Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Organization and logistics of the data collection

Personnel

Six working teams conducted the tests, examinations, measurements, and interviews (table 1).

TABLE 1. Field personnel

Two teams, of eight field workers each, collected data on the follow-up cohort, operating from a central location. Two were anthropometrists who, in addition to the anthropometric measurements, performed the hand and wrist radiographs and the hand-strength test. A physician conducted the medical examination and collected blood and urine samples. Three psychometrists were in charge of administering the behavioural tests: two for functional performance and intelligence and one for information processing. One enumerator conducted the life history interviews for both men and women, including the wives of adolescent men.

A logistic supervisor was in charge of coordinating the flow of subjects through the different tests, examinations, and measurements and was also responsible for identifying the subjects who would participate each day. He was assisted by two community workers in each village, who were in charge of visiting the subjects selected for study one day in advance and inviting them to attend the centre where the teams were located. In addition, one master's degree student from the University of California at Davis collected the school performance data from school records, with support from the behavioural component team.

Three enumerators conducted the life history interviews with the mothers of the follow-up cohort. They received help from the two enumerators who conducted the life history interviews with the follow-up cohort. Most of the data were collected through household visits. Two enumerators gathered income and wealth data on the heads of households through home visits. Two anthropometrists did the anthropometric measurements on the parents, using a combination of home visits and measurements at the central location.

Two anthropometrists made the anthropometric measurements on preschool children, generally at the central location but occasionally during home visits.

The work-capacity team consisted of a physician and a nutritionist, both with training in exercise physiology, a medical student, and a nurse. In addition, two master's degree students from Cornell University joined the team for a few months. The team operated in a physiology laboratory, which, with one exception (Espíritu Santo), was not located in the study villages because of a lack of appropriate facilities; the subjects were transported to the laboratory where the tests were performed.

A special team, consisting of a supervisor and two field workers, was put together to locate migrants and organize the collection of data from them. Only subjects who were known to have migrated to Guatemala City, Sanarate, or El Jícaro were included in the follow-up study.

Coordination and supervision

A technical coordinator (J. Rivera) was responsible for the overall coordination of the project. A field director (H. Castro) was in charge of the day-to-day coordination of the field work and also supervised the anthropometry data collection, the hand and wrist radiographs, and the medical examinations. The coordinator and supervisor of data collection for the behavioural component was Kathleen Gorman. Beth Conlisk and Elkin Martinez were in charge of the work capacity test, coordination of the bio-impedance and bone density examinations, and the physical activity questionnaire. In addition, three field work supervisors with extensive experience at INCAP were hired to supervise the life history interviews (Marta Amanda Barrera), income and wealth data collection (Victor Mejía-Pivaral), and the study of migrants (Elena de Ramírez).

Logistics

Rotation of teams

The data-collection teams rotated among the villages, following the programme outlined in table 2. Data were collected simultaneously in two villages. The second and third columns give the locations of the two follow-up cohort teams during each week of the study. The first column under "Work capacity" indicates the location of the laboratory and the second the villages of origin of the participating subjects. The last three columns give the locations of the adults teams each week.

TABLE 2. Data collection schedule, May 1988-May 1989

Atole villages: Conacaste, San Juan. Fresco: Santo Domingo, Espíritu Santo. Comparison: Subinal, Las Ovejas, El Caulote. Data collection in Guatemala City and Sanarate (except in the case of the work-capacity laboratory) was for migrants from the villages.

The purpose of rotating the teams was to diminish biases due to measurement differences between teams; therefore each team collected data in all the villages. Also, the data collection in each village was staggered over the duration of the study to coincide with both the rainy and the dry season.

The work-capacity laboratory was moved according to the rotation programme. It needed to be where there was adequate space, electricity was available, and air-conditioning could be installed. The closest locations where these conditions could be met were selected: the town of Sanarate for four of the villages in the western part of the province, and Espíritu Santo, one of the supplemented villages, for itself and one of the comparison villages. The laboratory was moved between these two places twice during the study. At the end it was moved to Guatemala City, where children from El Caulote, another comparison village, performed the tests; this was the only village for which the work-capacity data were collected in one stretch of time.

Programme of appointments

Lists of the target samples for participation in the various tests, measurements, and interviews were generated from the INCAP-Cornell master file. The families of the follow-up cohort members in the target sample were selected randomly for participation during the first or the second visit of the working teams to their village. The supervisors instructed the two community workers in each village to visit the individuals listed and invite them to be study subjects. The follow-up subjects were scheduled to complete all tests, measurements, and interviews in three visits. In practice, however, most chose to complete the entire battery in two days and in some cases in one day.

On arrival at the centre, the subjects received an explanation about the assessments in which they would participate. The person giving this explanation coordinated the flow of subjects into the different study areas. When possible, the behavioural tests were performed before the medical examination, which involved collection of a blood sample, since this caused anxiety in some of the subjects. When possible, the behavioural tests were spread across visits to avoid loss of attention due to fatigue.

The typical durations of the procedures conducted with the follow-up cohort were as follows: anthropometric measurements, hand-wrist radiographs, and hand-strength tests, 25 minutes; medical examination and collection of urine and blood samples, 25 minutes; functional competence and intelligence tests, 60 minutes; and life history interview, 20 minutes. In general, the subjects tolerated the time involved better than had been expected.

A few follow-up subjects who refused to attend the centre but were willing to participate (3.6%) were visited at their homes, where the tests and measurements—except for the radiographs and the information-processing test, both of which required the use of equipment that could not be transported— and the interviews were carried out. Most of the income and wealth interviews of heads of households and many of the life history interviews of mothers and the anthropometric measurements of the parents of members of the follow-up cohort were carried out at their home, since these subjects had less time to attend the centres.

The teams worked approximately eight hours a day on flexible schedules depending on the preferred hours of participation of the subjects in each village and season. In some villages and seasons, work was from 8 a.m. to 5 p.m., and in others it was as late as between noon and 9 p.m. In all the villages, the teams had to work some weekends in order to allow for the participation of subjects who worked in the field until late afternoon on weekdays or who worked out of the village during the week and returned on weekends and migrants who visited their families on weekends.

Training, supervision, and data flow

Training

Training took place in February and March 1988. Project staff and, in some cases, experts hired as consultants were in charge of training the field personnel in the different areas. A detailed description of those who were in charge of training is available elsewhere [2].

The final stage of training included standardization exercises in anthropometry and in administering the functional performance tests, the income and wealth questionnaire, and the life history questionnaire. In the anthropometry training exercise, the levels of technical errors of measurement obtained were as good as or better than those reported in the literature. For the life history and the income and wealth questionnaire as well as for the functional performance tests, the percentages of agreement between interviewers were in general above 95% [2].

Supervision and quality control

Supervision of the anthropometric measurements, radiographs, hand-strength tests, medical examinations, life history interviews, and behavioural tests was continuous. The supervisors spent at least two or three days per week in the field, providing direct supervision to the field workers and examining and correcting the data-collection forms weekly. Efforts were made to identify obvious coding errors, consulting with the field workers who had collected the data, and correcting them when possible.

When errors were suspected in anthropometric measurements, the anthropometrist responsible for data collection was asked to repeat the measurement. The supervisor then compared the measurements and specified which value to accept. This supervision involved direct observation of the technique and review of the data-collection forms. Ranges of permissible values were used to detect outliers; then, either obvious errors were corrected or the subject was re-examined. The tests in the behavioural area and the life history interviews required more privacy; therefore, supervision for those areas was done only through review of the data forms.

The supervisor of the income and wealth data collection visited the field less frequently because of other commitments. In this area, the two enumerators exchanged data forms after interviews were conducted. Apparent discrepancies or coding errors were discussed.

In the work-capacity area, the supervisors participated directly in the data collection. After each test, the team reviewed the results and so detected and corrected obvious errors.

Repeated measurements were made in a subsample of the subjects in order to establish reliability. The percentages of subjects remeasured were as follows: 10.3% for anthropometry, 3.9%-4.6% for the behavioural tests, 4.3% for work capacity, and 2%5% for the life histories and income and wealth questionnaires.

Detailed analysis of the reliability of the anthropometric measurements is presented elsewhere [3]. The reliability values were above .91 for all the anthropometric measurements. The intra-measurer reliabilities were above .96 in all measurements, and the inter-measurer reliabilities were above .98 in most measurements. These values are considered satisfactory.

Test-retest stability coefficients and internal consistency measures of the psycho-educational tests were in general similar to published results and are considered satisfactory. Some variables of the information-processing tests with low stability coefficients were dropped from the analysis. Differences between testers were large in many cases; however, analysis before and after controlling for tester effects showed similar results. Tn summary, reliability was found to be acceptable for key psycho-educational variables.

Data flow, entry, verification, and cleaning

After thorough revision and correction of the data forms, the data were key-punched twice at the INCAP computer centre. They were then cleaned using valid ranges of values to detect outliers suspected to be errors and also using consistency checks across variables. Finally, computer files were prepared for data analysis. Values suspected to be incorrect were sent back to the field, where the supervisor of each area corrected coding errors. In the areas of anthropometry and life history, when errors other than coding were found, the subjects were reexamined.

Primary health care activities

The study imposed time demands on and caused some inconvenience for the subjects. Since the principal objective was to study the long-term effects of undernutrition and supplementary feeding, the study was of little direct benefit to the people of the communities in improving their nutrition status and health. Consequently, it was considered necessary to contribute directly to the welfare of the villages, and a primary health care programme was established for this purpose. To avoid its dependence on the project, it was implemented in coordination with local clinics run by the Ministry of Health in five of the seven villages.

A paediatrician with a master's degree in public health and nutrition was in charge of the programme. The emphasis was on maternal and child health care, but it provided medical attention to persons of all ages in all the study villages. The project donated a supply of drugs that were needed in the clinics run by the Ministry of Health, and the paediatrician provided training to the nurses in the ministry's child survival programme. In the two villages without clinics, the project established clinics in buildings donated by the communities and hired a nurse to staff them. The paediatrician visited each village once a week and examined and gave medical attention to children identified by the nurses as requiring medical care as well as all those who came to the clinic; he also used these cases as opportunities to train the nurses.

In addition, campaigns were organized in collaboration with dentists from the University of San Carlos for the provision of dental care.

At the end of the project all the medical equipment was donated to the communities where the new health centres had been established.

Contents - Previous - Next