165
14.2 Materials and Methods
A literature review resulted in an early list of suitable farm-level indicators for each
of the three CSA pillars. The list was then discussed with extension officers of
Lushoto District and experts from CIAT and the Selian Agricultural Research
Institute before a final list of indicators was agreed upon.
For the food security and adaptation pillars, the indicators were weighted, scored
and finally combined into aggregated indices using a weighted sum of the indicators. Weights and scores were elicited via a survey carried out among a selection of
CCAFS project farmers. The data collection protocol involved pairwise ranking and
scoring, according to a Likert scale. The weights for each indicator were established
through pairwise comparison following the Analytic Hierarchy Process outlined in
Saaty (1980). Comparisons of the importance of the indicators were entered into a
matrix with a 1–9-point scale. Following this, a consistency ratio was calculated for
each pillar. When the consistency ratio was greater than 0.10, all comparisons were
reviewed and the inconsistent ones re-evaluated (Saaty 1980). The weight of each
indicator from each pillar was calculated using a normalised comparison matrix in
which each value present in the matrix was divided by the sum of its column.
Based on these weights, the aggregated food security and adaptation indices
were calculated following a three-step process. Firstly, the intervention was scored
for each indicator within the food security and adaptation pillars. Farmers were
asked to assess whether there had been an increase in the indicators since the beginning of the intervention. The scoring of the indicators was performed using a Likert
scale, with scores ranging from 0 to 5—a score of 1 meaning that the farmer strongly
disagrees, and 5 meaning that the farmer strongly agrees that the indicator has
increased since s/he began the intervention. Secondly, these scores were translated
into values ranging between −1 and +1, where: 0 means ‘no contribution’; −1
means ‘reduces overall score strongly’; −0.5 means ‘reduces overall score’; 0.5
means ‘increases overall score’; and 1 means ‘increases overall score strongly’. The
final step resulted in a weighted sum per CSA pillar.
Through this process we achieved a farmer-centric evaluation of the interventions. The establishment of indicator weights, based on the farmer’s perspective,
ensures that the assessment takes into account the indicators that are most relevant
to the farmer in his/her own context. The scoring of the indicators is based on the
changes the farmers observe on their own farms as a result of the improved practice
and allows for a comparison with the farmers’ previous or ‘business-as-usual’
practice.
The protocol was tested among 72 farmers in the climate-smart village of
Lushoto. The data was analysed using Microsoft Excel and different CSA interventions assessed for their contribution to adaptation and productivity.
14 A Participatory Approach to Assessing the Climate-Smartness of Agricultural…
14.2 Materials and Methods
A literature review resulted in an early list of suitable farm-level indicators for each
of the three CSA pillars. The list was then discussed with extension officers of
Lushoto District and experts from CIAT and the Selian Agricultural Research
Institute before a final list of indicators was agreed upon.
For the food security and adaptation pillars, the indicators were weighted, scored
and finally combined into aggregated indices using a weighted sum of the indicators. Weights and scores were elicited via a survey carried out among a selection of
CCAFS project farmers. The data collection protocol involved pairwise ranking and
scoring, according to a Likert scale. The weights for each indicator were established
through pairwise comparison following the Analytic Hierarchy Process outlined in
Saaty (1980). Comparisons of the importance of the indicators were entered into a
matrix with a 1–9-point scale. Following this, a consistency ratio was calculated for
each pillar. When the consistency ratio was greater than 0.10, all comparisons were
reviewed and the inconsistent ones re-evaluated (Saaty 1980). The weight of each
indicator from each pillar was calculated using a normalised comparison matrix in
which each value present in the matrix was divided by the sum of its column.
Based on these weights, the aggregated food security and adaptation indices
were calculated following a three-step process. Firstly, the intervention was scored
for each indicator within the food security and adaptation pillars. Farmers were
asked to assess whether there had been an increase in the indicators since the beginning of the intervention. The scoring of the indicators was performed using a Likert
scale, with scores ranging from 0 to 5—a score of 1 meaning that the farmer strongly
disagrees, and 5 meaning that the farmer strongly agrees that the indicator has
increased since s/he began the intervention. Secondly, these scores were translated
into values ranging between −1 and +1, where: 0 means ‘no contribution’; −1
means ‘reduces overall score strongly’; −0.5 means ‘reduces overall score’; 0.5
means ‘increases overall score’; and 1 means ‘increases overall score strongly’. The
final step resulted in a weighted sum per CSA pillar.
Through this process we achieved a farmer-centric evaluation of the interventions. The establishment of indicator weights, based on the farmer’s perspective,
ensures that the assessment takes into account the indicators that are most relevant
to the farmer in his/her own context. The scoring of the indicators is based on the
changes the farmers observe on their own farms as a result of the improved practice
and allows for a comparison with the farmers’ previous or ‘business-as-usual’
practice.
The protocol was tested among 72 farmers in the climate-smart village of
Lushoto. The data was analysed using Microsoft Excel and different CSA interventions assessed for their contribution to adaptation and productivity.
14 A Participatory Approach to Assessing the Climate-Smartness of Agricultural…