Determinants of WTP for Diarrhea Antibiotics.................... 16 
Table 9. Diarrhea Maximum Willingness to Pay, by Distance, Income, and 
Payment..................................................................................................... 16 
Table 10. Regression on the Determinants of WTP for Diarrhea Antibiotics.................. 18 
Figures 
Figure 1. Distribution of MWTP ........................................................................................ 9 
Figure 2. MWTP, by Distance, for Pneumonia ................................................................ 11 
Figure 3. MWTP, by Household Income, for Pneumonia ................................................ 11 
Figure 4. MWTP, by Paying Experience, for Pneumonia ................................................ 13 
Figure 5. MWTP, by Household Income, for Diarrhea .................................................... 15 
Figure 6. MWTP, by Paying Experience, for Diarrhea .................................................... 17 
5
Part One: Introduction 
Section 1.1: Background 
In the Takhtakupir and Nukus rayons (or 
districts) of Karakalpakstan, an autonomous 
region of Uzbekistan, respiratory complications, 
infectious and parasitic conditions, and diarrheal 
disease pose grave threats to child health, serving 
as the major contributors to the death of children 
under five. The Umir Nuri (Rainbow of Life) 
Child Survival Program, a U.S. Agency for 
International Development (USAID) initiative, is 
working to lower infant and child mortality rates 
in these areas and to increase the capacity of local 
organizations to provide care. 
Recognizing that access to medicines is vital to disease control, USAID’s 
contractor, Counterpart International, is examining the possibility of establishing 
pharmacies in local villages. Currently, residents must travel long distances to obtain 
drugs, or must buy them from unreliable vendors. Village pharmacies would ensure 
customers receive affordable, high-quality drugs for their children and clear instructions 
for their use. By designing questionnaires to assess drug purchase practices and to 
determine the willingness of residents to pay for drugs, RTI is helping Counterpart to 
conduct a willingness-to-pay (WTP) survey and analysis to determine the feasibility of 
establishing and maintaining such pharmacies. 
Section 1.2: Willingness-to-Pay Surveys 
The willingness-to-pay survey is a well-established method used by social 
scientists to measure how much consumers value a commodity or service that either is 
not available in the market or is being provided by noncommercial agents at subsidized 
prices. The purpose of conducting WTP surveys is to gauge the potential demand for 
such products or services and the financial sustainability of providing them in a market 
situation within a certain price range. The results can also be interpreted as welfare gains 
to the consumers from making such products or services available to them. 
Willingness-to-pay surveys usually are designed to ask people the maximum 
amount they are willing to pay for the product or service in question. Since such surveys 
hypothetically assume (or are contingent upon) the existence of a market for the product 
or service, they are also called contingent valuations. 
The mean and distribution of WTP from survey data can be used to estimate the 
number of people likely to pay for the product or service if it were offered. This 
information can be further used to determine an appropriate price level that may generate 
sufficient revenues to allow this new activity to be financially sustainable. Based on the 
To purchase medicines, many residents of
Karakalpakstan travel long distances to
pharmacies like the one shown above,
which is located in the Nukus rayon center
6
characteristics of the respondents, it is also possible to project future changes in WTP and 
even to design interventions to influence WTP. 
Part Two: Survey Implementation 
Section 2.1: Survey Design 
RTI designed two questionnaires for the 
willingness-to-pay study: an exit interview and a 
household interview. The exit interview was 
intended to capture households that recently (i.e., 
at the time of the interview) had a child 
experience either diarrhea or pneumonia. The 
household interview was intended to gather 
information from households that may have 
experienced these two diseases in the past. The 
main purpose of the overall survey was to find out 
the maximum amount of money that people were 
willing to pay for the availability, convenience, 
and reliability of getting prescription drugs at a 
nearby village pharmacy. These questionnaires also examined the current drug 
prescription and purchase practices in the region. 
In designing the questionnaires, we followed the guidelines for WTP surveys 
developed by Foreit and Foreit (2001) to ensure the consistency and reliability of the 
survey. One method is to use probing and sequencing to ask several questions about 
respondents’ willingness to pay for drugs. In addition, a final question about the 
maximum amount the respondent is willing to pay is added to extend the price range. For 
those who had paid for medicines in the past, we used the amount that they paid before as 
the starting point. For others, we used the market price of amoxicillin as the starting 
point. 
Drafts of the survey questionnaires were forwarded to Counterpart’s local office 
in Nukus for review and translation. A pretest was conducted in January 2003. Based on 
the results of the pretest, CSP staff worked with Dennis Chao to revise and finalize the 
questionnaires during his visit to Nukus in February 2003. 
Section 2.2: Field Work 
The training for interviewers (village health workers and field officers, as well as 
other staff members) took place on February 27–28, 2003. CSP staff members conducted 
the training. During the training, target communities for the survey were selected and a 
schedule of visits prepared. 
CSP staff along with Dr. Chao (RTI) visited
Takhtakupir and Nukus rayons to examine
how village-based pharmacies might best
operate and to oversee survey design and
7
Survey work started on March 3 and ended on March 18. The survey was 
conducted in two steps. First, the household interviews were conducted. The process took 
5 days; 21 people were involved in carrying out the interviews. Overall, 802 interviews 
were conducted. During the second phase, the exit interviews were conducted with 320 
mothers and other caretakers. Due to a lack of patients in the community health facilities 
on the interviewing days, an alternative method was used to find patients who had visited 
village clinics recently. Names and addresses of patients who had visited the village 
clinics within the past 2 weeks were collected from the health facilities. Using those lists 
of names, interviewers visited households and conducted interviews. Field officers and 
other program supervisory staff checked the questionnaires at the end of the day. 
Section 2.3: Data Preparation 
One major effort in preparing the data set for analysis was to combine several 
variables measuring the maximum willingness to pay (MWTP) into one variable. 
Depending on their experience with the two illnesses and whether they had paid for drugs 
before, respondents were asked about their MWTP in several different ways. Each 
answer was assigned a different code. We generated a combined file containing variables 
from the exit survey and the household survey, with a single variable representing the 
MWTP. With this combined file, we could analyze the whole sample or any subgroups. 
We also redefined the data set to include willingness to pay for antibiotics only. In 
the original data set, there were 275 records responding to nonantibiotic drugs, such as 
oral rehydration salts (ORS). The costs of these nonantibiotic drugs were much lower 
than those for antibiotics; therefore, answers on MWTP from these records were not 
comparable to the rest of survey. In order not to bias the MWTP estimates for antibiotics, 
which were the major drugs in question, we eliminated all 275 records of nonantibiotic 
cases and concentrated on the other 1,969 records.
In the combined data set, we identified six types of respondents, as shown in 
Table 1. 
Table 1. Categories of Survey Respondents 
Respondent Categories No. in Category 
Child pneumonia: 
No case in household in the past 703 
At least one case and paid for antibiotics 182 
At least one case and did not pay for antibiotics 206 
Child diarrhea: 
No case in household in the past 819 
At least one case and paid for antibiotics 14 
At least one case and did not pay for antibiotics 45 
 
8
Part Three: Data Analysis 
Section 3.1: Reliability and Validity of the Survey 
The reliability of a willingness-to-pay survey is defined by how well respondents 
are able to answer questions about how much they are willing to pay for the commodity 
or service in question and the degree of “yea-saying” (i.e., giving an answer the 
respondent believes the interviewer wants to hear) in answering probing questions. In the 
current survey, all the respondents understood the willingness-to-pay questions and were 
able to answer them. Furthermore, there appeared to be no yea-saying problem since the 
answers to the maximum willingness-to-pay question were all greater than or equal to the 
highest price that the respondent had agreed to pay previously, as revealed by probing. 
The average amount of reported actual payments for one unit of antibiotic drugs was 
almost identical to the market price. This seems to be evidence that the survey results are 
reliable. 
The validity of a willingness-to-pay survey is defined by how closely the 
willingness-to-pay results conform to the income level of the individual or household. 
The current survey shows strong and consistent evidence of a positive relationship 
between household income and willingness to pay for drugs. We present the results in the 
next section. 
9
Figure 1. Distribution of MWTP 
0
100
200
300
400
500
600
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
5,000
5000+
Total
Number of Respondents 
Section 3.2: Overall Level of Willingness to Pay 
We first examine overall willingness to pay for antibiotic drugs. Table 2 and 
Figure 1 show the range and distribution of the maximum amounts that respondents were 
willing to pay (MWTP). Details of the statistical analysis are reported in Appendix A. 
Table 2. Maximum Willingness to Pay (in Sum) 
 Frequency Mean S.D. Median Minimum Maximum 
Overall 1,969 1,689 1,092 1,500 100 10,000 
Pneumonia 1,091 1,734 1,122 1,500 100 10,000 
Diarrhea 878 1,632 1,051 1,500 100 10,000 
Figure 1. Distribution of MWTP 
The overall mean value of the maximum amount that the respondents were 
willing to pay was 1,689 Sum. The fact that the mean WTP value for all respondents was 
substantially higher than the prevailing market price of a unit of antibiotics (at 1,320 
Sum) was a primary indication that the villagers supported the idea of establishing village 
pharmacies and were willing to pay a premium for the convenience and reliability of 
village pharmacies. When examined separately, the differences in mean values of MWTP 
for diarrhea and pneumonia were small but statistically significant at the .039 level. This 
seems to suggest that in answering WTP questions, villagers were mainly responding to 
the antibiotic commodity itself and not to the purpose for which it would be used. 
Therefore, their expressed willingness to pay for antibiotics was similar, whether to cure 
diarrhea or pneumonia. 
In the next sections, we provide a more in-depth analysis of willingness to pay, 
for different groups. We examine results from the pneumonia survey and the diarrhea 
survey separately. 
MWTP Value (in Sum) 
10
Part Four: Analysis of Pneumonia Survey 
Section 4.1: Analysis of the Whole Group 
4.1.1 Willingness to Pay, by Illness Experience, Distance to Source, and 
Household Income 
Table 3 shows the distribution of MWTP by illness experience, distance to source 
of drug, and household income. Details of the statistical analysis are reported in 
Appendix B. 
Table 3. Pneumonia Maximum Willingness to Pay, by Illness Experience, Distance, and 
Income 
Category Frequency Mean S.D. F-test Sig. Level 
Illness Experience 0.358 .550 
No 703 1,719 1,127 
Yes 388 1,762 1,116 
Distance 9.122 .003 
Village 562 1,635 1,151 
Outside 529 1,840 1,082 
Income 2.983 .007 
Under 5,000 28 1,282 654 
5,000-15,000 327 1,631 1,179 
15,000-30,000 386 1,708 1,127 
30,000-45,000 226 1,837 1,100 
45,000-60,000 80 1,972 1,068 
60,000-100,000 34 1,957 755 
Above 100,000 10 2,360 1,325 
Illness Experience 
Sixty-four percent of the respondents had not experienced a case of child 
pneumonia in the household. The other 36 percent had experienced a child pneumonia 
case in the past. One might expect those who had experienced a child pneumonia case to 
express a higher MWTP. The results, however, showed a small and nonsignificant 
difference between the mean values of the MWTP of the two groups. This might be an 
indication that the seriousness of child pneumonia was well understood by those who had 
not yet experienced a case. 
Distance to Source 
Fifty-two percent of the respondents reported that they could obtain antibiotics 
within the village vicinity. The rest claimed that they had to travel to a rayon center or 
11
Figure 2. MWTP by distance
1635
1840
0
500
1000
1500
2000
Village Outside
Pneumonia
MWTP Value (in Sum) 
Figure 3. MWTP by income
1282
1631 1708
1837
1979 1957
2360
0
500
1000
1500
2000
2500
Level 1 Level 2 Level 3 Level 4 Level 5 Level 6 Level 7
Pneumonia
MWTP Value (in Sum) 
Income Level 
Nukus City to obtain the same drugs. Since travel outside the village has money and time 
costs, one might expect that the latter group would be willing to pay a higher price for the 
opportunity to purchase the drugs in a village pharmacy. Results in Table 3and Figure 2 
seem to support this hypothesis. There is a large and significant difference between the 
means of MWTP for the two groups. On average, those who had to travel outside the 
community to get the drugs were willing to pay 204 Sum more. 
Figure 2. MWTP, by Distance, for Pneumonia 
Monthly Household Income 
Economic theory suggests that there is a positive relationship between WTP and 
household income. Results reported in Table 3 and Figure 3 reveal that there is a clear 
pattern between the monthly income of the household and the WTP: the higher the 
income, the higher the WTP. The F-test of all the means is significant at the .007 level. 
Figure 3. MWTP, by Household Income, for Pneumonia 
12
4.1.2 Determinants of Willingness to Pay 
We carried out regression analyses to identify the main determinants of the 
willingness to pay and their individual and total impact on WTP. Table 4 presents the 
results of a regression analysis with three independent variables. 
Table 4. Regression on the Determinants of WTP for Pneumonia Antibiotics 
 Coefficients S.D. t-Test Sig. Level R-Square F-Test 
Model .023 8.666 
Constant 1,244.09 106.48 11.68 .000 
Income 115.50 29.37 3.93 .000 
Distance 210.32 68.29 3.08 .002 
Illness 
Experience 
73.84 71.29 1.03 .300 
Both household income and the distance factor have the expected sign and are 
highly significant. The illness experience factor has the expected sign that those who had 
experienced the illness were willing to pay more. However, the coefficient is small and 
not significant. This seems to suggest that those who had no experience of child 
pneumonia were equally aware of its danger. 
Section 4.2: Analysis of the Subgroup with Illness Experience 
To further our effort to identify factors that are important in determining WTP, 
this section focuses only on those who had previous experience with child pneumonia 
cases. In our survey, 388 households had had child pneumonia cases in the past. 
4.2.1 Willingness to Pay, by Distance, Income, and Payment 
Table 5 provides the differences in mean values of WTP by distance, income, and 
whether the household had paid for medicine in the past. Details of the statistical analysis 
are reported in Appendix C. 
Table 5. Pneumonia Maximum Willingness to Pay, by Distance, Income, and Payment 
Category Frequency Mean S.D. F-test Sig. Level 
Distance 11.815 .001 
Village 244 1,614 995 
Outside 144 2,012 1,259 
Income 1.369 .226 
Under 5,000 8 1,393 935 
5,000-15,000 118 1,693 1,181 
15,000-30,000 131 1,667 1,055 
13
Figure 4. MWTP by Paying Experience
1525
2030
0
500
1000
1500
2000
2500
No Paid
Pneumonia
MWTP Value (in Sum) 
Payers Nonpayers
Category Frequency Mean S.D. F-test Sig. Level 
30,000-45,000 83 1,891 1,119 
45,000-60,000 33 1,917 1,025 
60,000-100,000 11 2,073 1,001 
Above 100,000 4 2,775 1,953 
Payment 20.834 .000 
No 206 1,525 945 
Yes 182 2,030 1,231 
Distance to Source 
Sixty-three percent of the respondents reported that they could obtain the 
antibiotic drugs within the village vicinity and the rest claimed that they had to travel to 
the rayon center or Nukus City to obtain the same drug. As shown in Table 5, the results 
again support the hypothesis that those who had to travel to a rayon center or to Nukus 
City to obtain a drug would be willing to pay more for the opportunity to purchase the 
same drug in a village pharmacy. There is a large and significant difference between the 
means of MWTP for the two groups. On the average, those who had to travel outside the 
community to get the drug were willing to pay 398 Sum more. 
Monthly Household Income 
Table 5 shows a clear pattern of positive relationships between the monthly 
income of the household and the WTP—i.e., the higher the income, the higher the WTP. 
The F-test of all the means is significant at the .226 level. 
Payment for Medicine 
Within this group, about half paid for antibiotics and the other half did not. Other 
WTP studies have shown that those who had not paid for services or commodities usually 
expressed a lower WTP. Our results support this hypothesis. Results in Table 5 and 
Figure 4 indicate that the mean differences of WTP in these two groups are large and 
significant. Those who actually paid for drugs, on the average, would be willing to pay 
500 Sum more. 
Figure 4. MWTP, by Paying Experience, for Pneumonia 
14
4.2.2 Determinants of Willingness to Pay 
Table 6 reports the results of the regression analysis of WTP. Income remains a 
strong and significant factor in determining MWTP. The distance variable still has the 
expected sign but is no longer significant. One the other hand, whether the respondents 
paid for medicine influenced how much they were willing to pay for drugs in the 
proposed village pharmacy. The coefficient for payment is larger and highly significant. 
Table 6. Regression on the Determinants of WTP for Pneumonia Antibiotics 
 Coefficients S.D. t-Test Sig. Level R-Square F-Test 
Model .068 9.319 
Constant 1,127.144 169.167 6.645 .000 
Income 123.911 47.510 2.605 .010 
Distance 50.408 164.050 0.307 .759 
Payment 476.573 158.796 3.001 .003 
Part Five: Analysis of Diarrhea Survey 
Section 5.1: Analysis of the Whole Group 
Analysis of the whole group from the diarrhea survey reveals results similar to 
those of the pneumonia survey, with minor variations. 
5.1.1 Willingness to Pay, by Experience of Illness, Distance to Source, and 
Household Income 
Table 7 shows the distribution of diarrhea MWTP by illness experience, distance 
to source of drug, and household income. Details of the statistical analysis are reported in 
Appendix D. 
Table 7. Diarrhea Maximum Willingness to Pay, by Illness Experience, Distance, and 
Income 
Category Frequency Mean S.D. F-Test Sig. Level 
Illness Experience 0.486 .486 
No 819 1,625 1,049 
Yes 59 1,724 1,093 
Distance 1.643 .200 
Village 455 1,588 1,102 
Outside 423 1,679 994 
Income 2.559 .018 
Under 5,000 21 1,133 525 
15
Figure 2. MWTP by income
0
500
1000
1500
2000
2500
Level 1 Level 2 Level 3 Level 4 Level 5 Level 6 Level 7
Diarrhea
MWTP Value (in Sum)
Income Level
Category Frequency Mean S.D. F-Test Sig. Level 
5,000-15,000 271 1,549 1,053 
15,000-30,000 304 1,608 1,000 
30,000-45,000 180 1,691 1,217 
45,000-60,000 68 1,925 996 
60,000-100,000 26 1,854 629 
Above 100,000 8 2,106 764 
Illness Experience 
Again, the results of analyzing the diarrhea survey show a small and nonsigni￾ficant difference between the mean values of MWTP of the two groups. 
Distance to Source 
As shown in Table 7, there is a large but not significant difference between the 
means of MWTP for those who could obtain the antibiotic drug within the village vicinity 
and those who claimed that they had to travel to a rayon center or to Nukus City to obtain 
the same drug. On average, those who had to travel outside the community to get the drug 
were willing to pay 91 Sum more. 
Monthly Household Income 
Results reported in Table 7 and Figure 5 again reveal a clear pattern between the 
monthly income of the household and the WTP; the higher the income, the higher the 
WTP. The F-test of all the means is significant at the .018 level. 
Figure 5. MWTP, by Household Income, for Diarrhea 
16
5.1.2 Determinants of Willingness to Pay 
We carried out regression analyses to identify the main determinants of the 
willingness to pay and their individual and total impact on WTP. Table 8 presents the 
results of a regression analysis with three independent variables. All the coefficients have 
the expected sign but only income is significant. The income coefficient is similar to that 
from the pneumonia survey. This is another confirmation that the type of illness did not 
affect respondents’ MWTP for antibiotics. 
Table 8. Regression on the Determinants of WTP for Diarrhea Antibiotics 
 Coefficients S.D. t-Test Sig. Level R-Square F-Test 
Model .016 4.843 
Constant 1,246.52 107.52 11.59 .000 
Income 108.03 30.76 3.51 .000 
Distance 85.82 70.60 1.21 .224 
Illness 
Experience 
91.98 140.89 0.65 .514 
Section 5.2: Analysis of the Subgroup with Illness Experience 
This section focuses only on those who had previous experience with at least one 
child diarrhea case. In our survey, there were 59 households that had had child diarrhea 
cases in the past. 
5.2.1 Willingness to Pay, by Distance, Income, and Payment 
Table 9 shows the differences in mean values of WTP by distance, income, and 
whether the household had paid for medicine in the past. Details of the statistical analysis 
are reported in Appendix E. 
Table 9. Diarrhea Maximum Willingness to Pay, by Distance, Income, and Payment 
Category Frequency Mean S.D. F-Test Sig. Level 
Distance 0.001 .972 
Village 33 1,720 1,216 
Outside 26 1,730 939 
Income 0.949 .469 
Under 5,000 1 1,500 
5,000-15,000 18 1,489 981 
15,000-30,000 21 1,804 1,175 
30,000-45,000 8 1,713 1,406 
17
Figure 2. MWTP by Paying Experience
1089
1920
0
500
1000
1500
2000
2500
No Paid
Diarrhea
MWTP Value (in Sum) 
Payers Nonpayers 
Category Frequency Mean S.D. F-Test Sig. Level 
45,000-60,000 9 1,672 767 
60,000-100,000 1 3,300 
Above 100,000 1 3,500 
Payment for drugs 6.809 .012 
No 14 1,089 595 
Yes 45 1,921 1,142 
Distance to Source 
In this sub-sample, there is no difference between the means of MWTP for those 
who could obtain the antibiotic drug with in the village vicinity and those who claimed 
that they had to travel to a rayon center or to Nukus City to obtain the same drug. 
Monthly Household Income 
There is still a clear pattern between the monthly income of the household and the 
WTP; the higher the income, the higher the WTP. However, the F-test of all the means is 
not significant. 
Payment for Medicine 
Results show that those who did not pay for services or commodities expressed a 
lower WTP. In Table 9 and Figure 6, the mean differences of WTP in these two groups 
are large and significant. Those who actually paid for drugs, on the average, would be 
willing to pay 832 Sum more. 
Figure 6. MWTP, by Paying Experience, for Diarrhea 
18
5.2.2 Determinants of Willingness to Pay 
Table 10 reports the results of the regression analysis of WTP. The income 
coefficient remains in the same range as the earlier estimation, but it is not statistically 
significant. The distance variable is not significant. One the other hand, whether the 
respondent paid for medicine demonstrates again the strong influence on how much 
respondents were willing to pay for drugs in the proposed village pharmacy. 
Table 10. Regression on the Determinants of WTP for Diarrhea Antibiotics 
 Coefficients S.D. t-Test Sig. Level R-Square F-Test 
Model .068 9.319 
Constant 767.848 444.386 1.728 .090 
Income 122.029 114.995 1.061 .293 
Distance -127.009 288.675 -0.440 .662 
Payment 812.082 338.618 2.398 .020 
Part Six: Summary and Conclusions
Section 6.1: Clear and Strong Willingness to Pay 
From the analysis of the two surveys, the results clearly show a high willingness 
to pay for the convenience and reliability of buying antibiotics at a village pharmacy. 
From the whole sample to different subgroups, respondents consistently indicated that 
they valued having antibiotics available when needed and were willing to pay for them. 
The fact that the mean MWTP values for the different groups were higher than the 
prevailing market price of antibiotics reflects how strongly people felt the need to have 
this opportunity. 
Section 6.2: Factors Influencing Willingness to Pay 
The results seem to suggest that the MWTP for antibiotics was not influenced by 
respondents’ experience in dealing with the illness or the type of illness. On the other 
hand, whether the medicine could be obtained within the community seemed to be an 
important factor in determining how much a respondent was willing to pay. Those who 
had traveled to other rayon centers or to Nukus City to search for the medicine tended to 
favor a higher MWTP. For the group that had had experience with child pneumonia or 
diarrhea cases, whether they paid for the prescribed medicine turned out to be a very 
important factor in determining MWTP. 
The most important and consistent factor was household income level. The 
regression coefficient for income level remained positive and statistically significant for 
almost all the subgroups that we examined. Furthermore, the coefficient consistently 
19
stayed in the narrow range of 108 to 125. The pattern of a positive relationship between 
income level and MWTP is well supported by the survey. Given this relationship, it is 
reasonable to expect that future MWTP will rise as household income increases. 
Section 6.3: Financial Sustainability of the Village Pharmacy 
Since the mean value of MWTP was 1,689 Sum, which was 369 Sum higher than 
the commercial market price, there should be no shortage of demand if antibiotics were to 
be offered in village pharmacies at the same price as found in the market. If the price 
were set at 1,320 Sum, the distribution of MWTP (see Appendix A) suggests that at that 
price, 73 percent of our sample households would purchase antibiotics from a village 
clinic pharmacy. If the price of the antibiotic were set lower to reflect subsidies provided 
by the Child Survival Program, an even higher portion of the population would pay for 
them at a village pharmacy. For example, if the price were set at 660 Sum, then over 90 
percent of the households would participate in such an arrangement. Given such high 
potential demand at the near-market price, there seems no doubt that, assuming the 
village pharmacy were run with reasonable efficiency, the financial sustainability of the 
village pharmacy would be assured. 
Section 6.4: Improving the Welfare of the People 
By introducing village pharmacies and offering the antibiotics, the CSP and the 
Ministry of Health would improve the welfare of the local population. In addition to the 
obvious benefits to child health, there would also be substantial consumer welfare gains. 
By calculating the difference between the MWTP and the actual price, it is possible to 
estimate the consumer surplus generated. For example, if the price were set at 1,320 Sum 
per unit of antibiotics, the total consumer surplus would amount to 1 million Sum. This 
number is derived by adding the amount of MWTP exceeding 1,320 Sum of all the 
individuals who had indicated that they would be willing to pay at least 1,320 Sum for 
antibiotics. This total consumer gain would double to 2 million if the price were set at 
660 Sum. 
Section 6.5: Policy Recommendations 
The overwhelmingly positive results from our willingness-to-pay survey strongly 
support the establishment of pharmacies in village clinics. Having high-quality drugs 
available in the convenient setting of a village pharmacy would improve child survival 
and create consumer welfare gains. The strong potential demand for antibiotics at village 
pharmacies, together with uninterrupted supplies through donors’ support and 
coordination, would almost guarantee the success and the financial sustainability of the 
village pharmacies. 
20
References 
Foreit, Karen, and James Foreit. (September 2001). Willingness-to-pay surveys for setting 
prices for reproductive health products and services: A user’s manual.
Washington, DC: The Futures Group International and the Population Council. 
Foreit, James, and Karen Foreit. (2003). The reliability and validity of willingness to pay 
surveys for reproductive health pricing decisions in developing countries. Health 
Policy, 63: 37-47. 
Olsen, Jan Abel, and Richard Smith. (April 1999). Who has been asked to value what? A 
review of 54 “willingness-to-pay” surveys in health care. Working Paper 83. 
Clayton and Melbourne, Australia: Center for Health Program Evaluation, 
Monash University and University of Melbourne. 
Olsen, Jan Abel, Richard Smith, and Anthony Harris. (April 1999). Economic theory and 
the monetary valuation of health care: An evaluation of the issues as applied to 
the economic evaluation of health care programs. Working Paper 82. Clayton and 
Melbourne, Australia: Center for Health Program Evaluation, Monash University 
and University of Melbourne. 
A-1
Appendix A 
Statistics on Maximum Willingness to Pay
Statistics 
MWTP 
Valid 1969 N 
Missing 0
Mean 1688.6618
Median 1500.0000
Std. Deviation 1092.28214
Minimum 100.00
Maximum 10000.00
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
100.00 3 .2 .2 .2 
150.00 1 .1 .1 .2 
170.00 1 .1 .1 .3 
200.00 5 .3 .3 .5 
250.00 7 .4 .4 .9 
260.00 1 .1 .1 .9 
280.00 1 .1 .1 1.0 
300.00 15 .8 .8 1.7 
350.00 6 .3 .3 2.0 
375.00 1 .1 .1 2.1 
400.00 14 .7 .7 2.8 
420.00 1 .1 .1 2.8 
450.00 15 .8 .8 3.6 
475.00 1 .1 .1 3.7 
480.00 1 .1 .1 3.7 
500.00 82 4.2 4.2 7.9 
520.00 2 .1 .1 8.0 
550.00 10 .5 .5 8.5 
600.00 16 .8 .8 9.3 
650.00 4 .2 .2 9.5 
660.00 6 .3 .3 9.8 
670.00 6 .3 .3 10.1 
680.00 4 .2 .2 10.3 
700.00 75 3.8 3.8 14.1 
750.00 1 .1 .1 14.2 
770.00 1 .1 .1 14.2 
800.00 24 1.2 1.2 15.4 
Valid 
850.00 8 .4 .4 15.8 
A-2
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
900.00 13 .7 .7 16.5 
920.00 7 .4 .4 16.9 
930.00 1 .1 .1 16.9 
950.00 22 1.1 1.1 18.0 
1000.00 98 5.0 5.0 23.0 
1050.00 3 .2 .2 23.2 
1200.00 38 1.9 1.9 25.1 
1250.00 1 .1 .1 25.1 
1300.00 10 .5 .5 25.6 
1320.00 32 1.6 1.6 27.3 
1325.00 1 .1 .1 27.3 
1330.00 7 .4 .4 27.7 
1350.00 169 8.6 8.6 36.3 
1360.00 2 .1 .1 36.4 
1370.00 4 .2 .2 36.6 
1375.00 1 .1 .1 36.6 
1380.00 1 .1 .1 36.7 
1400.00 63 3.2 3.2 39.9 
1450.00 14 .7 .7 40.6 
1460.00 2 .1 .1 40.7 
1500.00 396 20.1 20.1 60.8 
1550.00 2 .1 .1 60.9 
1560.00 1 .1 .1 60.9 
1600.00 19 1.0 1.0 61.9 
1650.00 1 .1 .1 62.0 
1700.00 12 .6 .6 62.6 
1710.00 9 .5 .5 63.0 
1720.00 5 .3 .3 63.3 
1750.00 25 1.3 1.3 64.6 
1800.00 27 1.4 1.4 65.9 
1850.00 2 .1 .1 66.0 
1900.00 4 .2 .2 66.2 
1980.00 17 .9 .9 67.1 
2000.00 351 17.8 17.8 84.9 
2100.00 4 .2 .2 85.1 
2150.00 1 .1 .1 85.2 
2200.00 7 .4 .4 85.5 
2250.00 1 .1 .1 85.6 
2300.00 14 .7 .7 86.3 
2350.00 1 .1 .1 86.3 
2400.00 1 .1 .1 86.4 
2500.00 75 3.8 3.8 90.2 
2550.00 5 .3 .3 90.5 
A-3
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
2600.00 8 .4 .4 90.9 
2625.00 1 .1 .1 90.9 
2700.00 2 .1 .1 91.0 
2750.00 3 .2 .2 91.2 
2800.00 9 .5 .5 91.6 
3000.00 71 3.6 3.6 95.2 
3250.00 1 .1 .1 95.3 
3300.00 1 .1 .1 95.3 
3450.00 1 .1 .1 95.4 
3500.00 19 1.0 1.0 96.3 
3750.00 1 .1 .1 96.4 
4000.00 15 .8 .8 97.2 
4200.00 1 .1 .1 97.2 
4500.00 2 .1 .1 97.3 
5000.00 34 1.7 1.7 99.0 
6000.00 5 .3 .3 99.3 
7000.00 1 .1 .1 99.3 
10000.00 13 .7 .7 100.0 
Total 1969 100.0 100.0
B-1
Appendix B 
Statistical Analysis of Pneumonia Sample 
Statistics 
EXPERIEN DISTANCE INCOME MWTP 
Valid 1091 1091 1091 1091 N 
Missing 0 0 0 0 
Mean .3556 .4849 3.1329 1734.2117 
Median .0000 .0000 3.0000 1500.0000 
Std. Deviation .47893 .50000 1.14609 1122.39025 
Minimum .00 .00 1.00 100.00 
Maximum 1.00 1.00 7.00 10000.00 
Frequency Table 
EXPERIEN 
Frequency Percent Valid Percent Cumulative Percent 
.00 703 64.4 64.4 64.4 
Valid 1.00 388 35.6 35.6 100.0 
Total 1091 100.0 100.0
DISTANCE 
Frequency Percent Valid Percent Cumulative Percent 
.00 562 51.5 51.5 51.5 
Valid 1.00 529 48.5 48.5 100.0 
Total 1091 100.0 100.0
INCOME 
Frequency Percent Valid Percent Cumulative Percent 
1.00 28 2.6 2.6 2.6 
2.00 327 30.0 30.0 32.5 
3.00 386 35.4 35.4 67.9 
4.00 226 20.7 20.7 88.6 
5.00 80 7.3 7.3 96.0 
6.00 34 3.1 3.1 99.1 
7.00 10 .9 .9 100.0 
Valid 
Total 1091 100.0 100.0
MWTP 
B-2
Frequency Percent Valid Percent Cumulative Percent 
100.00 1 .1 .1 .1 
150.00 1 .1 .1 .2 
170.00 1 .1 .1 .3 
200.00 2 .2 .2 .5 
250.00 2 .2 .2 .6 
260.00 1 .1 .1 .7 
300.00 7 .6 .6 1.4 
350.00 3 .3 .3 1.6 
375.00 1 .1 .1 1.7 
400.00 5 .5 .5 2.2 
420.00 1 .1 .1 2.3 
450.00 13 1.2 1.2 3.5 
475.00 1 .1 .1 3.6 
480.00 1 .1 .1 3.7 
500.00 42 3.8 3.8 7.5 
520.00 2 .2 .2 7.7 
550.00 7 .6 .6 8.3 
600.00 10 .9 .9 9.3 
650.00 2 .2 .2 9.4 
660.00 3 .3 .3 9.7 
670.00 3 .3 .3 10.0 
680.00 2 .2 .2 10.2 
700.00 42 3.8 3.8 14.0 
770.00 1 .1 .1 14.1 
800.00 11 1.0 1.0 15.1 
850.00 8 .7 .7 15.9 
900.00 9 .8 .8 16.7 
920.00 3 .3 .3 17.0 
930.00 1 .1 .1 17.0 
950.00 12 1.1 1.1 18.1 
1000.00 51 4.7 4.7 22.8 
1050.00 3 .3 .3 23.1 
1200.00 21 1.9 1.9 25.0 
1250.00 1 .1 .1 25.1 
1300.00 10 .9 .9 26.0 
1320.00 17 1.6 1.6 27.6 
1325.00 1 .1 .1 27.7 
1330.00 3 .3 .3 28.0 
1350.00 77 7.1 7.1 35.0 
1360.00 1 .1 .1 35.1 
1370.00 2 .2 .2 35.3 
1375.00 1 .1 .1 35.4 
1400.00 39 3.6 3.6 39.0 
Valid 
1450.00 9 .8 .8 39.8 
B-3
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
1460.00 1 .1 .1 39.9 
1500.00 204 18.7 18.7 58.6 
1550.00 2 .2 .2 58.8 
1560.00 1 .1 .1 58.8 
1600.00 10 .9 .9 59.8 
1650.00 1 .1 .1 59.9 
1700.00 9 .8 .8 60.7 
1710.00 2 .2 .2 60.9 
1720.00 3 .3 .3 61.1 
1750.00 8 .7 .7 61.9 
1800.00 14 1.3 1.3 63.2 
1850.00 2 .2 .2 63.3 
1900.00 2 .2 .2 63.5 
1980.00 8 .7 .7 64.3 
2000.00 195 17.9 17.9 82.1 
2100.00 3 .3 .3 82.4 
2150.00 1 .1 .1 82.5 
2200.00 5 .5 .5 83.0 
2250.00 1 .1 .1 83.0 
2300.00 13 1.2 1.2 84.2 
2350.00 1 .1 .1 84.3 
2400.00 1 .1 .1 84.4 
2500.00 43 3.9 3.9 88.4 
2550.00 3 .3 .3 88.6 
2600.00 5 .5 .5 89.1 
2625.00 1 .1 .1 89.2 
2700.00 2 .2 .2 89.4 
2750.00 2 .2 .2 89.6 
2800.00 7 .6 .6 90.2 
3000.00 46 4.2 4.2 94.4 
3250.00 1 .1 .1 94.5 
3450.00 1 .1 .1 94.6 
3500.00 11 1.0 1.0 95.6 
3750.00 1 .1 .1 95.7 
4000.00 13 1.2 1.2 96.9 
4500.00 2 .2 .2 97.1 
5000.00 21 1.9 1.9 99.0 
6000.00 3 .3 .3 99.3 
7000.00 1 .1 .1 99.4 
10000.00 7 .6 .6 100.0 
Total 1091 100.0 100.0
B-4
Mean Analysis 
Case Processing Summary 
Cases 
Included Excluded Total 
N Percent N Percent N Percent 
MWTP * EXPERIEN 1091 100.0% 0 .0% 1091 100.0% 
MWTP * DISTANCE 1091 100.0% 0 .0% 1091 100.0% 
MWTP * INCOME 1091 100.0% 0 .0% 1091 100.0% 
MWTP * EXPERIEN 
Report 
MWTP 
EXPERIEN Mean N Std. Deviation
.00 1719.1110 703 1126.51517
1.00 1761.5722 388 1115.80603
Total 1734.2117 1091 1122.39025
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 450761.765 1 450761.765 .358 .550
Within Groups 1372687510.325 1089 1260502.764 
MWTP * 
EXPERIEN 
Total 1373138272.090 1090
Measures of Association 
Eta Eta Squared
MWTP * EXPERIEN .018 .000
MWTP * DISTANCE 
Report 
MWTP 
DISTANCE Mean N Std. Deviation
.00 1635.0089 562 1150.93845
1.00 1839.6030 529 1082.39866
Total 1734.2117 1091 1122.39025
ANOVA Table 
Sum of Squares df Mean Square F Sig.
B-5
Between 
Groups (Combined) 11406530.499 1 11406530.499 9.122 .003
Within Groups 1361731741.591 1089 1250442.371 
MWTP * 
DISTANCE 
Total 1373138272.090 1090
Measures of Association 
Eta Eta Squared
MWTP * DISTANCE .091 .008
MWTP * INCOME 
Report 
MWTP 
INCOME Mean N Std. Deviation
1.00 1281.7857 28 653.79266
2.00 1630.7798 327 1179.18765
3.00 1707.7073 386 1126.82285
4.00 1837.3009 226 1100.39155
5.00 1979.1875 80 1067.86651
6.00 1956.7647 34 755.09069
7.00 2360.0000 10 1325.14150
Total 1734.2117 1091 1122.39025
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 22303719.464 6 3717286.577 2.983 .007
Within Groups 1350834552.626 1084 1246157.336 
MWTP * 
INCOME 
Total 1373138272.090 1090
Measures of Association 
Eta Eta Squared
MWTP * INCOME .127 .016
Regression Analysis 
Variables Entered/Removed(b) 
Model Variables Entered Variables Removed Method 
1 INCOME, EXPERIEN, DISTANCE(a) . Enter 
a All requested variables entered. 
b Dependent Variable: MWTP 
B-6
Model Summary 
Model R R Square Adjusted R Square Std. Error of the Estimate 
1 .153(a) .023 .021 1110.73340 
a Predictors: (Constant), INCOME, EXPERIEN, DISTANCE 
ANOVA(b) 
Model Sum of Squares df Mean Square F Sig. 
Regression 32075180.437 3 10691726.812 8.666 .000(a) 
1 Residual 1341063091.652 1087 1233728.695
Total 1373138272.090 1090
a Predictors: (Constant), INCOME, EXPERIEN, DISTANCE 
b Dependent Variable: MWTP 
Coefficients(a) 
Unstandardized Coefficients Standardized Coefficients 
Model 
B Std. Error 
Beta t Sig.
(Constant) 1244.093 106.482 11.684 .000
EXPERIEN 73.844 71.296 .032 1.036 .301
DISTANCE 210.325 68.295 .094 3.080 .002
1 
INCOME 115.508 29.372 .118 3.933 .000
a Dependent Variable: MWTP 
C-1
Appendix C 
Statistical Analysis of Pneumonia Sample, with 
Households with Experience of Illness 
Statistics 
DISTANCE INCOME PAYMENT MWTP 
Valid 388 388 388 388 N 
Missing 0 0 0 0 
Mean .3711 3.1649 .4691 1761.5722 
Median .0000 3.0000 .0000 1500.0000 
Std. Deviation .48373 1.15631 .49969 1115.80603 
Minimum .00 1.00 .00 170.00 
Maximum 1.00 7.00 1.00 10000.00 
Frequency Table 
DISTANCE 
Frequency Percent Valid Percent Cumulative Percent 
.00 244 62.9 62.9 62.9 
Valid 1.00 144 37.1 37.1 100.0 
Total 388 100.0 100.0
INCOME 
Frequency Percent Valid Percent Cumulative Percent 
1.00 8 2.1 2.1 2.1 
2.00 118 30.4 30.4 32.5 
3.00 131 33.8 33.8 66.2 
4.00 83 21.4 21.4 87.6 
5.00 33 8.5 8.5 96.1 
6.00 11 2.8 2.8 99.0 
7.00 4 1.0 1.0 100.0 
Valid 
Total 388 100.0 100.0
PAYMENT 
Frequency Percent Valid Percent Cumulative Percent 
.00 206 53.1 53.1 53.1 
Valid 1.00 182 46.9 46.9 100.0 
Total 388 100.0 100.0
C-2
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
170.00 1 .3 .3 .3 
200.00 1 .3 .3 .5 
260.00 1 .3 .3 .8 
300.00 2 .5 .5 1.3 
375.00 1 .3 .3 1.5 
400.00 1 .3 .3 1.8 
420.00 1 .3 .3 2.1 
450.00 8 2.1 2.1 4.1 
475.00 1 .3 .3 4.4 
480.00 1 .3 .3 4.6 
500.00 17 4.4 4.4 9.0 
520.00 2 .5 .5 9.5 
550.00 3 .8 .8 10.3 
600.00 6 1.5 1.5 11.9 
660.00 1 .3 .3 12.1 
670.00 1 .3 .3 12.4 
700.00 17 4.4 4.4 16.8 
770.00 1 .3 .3 17.0 
800.00 4 1.0 1.0 18.0 
850.00 7 1.8 1.8 19.8 
900.00 8 2.1 2.1 21.9 
950.00 2 .5 .5 22.4 
1000.00 23 5.9 5.9 28.4 
1050.00 3 .8 .8 29.1 
1200.00 6 1.5 1.5 30.7 
1250.00 1 .3 .3 30.9 
1300.00 9 2.3 2.3 33.2 
1320.00 4 1.0 1.0 34.3 
1330.00 1 .3 .3 34.5 
1350.00 19 4.9 4.9 39.4 
1370.00 1 .3 .3 39.7 
1375.00 1 .3 .3 39.9 
1400.00 16 4.1 4.1 44.1 
1450.00 2 .5 .5 44.6 
1500.00 41 10.6 10.6 55.2 
1550.00 2 .5 .5 55.7 
1560.00 1 .3 .3 55.9 
1600.00 4 1.0 1.0 57.0 
1650.00 1 .3 .3 57.2 
1700.00 9 2.3 2.3 59.5 
1750.00 1 .3 .3 59.8 
1800.00 5 1.3 1.3 61.1 
Valid 
1850.00 1 .3 .3 61.3 
C-3
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
1900.00 1 .3 .3 61.6 
1980.00 1 .3 .3 61.9 
2000.00 37 9.5 9.5 71.4 
2100.00 3 .8 .8 72.2 
2150.00 1 .3 .3 72.4 
2200.00 4 1.0 1.0 73.5 
2250.00 1 .3 .3 73.7 
2300.00 12 3.1 3.1 76.8 
2350.00 1 .3 .3 77.1 
2400.00 1 .3 .3 77.3 
2500.00 23 5.9 5.9 83.2 
2550.00 3 .8 .8 84.0 
2600.00 5 1.3 1.3 85.3 
2625.00 1 .3 .3 85.6 
2700.00 1 .3 .3 85.8 
2750.00 2 .5 .5 86.3 
2800.00 7 1.8 1.8 88.1 
3000.00 15 3.9 3.9 92.0 
3250.00 1 .3 .3 92.3 
3450.00 1 .3 .3 92.5 
3500.00 8 2.1 2.1 94.6 
3750.00 1 .3 .3 94.8 
4000.00 9 2.3 2.3 97.2 
4500.00 2 .5 .5 97.7 
5000.00 5 1.3 1.3 99.0 
6000.00 2 .5 .5 99.5 
7000.00 1 .3 .3 99.7 
10000.00 1 .3 .3 100.0 
Total 388 100.0 100.0
Mean Analysis 
Case Processing Summary 
Cases 
Included Excluded Total 
N Percent N Percent N Percent 
MWTP * DISTANCE 388 100.0% 0 .0% 388 100.0% 
MWTP * INCOME 388 100.0% 0 .0% 388 100.0% 
MWTP * PAYMENT 388 100.0% 0 .0% 388 100.0% 
MWTP * DISTANCE 
C-4
Report 
MWTP 
DISTANCE Mean N Std. Deviation
.00 1614.0369 244 995.22791
1.00 2011.5625 144 1259.44619
Total 1761.5722 388 1115.80603
ANOVA Table 
Sum of 
Squares df Mean Square F Sig.
Between 
Groups (Combined) 14310368.874 1 14310368.874 11.815 .001
Within Groups 467513572.106 386 1211175.057 
MWTP * 
DISTANCE 
Total 481823940.979 387
Measures of Association 
Eta Eta Squared
MWTP * DISTANCE .172 .030
MWTP * INCOME 
Report 
MWTP 
INCOME Mean N Std. Deviation
1.00 1392.5000 8 934.98281
2.00 1693.3475 118 1180.60358
3.00 1667.2519 131 1054.85894
4.00 1891.3253 83 1119.10000
5.00 1916.5152 33 1025.23662
6.00 2072.7273 11 1001.34001
7.00 2775.0000 4 1953.41581
Total 1761.5722 388 1115.80603
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 10167130.897 6 1694521.816 1.369 .226
Within Groups 471656810.082 381 1237944.383 
MWTP * 
INCOME 
Total 481823940.979 387
Measures of Association 
Eta Eta Squared
C-5
MWTP * INCOME .145 .021
MWTP * PAYMENT 
Report 
MWTP 
PAYMENT Mean N Std. Deviation
.00 1524.5388 206 944.86954
1.00 2029.8626 182 1230.66252
Total 1761.5722 388 1115.80603
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 24674388.224 1 24674388.224 20.834 .000
Within Groups 457149552.755 386 1184325.266 
MWTP * 
PAYMENT 
Total 481823940.979 387
Measures of Association 
Eta Eta Squared
MWTP * PAYMENT .226 .051
Regression Analysis 
Variables Entered/Removed(b) 
Model Variables Entered Variables Removed Method 
1 PAYMENT, INCOME, DISTANCE(a) . Enter 
a All requested variables entered. 
b Dependent Variable: MWTP 
Model Summary 
Model R R Square Adjusted R Square Std. Error of the Estimate 
1 .261(a) .068 .061 1081.48042 
a Predictors: (Constant), PAYMENT, INCOME, DISTANCE 
ANOVA(b) 
Model Sum of Squares df Mean Square F Sig. 
Regression 32697583.638 3 10899194.546 9.319 .000(a) 
1 Residual 449126357.341 384 1169599.889
Total 481823940.979 387
a Predictors: (Constant), PAYMENT, INCOME, DISTANCE 
b Dependent Variable: MWTP 
C-6
Coefficients(a) 
Unstandardized Coefficients Standardized Coefficients 
Model 
B Std. Error 
Beta t Sig.
(Constant) 1127.144 169.617 6.645 .000
DISTANCE 50.408 164.050 .022 .307 .759
INCOME 123.911 47.560 .128 2.605 .010
1 
PAYMENT 476.573 158.796 .213 3.001 .003
a Dependent Variable: MWTP 
D-1
Appendix D 
Statistical Analysis of Diarrhea Sample 
Statistics 
EXPERIEN DISTANCE INCOME MWTP 
Valid 878 878 878 878 N 
Missing 0 0 0 0 
Mean .0672 .4818 3.1287 1632.0615 
Median .0000 .0000 3.0000 1500.0000 
Std. Deviation .25051 .49995 1.14733 1051.55983 
Minimum .00 .00 1.00 100.00 
Maximum 1.00 1.00 7.00 10000.00 
Frequency Table 
EXPERIEN 
Frequency Percent Valid Percent Cumulative Percent 
.00 819 93.3 93.3 93.3 
Valid 1.00 59 6.7 6.7 100.0 
Total 878 100.0 100.0
DISTANCE 
Frequency Percent Valid Percent Cumulative Percent 
.00 455 51.8 51.8 51.8 
Valid 1.00 423 48.2 48.2 100.0 
Total 878 100.0 100.0
INCOME 
Frequency Percent Valid Percent Cumulative Percent 
1.00 21 2.4 2.4 2.4 
2.00 271 30.9 30.9 33.3 
3.00 304 34.6 34.6 67.9 
4.00 180 20.5 20.5 88.4 
5.00 68 7.7 7.7 96.1 
6.00 26 3.0 3.0 99.1 
7.00 8 .9 .9 100.0 
Valid 
Total 878 100.0 100.0
D-2
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
100.00 2 .2 .2 .2 
200.00 3 .3 .3 .6 
250.00 5 .6 .6 1.1 
280.00 1 .1 .1 1.3 
300.00 8 .9 .9 2.2 
350.00 3 .3 .3 2.5 
400.00 9 1.0 1.0 3.5 
450.00 2 .2 .2 3.8 
500.00 40 4.6 4.6 8.3 
550.00 3 .3 .3 8.7 
600.00 6 .7 .7 9.3 
650.00 2 .2 .2 9.6 
660.00 3 .3 .3 9.9 
670.00 3 .3 .3 10.3 
680.00 2 .2 .2 10.5 
700.00 33 3.8 3.8 14.2 
750.00 1 .1 .1 14.4 
800.00 13 1.5 1.5 15.8 
900.00 4 .5 .5 16.3 
920.00 4 .5 .5 16.7 
950.00 10 1.1 1.1 17.9 
1000.00 47 5.4 5.4 23.2 
1200.00 17 1.9 1.9 25.2 
1320.00 15 1.7 1.7 26.9 
1330.00 4 .5 .5 27.3 
1350.00 92 10.5 10.5 37.8 
1360.00 1 .1 .1 37.9 
1370.00 2 .2 .2 38.2 
1380.00 1 .1 .1 38.3 
1400.00 24 2.7 2.7 41.0 
1450.00 5 .6 .6 41.6 
1460.00 1 .1 .1 41.7 
1500.00 192 21.9 21.9 63.6 
1600.00 9 1.0 1.0 64.6 
1700.00 3 .3 .3 64.9 
1710.00 7 .8 .8 65.7 
1720.00 2 .2 .2 65.9 
1750.00 17 1.9 1.9 67.9 
1800.00 13 1.5 1.5 69.4 
1900.00 2 .2 .2 69.6 
1980.00 9 1.0 1.0 70.6 
2000.00 156 17.8 17.8 88.4 
Valid 
2100.00 1 .1 .1 88.5 
D-3
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
2200.00 2 .2 .2 88.7 
2300.00 1 .1 .1 88.8 
2500.00 32 3.6 3.6 92.5 
2550.00 2 .2 .2 92.7 
2600.00 3 .3 .3 93.1 
2750.00 1 .1 .1 93.2 
2800.00 2 .2 .2 93.4 
3000.00 25 2.8 2.8 96.2 
3300.00 1 .1 .1 96.4 
3500.00 8 .9 .9 97.3 
4000.00 2 .2 .2 97.5 
4200.00 1 .1 .1 97.6 
5000.00 13 1.5 1.5 99.1 
6000.00 2 .2 .2 99.3 
10000.00 6 .7 .7 100.0 
Total 878 100.0 100.0
Mean Analysis 
Case Processing Summary 
Cases 
Included Excluded Total 
N Percent N Percent N Percent 
MWTP * EXPERIEN 878 100.0% 0 .0% 878 100.0% 
MWTP * DISTANCE 878 100.0% 0 .0% 878 100.0% 
MWTP * INCOME 878 100.0% 0 .0% 878 100.0% 
MWTP * EXPERIEN 
Report 
MWTP 
EXPERIEN Mean N Std. Deviation
.00 1625.4212 819 1048.85590
1.00 1724.2373 59 1093.46996
Total 1632.0615 878 1051.55983
ANOVA Table 
Sum of Squares df Mean Square F Sig.
MWTP * 
EXPERIEN 
Between 
Groups (Combined) 537398.331 1 537398.331 .486 .486
D-4
Within Groups 969229970.348 876 1106426.907 
Total 969767368.679 877
Measures of Association 
Eta Eta Squared
MWTP * EXPERIEN .024 .001
D-5
MWTP * DISTANCE 
Report 
MWTP 
DISTANCE Mean N Std. Deviation
.00 1588.2198 455 1101.92668
1.00 1679.2199 423 993.68390
Total 1632.0615 878 1051.55983
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 1815268.104 1 1815268.104 1.643 .200
Within Groups 967952100.575 876 1104968.151 
MWTP * 
DISTANCE 
Total 969767368.679 877
Measures of Association 
Eta Eta Squared
MWTP * DISTANCE .043 .002
MWTP * INCOME 
Report 
MWTP 
INCOME Mean N Std. Deviation
1.00 1133.3333 21 525.19838
2.00 1549.4834 271 1052.83520
3.00 1608.1908 304 999.51425
4.00 1690.9444 180 1216.66579
5.00 1925.4412 68 996.37318
6.00 1853.8462 26 629.11415
7.00 2106.2500 8 763.65545
Total 1632.0615 878 1051.55983
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 16799240.309 6 2799873.385 2.559 .018
Within Groups 952968128.370 871 1094108.069 
MWTP * 
INCOME 
Total 969767368.679 877
Measures of Association 
Eta Eta Squared
D-6
MWTP * INCOME .132 .017
Regression Analysis 
Variables Entered/Removed(b) 
Model Variables Entered Variables Removed Method 
1 INCOME, EXPERIEN, DISTANCE(a) . Enter 
a All requested variables entered. 
b Dependent Variable: MWTP 
Model Summary 
Model R R Square Adjusted R Square Std. Error of the Estimate 
1 .128(a) .016 .013 1044.71566 
a Predictors: (Constant), INCOME, EXPERIEN, DISTANCE 
ANOVA(b) 
Model Sum of Squares df Mean Square F Sig. 
Regression 15856846.849 3 5285615.616 4.843 .002(a) 
1 Residual 953910521.829 874 1091430.803
Total 969767368.679 877
a Predictors: (Constant), INCOME, EXPERIEN, DISTANCE 
b Dependent Variable: MWTP 
Coefficients(a) 
Unstandardized Coefficients Standardized Coefficients 
Model 
B Std. Error 
Beta t Sig.
(Constant) 1246.521 107.527 11.593 .000
EXPERIEN 91.984 140.893 .022 .653 .514
DISTANCE 85.828 70.602 .041 1.216 .224
1 
INCOME 108.035 30.765 .118 3.512 .000
a Dependent Variable: MWTP 
E-1
Appendix E 
Statistical Analysis of Pneumonia Sample, with Households 
with Experience of Illness 
Statistics 
DISTANCE INCOME PAYMENT MWTP 
Valid 59 59 59 59 N 
Missing 0 0 0 0 
Mean .4407 3.2203 .7627 1724.2373 
Median .0000 3.0000 1.0000 1500.0000 
Std. Deviation .50073 1.23271 .42907 1093.46996 
Minimum .00 1.00 .00 200.00 
Maximum 1.00 7.00 1.00 4200.00 
Frequency Table 
DISTANCE 
Frequency Percent Valid Percent Cumulative Percent 
.00 33 55.9 55.9 55.9 
Valid 1.00 26 44.1 44.1 100.0 
Total 59 100.0 100.0
INCOME 
Frequency Percent Valid Percent Cumulative Percent 
1.00 1 1.7 1.7 1.7 
2.00 18 30.5 30.5 32.2 
3.00 21 35.6 35.6 67.8 
4.00 8 13.6 13.6 81.4 
5.00 9 15.3 15.3 96.6 
6.00 1 1.7 1.7 98.3 
7.00 1 1.7 1.7 100.0 
Valid 
Total 59 100.0 100.0
PAYMENT 
Frequency Percent Valid Percent Cumulative Percent 
.00 14 23.7 23.7 23.7 
Valid 1.00 45 76.3 76.3 100.0 
Total 59 100.0 100.0
E-2
MWTP 
Frequency Percent Valid Percent Cumulative Percent 
200.00 1 1.7 1.7 1.7 
250.00 2 3.4 3.4 5.1 
280.00 1 1.7 1.7 6.8 
300.00 1 1.7 1.7 8.5 
350.00 1 1.7 1.7 10.2 
500.00 4 6.8 6.8 16.9 
550.00 1 1.7 1.7 18.6 
600.00 1 1.7 1.7 20.3 
650.00 1 1.7 1.7 22.0 
700.00 2 3.4 3.4 25.4 
900.00 1 1.7 1.7 27.1 
1000.00 8 13.6 13.6 40.7 
1350.00 3 5.1 5.1 45.8 
1500.00 3 5.1 5.1 50.8 
1600.00 1 1.7 1.7 52.5 
1700.00 1 1.7 1.7 54.2 
1900.00 1 1.7 1.7 55.9 
2000.00 4 6.8 6.8 62.7 
2100.00 1 1.7 1.7 64.4 
2200.00 2 3.4 3.4 67.8 
2500.00 3 5.1 5.1 72.9 
2550.00 2 3.4 3.4 76.3 
2600.00 3 5.1 5.1 81.4 
2750.00 1 1.7 1.7 83.1 
2800.00 2 3.4 3.4 86.4 
3300.00 1 1.7 1.7 88.1 
3500.00 5 8.5 8.5 96.6 
4000.00 1 1.7 1.7 98.3 
4200.00 1 1.7 1.7 100.0 
Valid 
Total 59 100.0 100.0
Mean Analysis 
Case Processing Summary 
Cases 
Included Excluded Total 
N Percent N Percent N Percent 
MWTP * DISTANCE 59 100.0% 0 .0% 59 100.0% 
MWTP * INCOME 59 100.0% 0 .0% 59 100.0% 
MWTP * PAYMENT 59 100.0% 0 .0% 59 100.0% 
E-3
MWTP * DISTANCE 
Report 
MWTP 
DISTANCE Mean N Std. Deviation
.00 1719.6970 33 1216.09900
1.00 1730.0000 26 938.57339
Total 1724.2373 59 1093.46996
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 1543.708 1 1543.708 .001 .972
Within Groups 69347696.970 57 1216626.263 
MWTP * 
DISTANCE 
Total 69349240.678 58
Measures of Association 
Eta Eta Squared
MWTP * DISTANCE .005 .000
MWTP * INCOME 
Report 
MWTP 
INCOME Mean N Std. Deviation
1.00 1500.0000 1 .
2.00 1488.8889 18 980.77934
3.00 1803.8095 21 1175.05522
4.00 1712.5000 8 1405.79159
5.00 1672.2222 9 766.93836
6.00 3300.0000 1 .
7.00 3500.0000 1 .
Total 1724.2373 59 1093.46996
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between Groups (Combined) 6842062.107 6 1140343.684 .949 .469
MWTP * INCOME Within Groups 62507178.571 52 1202061.126 
Total 69349240.678 58
Measures of Association 
Eta Eta Squared
E-4
MWTP * INCOME .314 .099
MWTP * PAYMENT 
Report 
MWTP 
PAYMENT Mean N Std. Deviation
.00 1089.2857 14 595.23059
1.00 1921.7778 45 1141.60033
Total 1724.2373 59 1093.46996
ANOVA Table 
Sum of Squares df Mean Square F Sig.
Between 
Groups (Combined) 7400290.043 1 7400290.043 6.809 .012
Within Groups 61948950.635 57 1086823.695 
MWTP * 
PAYMENT 
Total 69349240.678 58
Measures of Association 
Eta Eta Squared
MWTP * PAYMENT .327 .107
Regression Analysis 
Variables Entered/Removed(b) 
Model Variables Entered Variables Removed Method 
1 PAYMENT, INCOME, DISTANCE(a) . Enter 
a All requested variables entered. 
b Dependent Variable: MWTP 
Model Summary 
Model R R Square Adjusted R Square Std. Error of the Estimate 
1 .362(a) .131 .084 1046.72478 
a Predictors: (Constant), PAYMENT, INCOME, DISTANCE 
ANOVA(b) 
Model Sum of Squares df Mean Square F Sig. 
Regression 9089438.847 3 3029812.949 2.765 .050(a) 
1 Residual 60259801.831 55 1095632.761
Total 69349240.678 58
a Predictors: (Constant), PAYMENT, INCOME, DISTANCE 
b Dependent Variable: MWTP 
E-5
Coefficients(a) 
Unstandardized Coefficients Standardized Coefficients 
Model 
B Std. Error 
Beta t Sig.
(Constant) 767.848 444.366 1.728 .090
DISTANCE -127.009 288.675 -.058 -.440 .662
INCOME 122.029 114.995 .138 1.061 .293
1 
PAYMENT 812.082 338.618 .319 2.398 .020
a Dependent Variable: MWTP