Evaluation database

Evaluation report

2000 CBD: UNICEF Follow-Up Survey of Households in CASD Villages (May-June 2000)

Author: Kenefick, E.

Executive summary


In 1998, UNICEF Cambodia identified a need to collect data from of the Community Action for Social Development (CASD) program for monitoring and evaluation purposes as a Baseline survey. The survey included a sample of villages in Battambang, Kampong Spueu, Kampong Thom, Kratie, Prey Veng, Stung Treang, Svay Rieng and Takeo provinces. UNICEF and the UN World Food Programme (WFP) jointly sponsored the Baseline survey.

The six main components of the CASD program for 1996-2000 are:
1. Capacity building
2. Community education and child care
3. Food, water and environment
4. Health, hygiene and caring practices
5. Protection and care of vulnerable children and women
6. Credit, employment and income

Purpose / Objective

The purpose of the evaluation is to assess programme impact by comparing changes in the status of children and mothers in UNICEF's and control districts. The objectives were to assess post-intervention changes in nutritional status from those households sampled as determined by z-scores, Body Mass Indices (BMI), and hemoglobin levels as well as with other measures of food security, vulnerability, and health status.


The Follow-up survey was to be conducted in the original 8 provinces with a separate sample of approximately150 households in the 2 districts of Otdar Mean Chey and would begin in May 2000.

The same village lists were used in 1998 and in 2000. The UNICEF list contained 581 villages of which approximately 50 were randomly chosen for inclusion in the survey. The other list was 2255 villages in the WFP Regular Target Communes (RTC) from their 1998 targeting exercise. Not every village in the WFP RTC list is or was necessarily the poorest in Cambodia since WFP's own first stage targeting is only at the commune level. The Baseline survey interviewed 1230 households from 124 villages: 62 from the WFP communes, 49 from CASD and 13 from Kratie and Stung Treang while the Follow-up sample of 2000 included 1298 households from 124 villages: 59 from the WFP communes, 52 from CASD and 13 from the northeast.

The second stage included random selection of households with children less than 5 years of age as the secondary sampling unit. The number of households selected depended on the size of the village, which was taken from the 1998 Census. If the village had less than 80 households, 8 families were interviewed. If there were between 80 and 120 households, then 10 families were included and any village with more than 120 households had 12 families interviewed for the survey.

Key Findings and Conclusions

Knowledge of the presence of Village Development Committees (VDCs) is very high among the CASD survey population. Attendance at VDC meetings is also quite high, with 70% of households in the CASD sample participating in the past month. However, the participation in activities (non-specific) is only 47% but perhaps some households are not aware that an activity is an "official" VDC function.

VDC participants also spend more on health care and are more likely to receive antenatal care from health centers, receive tetanus toxoid injections, and have children that are fully immunized. They are more likely to own soap, know the benefits of colostrum, and to feed their children more often and to feed them more nutritious foods. There are no direct differences in health and nutrition outcomes between VDC participants and non-participants.

Women participating in VDC functions have more pregnancies and children but also have higher survivorship of those children. They know more about fertility as well. Both mothers and children are more likely to have received vitamin A supplements in the past year. Households have higher diversity of income sources, and are more often engaged in rice farming. Hence, they spend less on food and borrow less often, except from Village Credit programs.

Access to a functioning primary school was no better in the CASD sample when compared to the Control sample, with 10% of households claiming the nearest primary school was outside the commune boundary. Furthermore, only 77% of school-aged children in the CASD households were attending school, which was lower than the 82% in the Control sample.

More than 60% of the households knew at least one importance of iodized salt. Most well-known health messages were the prevention of goiter and cleverness in children with messages being remembered most often from the television and radio. The knowledge is high but the utilization of iodized salt is very low.

The benefits of vitamin A are better known in CASD villages than the Control but not as high as iodine. Sources of this information are primarily television and health workers, with the latter presumably taking place during distribution of supplements. Expanding on this knowledge could easily be improved.

Recognition of the term "HIV/AIDS" is very high in the entire sample. Knowledge of how HIV/AIDS is transmitted is significantly higher in CASD but not yet known by the entire population. The same holds true for prevention, with greater knowledge in CASD than Control but not universally known. Only 50% of the sample had knowledge of non-sexual transmission of HIV/AIDS.

For the survey sample, very few households were without agricultural land - 6% in the Control and 9% in the CASD villages. Ownership of house plot, wet season rice land, and home gardens were the same for both CASD and Control, and had increased from Baseline for both groups. At Follow-up, ownership of chamkar land was significantly less in CASD households compared to Control, and had decreased from 1998.

Annual paddy production is significantly higher in the CASD households - 179 kg/capita as compared to 139 kgs/capita for the Control group. Consequently, significantly more CASD households (27%) sold part of their rice harvest for income. About 40% of producing households in both groups used a part of their harvest to pay debts. The median kilograms of rice left from sales and debt payment was 144 kgs/capita for CASD and 113 kgs/capita for the Control group,  a statistically significant difference.

The use of safe sources of drinking water was significantly higher in the CASD villages (58%) than the Control group (25%) and had increased from 45% at Baseline. Further investigation showed that approximately 60% of people who lived within 10 minutes of their main drinking water source were using safe water. That percentage dropped to around 40 for families more than 10 minutes from the source.

More women in the CASD sample were receiving some form of antenatal care in the Follow-up survey when compared to Baseline but not more than the Control group. The number of antenatal care visits was higher in the CASD sample at Follow-up. Significantly more CASD women had received at least one tetanus toxoid injection at Follow-up (59%) than at Baseline (50%) or than the Control group (51%).

The percentage of children (12-59 months) in CASD villages receiving at least one DPT injection or a BCG injection was significantly higher than the Control and had increased since Baseline. Although the immunization rate for measles was higher in 2000, it was still the lowest of all child vaccines, at 70% for CASD and 65% for Control. The immunization rate for polio had dropped for both groups from around 96% in 1998 to 90% in the Follow-up. The percentage of children 12-59 months of age who were fully immunized was higher at the Follow-up survey with 41% in the CASD villages and 33% for the Control group - a significant difference.

The 2-week period prevalence of fever was lower for both CASD (52%) and Control (58%) at Follow-up, in all age groups. The prevalence of acute respiratory infection was lower at Follow-up for CASD but was not different from the Control group. The prevalence of diarrhea decreased significantly for both groups and was significantly lower in the CASD sample (20%) when compared to the Control (25%).

The majority of cases of fever and ARI were treated with "other" medicine implying little knowledge of specific treatment regimens prescribed by either the pharmacist or the health center. At Follow-up there was a decrease in the use of drug sellers or pharmacists for treatment and slight increases in the use of private doctors and health clinics for both groups.
Data on treatment of diarrhea showed that significantly more CASD mothers had increased the breastfeeding of children during the episodes than in the Control group. Many cases were treated with medicine with little use of Oral Rehydration Solution (ORS) or home fluids for treatment. Most cases were treated by drug sellers or pharmacists, but fewer than at Baseline.

Both samples showed significant increases in soap ownership at Follow-up. The practice of washing hands with soap and water was higher overall in the CASD sample but still very low. Only 19% of CASD women reported washing their hands with soap and water after defecation.

The ownership of latrines was significantly improved at Follow-up for the CASD sample (14%) but still very low. Latrine ownership was significantly higher than the Control (6%) at Follow-up and higher than the national average for rural households (9%). For those households owning latrines, nearly all reported using them regularly for defecation for both adults and children.

Knowledge of the benefits of colostrum for babies was significantly higher in the CASD sample than the Control group but had not changed since Baseline. There were slight increases in the immediate initiation of breastfeeding after delivery but there was no difference between the CASD and Control samples. However, significantly more women initiated breastfeeding within 12 hours in CASD than Control. No women with very young children were practicing exclusive breastfeeding - nothing but breast milk since birth. For the CASD sample, most children were weaned at about 18 months of age, while weaning was earlier for the Control sample.

Women in the Control group had significantly more pregnancies and living children than the CASD group. However, the percentage of women who had experienced a miscarriage or child death was quite high for both groups - 29% in CASD and 34% in the Control group.

Significantly more non-pregnant women in CASD were using modern family planning methods (29%) at Follow-up than the Control (22%) or at Baseline (21%). The main methods preferred are the injection and birth control pills. At Follow-up, 19% of the women in each group knew when in their menstrual cycle they were most fertile. This knowledge was significantly higher for the Control group when compared to Baseline. For both groups and in both rounds, more than 80% of women interviewed wanted more information on family planning and birth spacing.

Overall, for children 6-59 months of age, the prevalence of wasting was higher in CASD (15%) than in Control (13%) but less than the Baseline. The prevalence of underweight was significantly reduced for both Control (50%) and CASD (49%) at Follow-up, as was the mean weight-for-age z-score. Stunting was significantly reduced for both groups at Follow-up but was also significantly lower in the CASD (38%) sample when compared to the Control group (48%).

The prevalence of low BMI (maternal malnutrition) was lower at Follow-up for both groups but with greater reductions in CASD (14%) than Control (17%) while the mean BMI was significantly higher. The prevalence of maternal stunting (< 1.45 m) was the same at Follow-up in both groups but the prevalence of maternal underweight (< 45 kgs) was significantly lower for CASD at Follow-up. Multiple risk factors in mothers (underweight, stunting, and anemia) are related to poor nutritional outcomes in their children, especially in underweight and stunting.

In 2000, the rate of nightblindness in pregnant mothers was significantly less for CASD (7%) than for the Control (11%) women, and a bit lower than at Baseline. Significantly more CASD mothers received vitamin A capsules after the delivery of their child (24%) than the Control mothers (17%). Vitamin A capsule distribution was significantly higher for all age groups in the CASD villages when compared to the Control and to the Baseline sample. Consumption of vitamin A rich foods was also significantly higher in both groups at Follow-up with families from CASD villages consuming eggs and kang kong significantly more often than the Control households.

There were reductions in child anemia from Baseline to Follow-up for all age groups - the overall prevalence of anemia in CASD was 72% or just slightly better than the 75% prevalence from the Control group. The prevalence of anemia in non-pregnant women was significantly lower than Baseline for CASD and was also lower (57%) than the Control (65%) at Follow-up. Anemia prevalence in pregnant women was also lower in CASD (69%) than the Control (83%) group.

Nearly 20% of households in both groups had no household assets in the Follow-up sample, showing them to be similar in socio-economic terms. Overall, significantly fewer households reported 3 income sources in the Follow-up survey, with the figure lower in the CASD sample (34%) than the Control (54%). Increased diversification of income sources was found to be related to housing type and relative wealth. Nearly 90% of the CASD households reported rice farming to be their primary source of income in 2000, as compared to 78% at Baseline. Animal raising accounted for 28% of income activities, followed by small business and wage labor for both groups. Women are very active in income generating activities, primarily in small business, animal raising and rice farming, with less involvement in fishing activities.

It appears that the rural economy had improved greatly from Baseline (1998) to Follow-up (2000) and had an impact on both groups because there was a much lower mean and median percentage of total expenditure on rice and other food, which is a proxy indicator of relative wealth. Less than 50% of the households reported expenditures for debt repayments. Gross median monthly expenditure was significantly less for the CASD families at Follow-up.

Household participation in Food-for-Works programs was much higher in the Control group (37%) than the CASD sample (19%). For participation in both the home village and other villages, the average amount of rice earned was 19 kg/capita in the Control group and 17 kg/capita in the CASD villages.

The rate of borrowing money from moneylenders or relatives was significantly less for both groups at Follow-up, with no difference between CASD and Control. However, significantly fewer families from CASD reported borrowing rice from non-family members (32%) than in the Control (38%) group but rice borrowing was less overall at Follow-up when compared to Baseline. Thirty-three percent of CASD families borrowed from village credit schemes, which was significantly higher than the 18% reported from Control families. Participation in village rice banks was also higher in CASD villages (14%) than in Control villages (7%). Households that borrow from more sources have fewer assets and higher percentage expenditure on rice.

Significantly more families had household members who migrated for 6 months or more to work at Follow-up - 20% in each group. In addition, the rate of migration for work was highest in households with no assets and decreased with increased asset ownership. At Follow-up significantly fewer families in CASD villages were taking risks by working in forested areas or in areas with land mines, when compared to the Control group.


Capacity Building:
Since a quantitative household survey alone cannot measure capacity building at all levels, further qualitative work should be conducted to measure the progress in achieving this objective.

The survey results show that families living in temporary houses on the ground are much less likely to participate in VDC meetings and activities. Targeting of vulnerable groups for inclusion in the VDC process is vital for complete community development. 

Community Education and Child Care:
Monitoring systems should be in place to ensure 100% enrolment of children in UNICEF-supported villages in functioning primary schools. Educational activities should include some form of assessment exercise to show that the participants have understood the message and can apply any new skills learnt to their own lives.

UNICEF should consider supporting school health programs that include initial growth monitoring and anemia testing of children followed by multi-vitamin supplementation, de-worming and school feeding. The local community in cooperation with the line ministries and WFP should support these programs. Progress can be monitored through repeated growth monitoring and blood hemoglobin assessment. A pilot program is currently underway in Kampot province through the support of GTZ.  There should be an intensive educational program to promote the benefits and use of iodized salt amongst all villagers.  

Food, Water and Environment:
Although only 9% of the households sampled had no agricultural land it is still important to identify the landless households in a CASD village and to target them for special training and programs in order to assist them to acquire land, raise animals, or to run small businesses. Some investigation into the significant decrease in chamkar farming should be investigated. Care should be taken not to have complete focus on rice farming as a means of income and a means of food production in order to mitigate the effects of flood or drought on a crop as well as to promote dietary diversity for the families. 

Health, Hygiene and Caring Practices:
Availability and access to functioning health clinics is low in the CASD villages. With the new Country Program and strengthening of the district level health centers, this should improve. Continued support and education is needed for women to understand how to ensure a safe pregnancy and delivery, especially targeting remote villages and the poorest women in the village. Health programs should emphasize prevention of illness first and foremost, including improvements in sanitation and personal hygiene. Secondly, campaigns to treat illnesses such as diarrhea and fever at home should be developed and promoted in the villages. Very few mothers reported using Oral Rehydration Solution to treat diarrhea but preferred to get medicine from the drug seller

There should be a re-evaluation of the program to promote boiling of children’s drinking water. The percentage of households doing so did not increase from Baseline and there was no relationship to boiling water and using drinking water from safe sources. The practice of boiling drinking water should be part of a complete routine of food and drink hygiene for children.

Based on the analysis of the breastfeeding, introduction of complimentary feeding foods and trends in malnutrition of rural Cambodia children, the most important component of a strategy to reduce child illness and malnutrition in this country is child feeding practices. There is lack of knowledge on appropriate breastfeeding, timely introduction of nutritious and appropriate complimentary weaning foods, and on the need to feed young children a diverse diet, more than three times a day. 

Protection and Care of Vulnerable Children:
The prevalence of child anemia is still very high. Very few of the children suffer from moderate or severe anemia, indicating that the problem is not due to malaria or helminth infections but rather to lack of iron in the diet and poor maternal iron stores. Supplementation of women is a first step to reducing the high prevalence of anemia in children. Secondly, in conjunction with the program on child feeding, specific steps should be outlined to improve dietary consumption of iron for young children. Experts at Micronutrient Initiative (MI) indicate that de-worming programs are best for reducing the severity of IDA but not for eliminating anemia in women or children. Reduction in child anemia should be a priority in the next Country Programme. 

Credit, Employment and Income:
UNICEF should consider developing a strategy of social safety nets to prevent or mitigate the effects of a natural disaster. Such a program is already a part of the UNDAF and the World Food Programme’s PRRO extension for 2001-2003.  Expansion of the village credit program could assist these poorer families by helping them to consolidate their debts as well as to help them have lower interest rates. Rice banks also help to provide a better safety net for the poorer families who rely too much on borrowing. 

Full report in PDF

PDF files require Acrobat Reader.



Report information





ECD - Area-based Programme

Royal Government of Cambodia, Ministry of Planning


Follow Up:


Sequence Number:

New enhanced search