Fig 1.
Each dot contains the number of individuals interviewed in that specific area. Note that while region 1 may appear as coastal, it is classified as hinterland by the government given its economic activity and topography. Source: the map was created from the data we collected using the KoBoToolbox.
Table 1.
Descriptive statistics of the respondents interviewed in private houses.
Fig 2.
Path diagram of the system of simultaneous equations.
Circled variables are the latent ones and boxed variables are the observed ones. The arrow from the circled variables to the boxed variables indicates the quality coefficient.
Table 2.
Knowledge level per disease.
Fig 3.
Cumulative frequency of self-reported risk perception across diseases.
As this graph shows the cumulative frequency of risk perception, we start by including the people who had a risk perception of at least 1 (on a scale from 0 to 10). The percentage of people who had a risk perception of 0 can be computed for each disease by subtracting to the sample the proportion of people who perceived a risk of at least 1. For instance, for malaria, 100%-84% = 16% of the sample believed the risk to be 0 (although knowing about the disease).
Table 3.
Self-reported vector control practices.
Table 4.
Results of the structural model.