(Awumbila et al. 2015; Songsore 2009). Similarly, women in the study area tend to
have limited control and ownership of productive resources such as land and capital
(Rademacher-Schulz and Mahama 2012). Women also participate less in decisionmaking, although they contribute significantly to labour and housekeeping. In fact,
in the study area, men have the major say in household decision-making and
practically control most productive resources and assets (Zakaria et al. 2015; Zakaria
2017). Participation in the FGDs varied from 6 to 12 people (Table 3). Even though
we sought to obtain a representative sample of each target group in each community,
the actual participation was ultimately influenced by the availability and willingness
of invitees to participate.
During the FGDs, we aimed to elicit perceptions about the performance of each
age group across the different resilience elements. To elicit a qualitative estimate, we
used a Likert-type scale where 1 ¼ very low performance, 2 ¼ low performance,
3 ¼ moderate performance, 4 ¼ high performance and 5 ¼ very high performance.
We clarified to all participants that their answer should represent their group’s
performance in that specific resilience element, with lower scores denoting poor
performance in a specific element, whilst higher scores denoting a better performance. Research assistants, who were native to the study areas and proficient in the
local dialects, translated the resilience elements to the local dialects. Prior to each
FGD, the scoring rules and approach were explained to all participants using five
rounded, equally sized stones.
The FGDs generated dialogue among participants and sought to capture consensus opinions. Thus, each of the value elicited during Step 4 characterizes the entire
group, and not the individual respondent. When the group reached an agreement for
a given resilience element, the moderator moved on to the next element. When
participants failed in reaching a consensus immediately, the moderator brought up a
shared experience of an impact associated with a flood or drought event, as a means
of resolving the lack of consensus over the score. For most resilience elements,
participants were asked to provide tangible and practically relevant evidence to
support their proposed performance scores.
6.2.4 Data Analysis
The FGD results are aggregated at three levels: (a) social group (i.e. elderly men,
elderly women, young adults), (b) location (i.e. disaster proneness, community) and
(c) resilience category (i.e. dimension, sub-dimension, element). We calculate the
mean resilience scores across communities and interest groups and use the outcomes
from each FGD to estimate the perceived resilience score.
We obtain the scores for each sub-dimension through averaging the scores for
each resilience element. Finally, the resilience scores for each resilience dimension
(i.e. ecological, engineering, socio-economic) are calculated as the average of each
sub-dimension. The average of averages is used to obtain the resilience score for
each resilience dimension and sub-dimension to provide the same weight for each
6 Perceived Community Resilience to Floods and Droughts Induced by Climate Change. . . 203
have limited control and ownership of productive resources such as land and capital
(Rademacher-Schulz and Mahama 2012). Women also participate less in decisionmaking, although they contribute significantly to labour and housekeeping. In fact,
in the study area, men have the major say in household decision-making and
practically control most productive resources and assets (Zakaria et al. 2015; Zakaria
2017). Participation in the FGDs varied from 6 to 12 people (Table 3). Even though
we sought to obtain a representative sample of each target group in each community,
the actual participation was ultimately influenced by the availability and willingness
of invitees to participate.
During the FGDs, we aimed to elicit perceptions about the performance of each
age group across the different resilience elements. To elicit a qualitative estimate, we
used a Likert-type scale where 1 ¼ very low performance, 2 ¼ low performance,
3 ¼ moderate performance, 4 ¼ high performance and 5 ¼ very high performance.
We clarified to all participants that their answer should represent their group’s
performance in that specific resilience element, with lower scores denoting poor
performance in a specific element, whilst higher scores denoting a better performance. Research assistants, who were native to the study areas and proficient in the
local dialects, translated the resilience elements to the local dialects. Prior to each
FGD, the scoring rules and approach were explained to all participants using five
rounded, equally sized stones.
The FGDs generated dialogue among participants and sought to capture consensus opinions. Thus, each of the value elicited during Step 4 characterizes the entire
group, and not the individual respondent. When the group reached an agreement for
a given resilience element, the moderator moved on to the next element. When
participants failed in reaching a consensus immediately, the moderator brought up a
shared experience of an impact associated with a flood or drought event, as a means
of resolving the lack of consensus over the score. For most resilience elements,
participants were asked to provide tangible and practically relevant evidence to
support their proposed performance scores.
6.2.4 Data Analysis
The FGD results are aggregated at three levels: (a) social group (i.e. elderly men,
elderly women, young adults), (b) location (i.e. disaster proneness, community) and
(c) resilience category (i.e. dimension, sub-dimension, element). We calculate the
mean resilience scores across communities and interest groups and use the outcomes
from each FGD to estimate the perceived resilience score.
We obtain the scores for each sub-dimension through averaging the scores for
each resilience element. Finally, the resilience scores for each resilience dimension
(i.e. ecological, engineering, socio-economic) are calculated as the average of each
sub-dimension. The average of averages is used to obtain the resilience score for
each resilience dimension and sub-dimension to provide the same weight for each
6 Perceived Community Resilience to Floods and Droughts Induced by Climate Change. . . 203