Transcript Creation and validation of FCS (2011)

Food Consumption Scores and
Food Consumption Groups
Creation and Validation
Data Analysis plan
 Create the Food Consumption Score
 Group foods by category (cereals, pulses, vegetables, fruit, meat &
fish, milk, sugar, oil
 Create the score based on frequency of consumption by group times
weight of each group and summing each group
 Validate the FCS as a Food Security Indicator
 Correlation with various other indicators
 Create Food Consumption Groups to answer: Who are the food
insecure? How many are they? Where are they? Why are they
insecure?
 Create basic statistics on how many in each group
 Explore further who they are by various strata (agro-ecological zone,
governate, urban vs. rural, etc)
Validating the FCS
Correlations with FCS
Reduced CSI
Wealth Index
Percent of total expenditure
on food
How many times In the past 7
days did not have enough food
or money to buy food
Correlation Coefficient
-0.329
Sig (2-tailed)
0.00
Correlation Coefficient
0.546
Sig (2-tailed)
0.00
Correlation Coefficient
-0.124
Sig (2-tailed)
0.00
Correlation Coefficient
-0.321
Sig (2-tailed)
0.00
FCG Thresholds
 An FCS of 21 in Yemen is composed of oil, sugar and
cereals (staple)
 Because the value of sugar and oil consumption was
daily (7 times a week), we have used a higher threshold
for grouping. In order to properly evaluate this high
sugar and oil diet as poor, we have changed our
thresholds as follows:
 Poor < 28
 Borderline 28.5 – 42
 Acceptable >42.5
FCG by Governate
‘Rayma’
‘Ad Daleh’
‘Amran’
‘Al Mahra’
‘Al Mahweet’
‘Mareb’
‘Lahej’
‘Aden’
‘Sana'a’
‘Shabwa’
‘Dhamar’
‘Hadramout’
‘Hodeidah’
‘Hajja’
‘Taiz’
‘Al Bayda’
‘Sana'a City’
‘Abyan’
‘Ibb’
poor
borderline
acceptable
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
 Rayma, Ad Daleh, Amran and Al Marha show the highest
proportion of poor food consumption group
FCG by Agro-ecological zone
‘North Highland’
‘Desert ’
‘Valley’
poor
borderline
‘South Coast’
acceptable
‘Red Sea Coast’
‘Highland’
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
FCG by Urban / Rural
FCG by Main Livelihood Activity
FCG by Household Status
FCG
Literacy Rate
High
Dependency
Poor
39.1%
30.1%*
Borderline
50.3%
26.9%*
Acceptable
62.2%
18.0%
*The difference between Poor and Borderline in regards to high
dependency was not found to be statistically significant
Conclusions
 FCS is strongly correlated with other key food security
indicators
 The dietary patterns of the poorest households (in terms
of food consumption) are highly reliant on sugars and oil
 Generally, rural households are worse off as are
households in the North Highlands
 More analysis is needed to further profile the most food
insecure