An Economic Assessment of Banana Genetic Improvement and ...
An Economic Assessment of Banana Genetic Improvement and ... An Economic Assessment of Banana Genetic Improvement and ...
178 APPENDIX D Table D.1 Sampling fractions for primary sampling units in the survey domain Population of primary sampling unit Elevation Sample of primary sampling units Elevation Stratum Low High Total Stratum Low High Total Exposed Row (percent) 49 (91) 5 (9) 54 (100) Exposed s.f. 17 0.367 3 0.40 20 0.37 Not exposed Row (percent) 155 (76) 49 (24) 204 (100) Not exposed s.f 15 0.097 5 0.10 20 0.098 Total 204 54 258 Total 32 8 40 Note: s.f. is the primary sampling unit sampling fraction (n ij /N ij ), where i =elevation (1 or 2) and j = exposure (1 or 0). mally within and among PSUs by choosing their number and the number of households per PSU to minimize the survey (sampling and nonsampling) error, given a fixed budget (De Groote 1996). In this case, the variances in the multiple population parameters of research interest were unknown. The minimum sample size for conducting hypothesis tests on variables measured at the community level (such as social capital and some physical capital) is 20 each in exposed and nonexposed areas (corresponding to a student’s t-test). Though a larger sample of communities would have been preferred for statistical precision, the cost of conducting the research in more than 40 communities scattered across the domain exceeded the budget. The total number of PSUs was therefore fixed at 40, with half distributed in the exposed areas and the other half in nonexposed. The 40 primary sampling units were then allocated between the two elevation levels and two countries proportionate to the probability of selection. Selection of Primary Sampling Units PSUs were drawn using systematic random sampling from a list frame with a random start. The sampling fractions for the primary sampling units among the four strata in the domain are shown in Table D.1. here> 1near
DETAILS OF SAMPLE SURVEY DESIGN 179 Table D.2 Number of survey sites (primary sampling units), by elevation and diffusion status, Uganda and Tanzania Stratum Low elevation High elevation Country Not exposed Exposed Not exposed Exposed Total Uganda 14 8 4 1 27 Tanzania 1 9 1 2 13 Total 15 17 5 3 40 Selection of Secondary Sampling Units The SSU both in Uganda and Tanzania is a village. In Uganda, in each LC3, there are several parishes (LC2s), and each parish consists of several villages (LC1s). One SSU was selected per PSU. The probability of selection (or sampling fraction) of a SSU varies by PSU and is equal to 1/M p , where M p represents the number of villages in the pth PSU (p = 1, . . . , 40 in the sample). For most exposed LC3s in Uganda, there is only one exposed LC1 per PSU. Where there is more than one exposed village per PSU, the SSU was drawn with a random number from the list of those villages with more than 100 households according to the 1991 census. Whether a community selected in the sample had been properly classified as exposed or nonexposed was then verified at the site. Two communities were replaced because they were fishing villages with little banana production. Selection of Households within a Community A constant sample size was maintained for two reasons. First, as mentioned above, there was no information on variation in underlying population parameters to guide optimal sample allocation. The second reason was to maintain comparable workloads among enumerators, who resided in the communities for 1 year to complete the survey. The sample villages were visited, and a current list of households in each village was requested from the chairman of the LC1. The total number of households selected per village was 20. The probability of selection (or a sampling fraction) of a household varies by village and is equal to 20/H s , where H s is the number of households in village s (s = 1, . . . , 40 in the sample). If there was an order in the list of households, random numbers were used for selection. Otherwise, a random start with systematic random sampling from the list was employed. Unit of Observation The basic unit of observation for the sample survey is the farm household. A farm household is defined according to the culture of which the household is a part, and includes female-headed and child-headed (orphaned) households, as well as male-headed households with more than one wife. The total sample size surveyed comprises 800 households (540 in Uganda and 260 in Tanzania). In some analyses conducted in this report, the plot or the banana cultivar is the unit of observation. Weights The overall probability of selecting a household in the subsample (denoted as PSH) is a unique number, defined as the product of the sampling fractions at each level: PSH =
- Page 144 and 145: 128 CHAPTER 8 Thomas, G. W. 1982. E
- Page 146 and 147: 130 CHAPTER 9 (Nkuba et al. 1999).
- Page 148 and 149: 132 CHAPTER 9 be reduced through la
- Page 150 and 151: 134 CHAPTER 9 Table 9.2 Summary sta
- Page 152 and 153: 136 CHAPTER 9 Table 9.3 Coefficient
- Page 154 and 155: 138 CHAPTER 9 Table 9.4 Mean compar
- Page 156 and 157: 140 CHAPTER 9 References Anandajaya
- Page 158 and 159: 142 CHAPTER 10 Table 10.1 Ugandan c
- Page 160 and 161: 144 CHAPTER 10 (Kangire and Rutherf
- Page 162 and 163: 146 CHAPTER 10 Table 10.4 Predomina
- Page 164 and 165: here> 7near 10. 148 CHAPTER 10 Tabl
- Page 166 and 167: 150 CHAPTER 10 Table 10.8 Present v
- Page 168 and 169: 152 CHAPTER 10 In terms of specific
- Page 171: Part IV. Conclusions
- Page 174 and 175: 158 CHAPTER 11 in cooking, brewing
- Page 176 and 177: 160 CHAPTER 11 sumption and income
- Page 178 and 179: 162 CHAPTER 11 4. 5. for which the
- Page 181 and 182: APPENDIX A Banana Taxonomy for Ugan
- Page 183 and 184: BANANA TAXONOMY FOR UGANDA 167 Tabl
- Page 185 and 186: BANANA TAXONOMY FOR UGANDA 169 Tabl
- Page 187 and 188: BANANA TAXONOMY FOR TANZANIA 171 Ta
- Page 189 and 190: BANANA TAXONOMY FOR TANZANIA 173 Ta
- Page 191 and 192: APPENDIX C Use of Improved Banana V
- Page 193: DETAILS OF SAMPLE SURVEY DESIGN 177
- Page 197 and 198: DETAILS OF SAMPLE SURVEY DESIGN 181
- Page 199 and 200: VILLAGE SOCIAL STRUCTURE IN UGANDA
- Page 201 and 202: List of Principal Authors and Contr
- Page 203 and 204: ABOUT THE AUTHORS 187 rigation, and
DETAILS OF SAMPLE SURVEY DESIGN 179<br />
Table D.2 Number <strong>of</strong> survey sites (primary sampling units), by elevation <strong>and</strong> diffusion<br />
status, Ug<strong>and</strong>a <strong>and</strong> Tanzania<br />
Stratum<br />
Low elevation<br />
High elevation<br />
Country<br />
Not exposed Exposed Not exposed Exposed<br />
Total<br />
Ug<strong>and</strong>a 14 8 4 1 27<br />
Tanzania 1 9 1 2 13<br />
Total 15 17 5 3 40<br />
Selection <strong>of</strong> Secondary<br />
Sampling Units<br />
The SSU both in Ug<strong>and</strong>a <strong>and</strong> Tanzania is a<br />
village. In Ug<strong>and</strong>a, in each LC3, there are<br />
several parishes (LC2s), <strong>and</strong> each parish<br />
consists <strong>of</strong> several villages (LC1s). One<br />
SSU was selected per PSU. The probability<br />
<strong>of</strong> selection (or sampling fraction) <strong>of</strong> a SSU<br />
varies by PSU <strong>and</strong> is equal to 1/M p , where<br />
M p represents the number <strong>of</strong> villages in the<br />
pth PSU (p = 1, . . . , 40 in the sample). For<br />
most exposed LC3s in Ug<strong>and</strong>a, there is only<br />
one exposed LC1 per PSU. Where there is<br />
more than one exposed village per PSU, the<br />
SSU was drawn with a r<strong>and</strong>om number<br />
from the list <strong>of</strong> those villages with more<br />
than 100 households according to the 1991<br />
census. Whether a community selected in<br />
the sample had been properly classified as<br />
exposed or nonexposed was then verified at<br />
the site. Two communities were replaced<br />
because they were fishing villages with little<br />
banana production.<br />
Selection <strong>of</strong> Households<br />
within a Community<br />
A constant sample size was maintained for<br />
two reasons. First, as mentioned above, there<br />
was no information on variation in underlying<br />
population parameters to guide optimal<br />
sample allocation. The second reason was to<br />
maintain comparable workloads among<br />
enumerators, who resided in the communities<br />
for 1 year to complete the survey. The<br />
sample villages were visited, <strong>and</strong> a current<br />
list <strong>of</strong> households in each village was requested<br />
from the chairman <strong>of</strong> the LC1. The<br />
total number <strong>of</strong> households selected per village<br />
was 20. The probability <strong>of</strong> selection (or<br />
a sampling fraction) <strong>of</strong> a household varies<br />
by village <strong>and</strong> is equal to 20/H s , where H s<br />
is the number <strong>of</strong> households in village s<br />
(s = 1, . . . , 40 in the sample). If there was<br />
an order in the list <strong>of</strong> households, r<strong>and</strong>om<br />
numbers were used for selection. Otherwise,<br />
a r<strong>and</strong>om start with systematic r<strong>and</strong>om<br />
sampling from the list was employed.<br />
Unit <strong>of</strong> Observation<br />
The basic unit <strong>of</strong> observation for the sample<br />
survey is the farm household. A farm household<br />
is defined according to the culture <strong>of</strong><br />
which the household is a part, <strong>and</strong> includes<br />
female-headed <strong>and</strong> child-headed (orphaned)<br />
households, as well as male-headed households<br />
with more than one wife. The total<br />
sample size surveyed comprises 800 households<br />
(540 in Ug<strong>and</strong>a <strong>and</strong> 260 in Tanzania).<br />
In some analyses conducted in this report,<br />
the plot or the banana cultivar is the unit <strong>of</strong><br />
observation.<br />
Weights<br />
The overall probability <strong>of</strong> selecting a household<br />
in the subsample (denoted as PSH) is a<br />
unique number, defined as the product <strong>of</strong><br />
the sampling fractions at each level: PSH =