 |
| This website is best viewed in IExplorer 4.0 or higher.
Land, People and Environment The Problem The Research Geo-Cultural Visual Tour About Me and My Reserach Interests
This website is best viewed in IExplorer 4.0 or higher.
|
Website
Design and Construction by Carlalynne C. Melendez |
|
DATA
ANALYSIS
|
||||||||||||||||||||||||||||||
|
The data analysis is guided by the notion that no single analytical approach is best when evaluating the relationship between culture, environment, and health. Accordingly, the research incorporates several analytical approaches that contribute to a broader understanding of the factors contributing to adverse health events, in particular, those associated with pesticide exposure. The term "exposure", which is used throughout this dissertation, refers to the evidence supporting pesticide intoxication or poisoning (accidental or intentional) and its associated symptoms and adverse health effects. Due to the meager public health and epidemiological information available on pesticide exposure in the study region, the research relied on multiple health indicators such as health survey-questionnaire, proximity analysis, cholinesterase test results, water testing, and pesticide dispersal or drift analysis.
SURVEYS In June 1998, I conducted a Reconnaissance / Exploratory Survey of the region to establish an understanding of local environmental conditions, community health problems and cultural characteristics. The Rapid Rural Appraisal approach (McCracken, Pretty, and Conway 1988; Ellman 1981; Beebe 1985) used in the study generated basic information on the feasibility of beginning the survey in the Constanza region. In 1998-1999, an Environmental Survey of community members’ attitudes, perception and knowledge regarding environmental change was carried out. The survey took place in nine communities: Tireo al Medio-La Altagracia, Suriel, Tireo Abajo, Tireo, La Culata, Valle Nuevo, Colonia Kennedy, Palero, and La Sabina. An environmental questionnaire-interview was also administered in the nine communities. The survey’s objective was to gauge local perception and knowledge of environmental change, and how that change has affected the health and well being of community members. The survey population included parcela owners, farm workers and their families, agrochemical mixers and applicators and their families. These were considered a more accurate representation of the population and the landuse activities of the communities under investigation. Semi-structured interviews were conducted simultaneously to identify and define the issues and problems affecting health in the communities. I attended community meetings to hear community members voice their opinions and concerns regarding a multitude of topics. At these meetings, community leaders and other individuals actively involved in the community, who appeared qualified to serve as key informants, were identified. I assisted and participated in various chats sponsored by agrochemical companies that targeted landowners and agricultural workers. At these community chats, I presented the objectives of my research and interviewed landowners, farm workers and other community members that could potentially contribute to the overall research objectives.
In the 1998-1999 survey, a randomly selected sample (N=262) from an approximate population size of 550 were asked general questions on environmental changes they had observed in their communities. The sample was selected based on the following eligibility criteria: participant lived in the community for 20 years or more, participant’s house is located close to pesticide sprayed fields; participant had general knowledge of his/her natural environment. Basic information was generated about local knowledge and perception regarding the causes of environmental change in their communities Questions were asked about fluctuations in the number of insects, temperature and precipitation, loss of biodiversity and prevalent diseases in the region. Table 1 shows the composition of the population samples for all surveys conducted between 1998 and 2002.
Statistical Tests Several statistical tests were used to estimate the strength of association between cholinesterase levels, symptom severity and probable level of pesticide exposure. I used linear regression for finding out whether or not there was a significant relationship between symptom severity / cholinesterase test levels and proximity to pesticide source (distance). Based on health survey questionnaire and field observations, two questions were examined: (1) what is the likelihood that a self-reported symptom is caused by pesticides? and (2) what is the likelihood that a low cholinesterase level (<50% of baseline level) is due to pesticide exposure?2.
Relational Database It is important to note that households and not individuals were surveyed. The person who was interviewed or who was administered the questionnaire, was not always the same person who suffered from symptoms, provided cholinesterase test results, or was in contact with agrochemicals. To get a better understanding of what was the health status of the household, I would asked other members of the households questions that were related to the survey questionnaire. These secondary data sources, along with my observational data, were entered into a relational database. The database was created to store health, cultural, biogeophysical and environmental data. Tables are the fundamental element of the database, and all database operations were performed through actions on tables. Essentially, tables served as storage containers, which facilitated querying and manipulating the different data sets. The data were stored on three 250mb zip disks containing interrelated multiple files and tables, which facilitated data analysis. The data analysis is divided into three sections:
|