This week’s post dives further into the 2016 Integrated Household Survey (IHS) collected by GBoS. This post concentrates on crime and crime perception of household members in a community. The questions that we want to answer in this post are;
- In which district is the crime rate highly prevalent?
- What are the characteristics of households that suffered from crime the most, for example the household size and proportion of male youths?
- What is the correlation of crime prevalence with education and income?
Crime rates were recorded if any household member had experienced any crime within the 5 years prior to 2016. We will start off by looking at the occurrence of crime at household level. Then an analysis on the perception of crime prevalence in the community; whether households feel that the rate of crime has increased or decreased. Finally, the occurrence of crime to household members due to household characteristics.
Figure 1: Occurrence of crime at household level at least once between 2011 and 2016
Note: Darker colors indicate a higher crime rate.
Figure 1 above is a heat map of the occurrence of crime at household level at least once between 2011 and 2016. Darker colors indicate higher crime rate and areas on the map with lighter colors indicate that the crime rate is lower in those areas as compared to other areas. It can be seen in the graph that Lower Niumi, Kombo Central, Jokadu and Niani recorded the highest rate of crime (the dark blue colored districts on the map). We learned from the figure that Banjul and Kanifing recorded only the second highest crime rates (the light blue colored districts on the map) there were no high crime rates recorded. The lowest crime rates were recorded in Niamina West, Niamina East, Nianija and Upper Saloum between 2011 and 2016.
Figure 2 below is on the perception of households on the prevalence of crime in the community. If households feel that the crime rate in the community is increasing the color is darker and if households feel that there is a reduction in the crime rate in the community the color is lighter as this can be shown in the figure 2 below. We learned from the figure that households in Niamina West, Janjanbureh, Wuli and Sandu feel that there is an increase in the crime rates in their communities as this can be shown by the dark blue color in these districts. Households in Nianija, Upper Saloum, Kombo South and Kombo Central perceived that there is drastic reduction in crime rates in their communities.
Figure 2: Perception of crime prevalence in the community
Note: If households feel crime is increasing, then color is darker and lighter if households feel crime is decreasing.
Table 1 shows characteristics of urban households that suffer from crime. The results for Table 1 are obtained analyzing only households that live in Banjul, Kanifing, Kombo North and Kombo South. The table indicates that the difference in household size plays a significant role in a household being victimize of crime or not. On average larger households are more likely to be victims of crime as compared to smaller household. Another household characteristic that was looked into was whether the household has a male youth. Male youth refers to a male in the household between 15 to 35 years of age. Results indicate that the number of male youths in a household plays a significant role in a household being victimize of crime, with households that have more male youth being more likely to suffer from a crime than those with less. The result in Table 1 reveals that even at 10 percent significant level average education years in household and total household income are not significant variables in explaining the prevalence of crime to a household.
Table 1: Characteristics of households by whether they were a victim of crime (Yes) or not (No)
|Has Male youth in household||0.588||0.663||0.037**|
|Average Education years in household||7.559||7.498||0.701|
|Total Household Income||99,308.345||122,903.00||0.226|
Note: Standard errors are clustered at variable district.
Note: ***, **, and * indicate significance at the 1, 5, and 10 percent critical level.
What we’ve seen from above is that small households, with a small proportion of male youth are less likely to suffer from crimes. The household income level and average education years of the household are not important determinants of being a victim of crime. This was counter to our expectations but does make sense if one assumes that the perpetrators have knowledge of the household. The perpetrator would be more likely to know a household member, the larger the household is. Given that the crimes reported in the IHS are most likely to be done by male youth, the perpetrator is also more likely to know households with a high proportion of male youth. Although Table 1 shows that income is not a significant determinant of being a victim of crime, households who suffer from a crime still earn more as a group (D122,903), which is not surprising.