3.1 Introduction
3.2.2 Data collection
Semi- structured and structured face-to-face interviews were administered to the 260 sample households to make possible the inclusion of participants who could not read nor write (Trochim, 2001). The questionnaire that was used to guide the interviews is presented in Appendix 3.2.The interviews were conducted in Swahili, the official language in Tanzania, at the participants own homes or venues of their choice in order to make easier the participation of women. In Rungwe district most of the farming activities including processing and crop storage are done by women {Food and Agriculture Organisation of the United Nations (FAO), 1994} just as it is in the whole of Tanzania (Government of the United Republic of Tanzania, 2006). Thus, participation of women in this study was considered crucial.
65
Seasonal calenders were used for identifying the time in a year when each sample household experienced maize shortage, and for studying farm household’s coping strategies in the indicated times of maize shortage. Among other participatory techniques, seasonal calenders drawn from the participants knowledge of the issue being studied enable involvement of the participants in planning and managing matters pertaining to the concerned study (Preece, 2006; King, 1994).
Data that wascollected in order to answer the specific objectives for the study included information regarding the amount of land that farm households allocated to growing maize, the amount of land that the farm households allocated to growing other types of foods, the time in a year when the farm households started consuming maize and the time in a year when the farm households ran out of maize. Information regarding the coping strategies that the farm households used when maize supplies ran out was also collected. The amounts of maize consumed by farm households and the amounts of maize they used for purposes other than consumption were recorded and used for answering specific objective two. The level of formal education, gender and age of farm household heads were also recorded for investigating their influence on maize consumption.
Data set concerning the type of maize that the farm households grew and the type of maize meal that the farm households often used was also collected. The number of days per week during which each farm household consumed the maize meal, the number of maize meals that each household consumed daily, the total number of meals that farm households consumed per day and quantities of maize flour that they used per meal were recorded. Data set concerning the reasons pertaining to the frequencies at which the farm households consumed maize meals was collected, farm households’ perceptions concerning the importance of maize and reasons for the perceptions, were also recorded.
3. 2.3 Statistical analyses
Data was analyzed using theStatistical Programme for Social Sciences (SPSS) version 15 by Pallant (2005), which required coding and capturing of the participants’ responses. Frequencies were calculated in order to find the number of scores per response. Pearson product-moment correlation was used to explore relationships between variables. In line with Pallant’s (2005) recommendation, preliminary analyses were executed prior to exploring relationships in order to
66
make sure that assumptions of normality, linearity and homoscedasticity are not violated. The preliminary analyses were done through scatter plots to check for outliers, and no outliers were found, hence the use of Pearson product-moment correlation. Parts of the outputs from data analysis are shown in Appendices 3.3 - 3.6.
The amount of nutrients that the farm households in Katumba ward could obtain from maize consumption per meal was calculated through converting the average amount of maize flour that farm households in Katumba ward utilized per meal into grams, followed by dividing the result by 100 g, then multiplying the obtained figure by FAO’s (1968) estimated figures for the quantity of each nutrient in 100g of maize flour as follows:
Amount of nutrients that the farm households can obtain per day from maize meals
Furthermore, the amount of nutrients required daily per individual differs with age, gender and activity (Food Nutrition Board, Institute of Medicine of the National Academies, 2005).
However, the quantities of nutrient intake recommended by the Food Nutrition Board, Institute of Medicine of the National Academies (2005) are also used as general estimates for the specified age groups of people who engaged in moderate activity. Thus, in this study, the estimated quantities of nutrients a farm household in Katumba ward could access from maize were calculated based mainly on the ages of the individuals in the farm households. The total amount of energy that an average farm household in Katumba ward required per day was calculated using the following equation:
Percentage of the daily required energy that the farm household obtains from maize=
Likewise, the percentage of protein that a farm household in Katumba ward can access from maize was calculated based on the amount of maize flour that a farm household consumes per
67
meal, and the Food and Nutrition Board, Institute of Medicine of the National Academies’ (2005) estimates for the amount of required protein per individual. Thus, the estimated percentage of protein that a farm household in Katumba ward could get from maize per day was calculated using the following equation:
Percentage of the daily required protein that the farm household obtains from maize =
Total estimated amount of maize that the farm households in Katumba ward produced per year was calculated by multiplying the average amount of maize that such a household produced per annum by the total number of farm households in the ward.