Geoscience Reference
In-Depth Information
Table 13.1
Summary Statistics
Variable
Obs.
Mean
Std. dev.
Min
Max
Total monthly income (rupees)
640
27,749.31
61,315.5
3,000
724,000
Savings
640
4,095.7
4,971.79
0
200,000
Hasloodexperience
640
0.79
0.41
0
1
Past flood was worst
640
0.07
0.26
0
1
Number of floods experienced
627
1.16
1.04
0
6
Feels better prepared than others
640
0.31
0.46
0
1
Hasheldinsurance
640
0.10
0.30
0
1
Changed precautions since 2010
640
0.47
0.50
0
1
Learned new mitigation techniques
640
0.23
0.42
0
1
Lives in flood cluster
640
0.50
0.50
0
1
Hasalocal“patron”
640
0.18
0.39
0
1
Adopted new techniques
640
0.51
0.50
0
1
Ability to recover faster than others
640
0.20
0.40
0
1
Took any mitigation action
640
0.33
0.47
0
1
Reinforced house structure
640
0.15
0.36
0
1
Used sandbag barriers
640
0.18
0.38
0
1
Moved possessions to a higher level
640
0.22
0.41
0
1
Received any warning
640
0.76
0.43
0
1
Received warning from radio
640
0.20
0.40
0
1
Received warning from television
640
0.28
0.45
0
1
Received warning from neighbors
640
0.48
0.50
0
1
Received warning from relatives
640
0.20
0.40
0
1
Received warning from mosque
640
0.30
0.46
0
1
Received warning from govt. worker
640
0.02
0.15
0
1
Preparation time (hours)
640
50.52
79.98
0
492
Had24+hourspreparationtime
640
0.39
0.48
0
1
Time spent evacuated (hours)
640
947.56
2,810.6
0
34,560
Householdlossvalue
640
61,827.03
146,803.4
0
2,500,000
Surprised at the level of damages
501
0.87
0.33
0
1
Flood propensity score
640
0.40
0.19
0.15
0.778
Age
384
37.77
12.58
16
80
Female
640
0.50
0.50
0
1
As shown in Table
13.2
, receiving any warning significantly increases the
probability of taking mitigation actions. Only the face-to-face information sources
(neighbors, government officials, mosque announcements) have a significant effect
on mitigation behavior. Remote information sources (radio, television, relatives
from other towns) do not have a significant effect on mitigation behavior.
Most people report neighbors and friends as the most important information
source, followed by mosque announcements. This self-reported priority order is con-
sistent with our empirical observation of the sources that actually prompt mitigation
behaviors; however, we find that different sources of information are likely to prompt
different types of mitigation behavior. When disaggregating the types of mitigation
action taken, we see that radio, neighbor, and mosque warnings significantly increase
Search WWH ::
Custom Search