289
15 The Damage Schedule Approach
direct attachment to conservation, while ‘programme [sic] to promote scienti fi c
research on lake fi sheries ecosystem’, which draws on the well-known precautionary
principle, also deems conservation highly relevant. Valuing these over the other
community programs, which may largely be driven by other interests and motivations that show little compatibility with the conservation objective, can be interpreted as a fair indication of their genuine inclination towards conservation. Each of
the community programs has an associated value, which is suggested in brackets in
Table 15.1 . The programs are presented to the respondents in a hypothetical sense
as something that could be implemented, but without the promise of implementation. This was to prevent immediate expectation from in fl uencing their choices. An
example of the paired comparison used in Set B is displayed in Fig. 15.1b .
Survey Information
The survey was directed at seven groups of respondents and involved multiple sites, as
shown in Fig. 15.2 . Active fi shing villages on the eastern shore of the SEA were chosen
to be the main location for surveying the resource-dependent groups, that is, gear owners, crew members, fi sh processors/traders and community members. This side of the
water body is believed to yield better catch than the stock-depleted western shore
(Njaya 2008 ) . Also, due to its relative remoteness, coupled with the shortage of infrastructure and tourism development, fi shing still remains a key economic activity supporting people’s livelihoods on the eastern shore. To investigate any potential disparity
that may arise from the east-west geographical distinction, two more resource-dependent groups, gear owners and crew members on the western shore, were added to the
survey. Thus, together with the managers/scientists group, seven respondent groups
were formed. The survey was conducted with the assistance of a local person, who is
native to the area and, in addition to being pro fi cient in English, has fl uency in two of
the most widely spoken local languages, Chichewa and Chiyao. The survey with the
scientists/managers group, comprising of various government of fi cials such as planners, researchers, statisticians, enforcement of fi cers, lecturers, and also scientists from
several nongovernmental organizations (NGOs) working in the fi eld of fi sheries, took
place in various locations around the lake. These were conducted without the assistance
of the local translator, because of their pro fi ciency in the English language.
All 21 pairs from each set were included in the survey booklet, resulting in a total of
42 pairs. The sequential order of the pairs in the booklet and the left-right position of
the two choices in each pair were both randomly generated to ensure the uniqueness
of each booklet, as well as avoiding any possible order-related bias. Due to concerns
about the availability and reliability of a census database, random sampling of respondents was not feasible. Instead, quota sampling was employed to obtain the total
number of respondents, 144. Approximately 20 respondents were surveyed for each
group. As shown in the demographic breakdown of the survey respondents (Table 15.2 ),
only modest differences exist in the number of respondents, their average age and the
average years of fi shery experience. One may think that the small sample size of each
15 The Damage Schedule Approach
direct attachment to conservation, while ‘programme [sic] to promote scienti fi c
research on lake fi sheries ecosystem’, which draws on the well-known precautionary
principle, also deems conservation highly relevant. Valuing these over the other
community programs, which may largely be driven by other interests and motivations that show little compatibility with the conservation objective, can be interpreted as a fair indication of their genuine inclination towards conservation. Each of
the community programs has an associated value, which is suggested in brackets in
Table 15.1 . The programs are presented to the respondents in a hypothetical sense
as something that could be implemented, but without the promise of implementation. This was to prevent immediate expectation from in fl uencing their choices. An
example of the paired comparison used in Set B is displayed in Fig. 15.1b .
Survey Information
The survey was directed at seven groups of respondents and involved multiple sites, as
shown in Fig. 15.2 . Active fi shing villages on the eastern shore of the SEA were chosen
to be the main location for surveying the resource-dependent groups, that is, gear owners, crew members, fi sh processors/traders and community members. This side of the
water body is believed to yield better catch than the stock-depleted western shore
(Njaya 2008 ) . Also, due to its relative remoteness, coupled with the shortage of infrastructure and tourism development, fi shing still remains a key economic activity supporting people’s livelihoods on the eastern shore. To investigate any potential disparity
that may arise from the east-west geographical distinction, two more resource-dependent groups, gear owners and crew members on the western shore, were added to the
survey. Thus, together with the managers/scientists group, seven respondent groups
were formed. The survey was conducted with the assistance of a local person, who is
native to the area and, in addition to being pro fi cient in English, has fl uency in two of
the most widely spoken local languages, Chichewa and Chiyao. The survey with the
scientists/managers group, comprising of various government of fi cials such as planners, researchers, statisticians, enforcement of fi cers, lecturers, and also scientists from
several nongovernmental organizations (NGOs) working in the fi eld of fi sheries, took
place in various locations around the lake. These were conducted without the assistance
of the local translator, because of their pro fi ciency in the English language.
All 21 pairs from each set were included in the survey booklet, resulting in a total of
42 pairs. The sequential order of the pairs in the booklet and the left-right position of
the two choices in each pair were both randomly generated to ensure the uniqueness
of each booklet, as well as avoiding any possible order-related bias. Due to concerns
about the availability and reliability of a census database, random sampling of respondents was not feasible. Instead, quota sampling was employed to obtain the total
number of respondents, 144. Approximately 20 respondents were surveyed for each
group. As shown in the demographic breakdown of the survey respondents (Table 15.2 ),
only modest differences exist in the number of respondents, their average age and the
average years of fi shery experience. One may think that the small sample size of each