Introduction
For wealthy people from wealthy countries, it may not always be easy, though usually quite feasible to be socially distant: Just stay home as much as possible, maybe read more books or watch a little more Netflix, and keep a distance from others when leaving the house.
Things are not that simple for the more than 800 million people living in informal settlements in the Global South. Many of them rely on daily wages, meaning that any day they don’t leave the house is one they will not find food. Further, informal settlements are densely populated, have shared toilets and showers often located outside the house, making social distancing extremely difficult.
The tension between meeting basic needs and social distancing
In recent months, many efforts have been made to find out how social distancing can be applied in such neighborhoods and what effect it has on residents. A study conducted with inhabitants of Nairobi’s informal settlements in May, for example, found that social distancing measures were widely applied during that time, and that they were quite effective in reducing physical contacts, but they came at large costs. Most respondents stated that the pandemic had caused them to skip meals more often than usual, and 86% reported having suffered a partial or complete loss of income. Overall, very few scientific articles have been published on this subject so far. Field studies are currently impossible in many countries, and not many researchers have the infrastructure to reach informal dwellers remotely.
With a second wave of COVID-19 cases hitting many emerging markets, governments must re-evaluate whether the measures they have taken to mitigate the spread of the virus are sufficient. They have a better, though still limited, information base than they did in March. Against this background, we decided to use our experience and infrastructure to ask residents of the Kibera informal settlement in Nairobi about their practices and perceptions related to social distancing. We conducted four survey rounds with the first taking place on 23 July 2020 and the last taking place on 15 October 2020.
Not surprisingly, we found that the respondents have been suffering from the restrictions imposed upon them and that their discipline in practicing social distancing has been declining. Towards the end of the study, they left their homes more often and tended to have more physical contacts. A majority of them said that social distancing had negatively affected their satisfaction with life. As we also wanted to examine the mechanisms behind these findings, we additionally sought to answer the following questions:
- Why do respondents leave their homes and experience physical contacts?
- What are their perceptions of the social distancing measures?
- Have they recovered from the income shocks?
- And how do all of these factors vary by demographic group and over time?
Our Research Design
We created a web-based questionnaire with 28 questions about different aspects of social distancing. We then randomly selected 1000 respondents from the informal settlement Kibera and sent them SMS invitations on four different dates with four weeks in between. Respondents could fill out the questionnaire on their phones and received KES 100 upon completion.
Between July and October, when data collection took place, many social distancing measures had already been relaxed or lifted by the Kenyan government. Nonetheless, a nighttime curfew was still in place, most educational facilities were closed and all bus drivers were ordered to operate at half capacity. The number of confirmed COVID-cases per day peaked at 960 on July 26, then decreased to around 100 and started to rise again in the second half of September. Although there is hardly any public data on the geographical distribution of the virus in Kenya, it is evident from the daily notices published by the Ministry of Health that most confirmed cases come from Nairobi, and that the research environment of Kibera is also affected.
Our Results
Across all four waves, we collected 750 complete surveys. Here are our five key conclusions:
- When respondents leave their homes and experience physical contacts, it’s mostly because they are fulfilling their basic needs
Food shopping, using a shared toilet and getting water from a water joint — these were the three options that were chosen by far most frequently when we asked respondents why they had left their homes on the day before the survey. Across all four waves, places that refer to leisure activities such as sports pitches, churches or restaurants were barely chosen. This is very similar to the answers we received when we asked respondents where they had experienced physical contacts with non-household members on the day before the survey: Most contacts had happened in food shops, shared water joints, shared toilets and in public transport.
But, looking at how the contact patterns changed over the course of the study, it appears that some respondents have started visiting leisure facilities again. The share of respondents who experienced physical contacts in a church or mosque more than doubled between wave 1 and wave 4, and increased by 40% for sports pitches. This is roughly in line with the lifting of restrictions. Churches started to slowly re-open at the end of July, and sporting events partly resumed in September.
2. It’s harder for men to be socially distant than for women, and they seem to practice social distancing less strictly
According to our data (fig.2), men take social distancing less seriously than women. In all four waves, men on average stated that they practice social distancing less strictly (0.02<p<0.92) and that they leave their homes more often (p<0.05 for all waves). They also tend to have more physical contacts than women, although the evidence is weak.
However, this does not explain why men experience social distancing and the lack of physical contacts much more negatively than women. Between 59% and 63% of men said that it negatively affects their satisfaction with life (fig. 4), compared to 44% to 47% for women (Fig. 3). And whereas up to 37% of women experienced a positive effect, only 11% (July) to 24% (September) of men did. It’s impossible to pinpoint exactly why this is the case, but one possible explanation is linked to money.
3. Respondents’ income seems to have increased over the course of the study
In Kenya it is still often the case that men are the main earners and women run the household. This also seems to be true for many respondents. When we asked how much they had earned in the month prior to the survey (fig.5), men’s income was about twice as high as that of women. It is therefore likely that the income shocks caused by the pandemic have hit men harder than women, and this may have affected their satisfaction with life. This thesis is supported by the fact that life satisfaction in our sample is positively correlated with income (p=0.02).
If looking at how income has changed over time, no trend can be seen at first sight. Respondents’ median income was KES 3,000 in July and KES 4,000 in all other months, and the mean monthly income was highest in July (KES 6,252) and lowest in August (KES 4,811). But these numbers are strongly skewed by the answer of one respondent who reported to have earned KES 200,000 in July. When removing this outlier as well as the highest income data point for each of the other waves, there is a statistically significant increase in income of KES 308 per wave (p=0.018). It seems as if respondents are slowly recovering from the income shocks they experienced in March.
4. Most respondents believe that social distancing measures are effective, but their confidence seems to be declining
Respondents were divided on the question whether the social distancing measures are effective to slow down the spread of the virus (fig. 6). A majority said that they were rather effective or very effective, but between 24% (July) and 33% (October) said that they were rather ineffective or very ineffective. There’s a trend towards the latter answers, but it’s not statistically significant (p=0.13).
However, fewer and fewer respondents believe that their neighbors are socially distant. When we asked them in July how many of their ten closest neighbors practice social distancing, the average answer was 4.4. This number declined steadily and in October was only 3.38 (p<0.0001). Taken together, the answers to these two questions indicate that respondents’ confidence in their community’s ability to control the spread of the virus is slowly declining. It also implies that future restrictions by the government might not be taken as seriously as during the first wave of COVID-cases.
5. Special consideration should be paid to public transport
The share of respondents who experienced physical contacts in public transport increased enormously over the course of the study. In October, one out of two respondents reported to have experienced at least one physical contact in public transport on the day before the survey. This indicates an enormous potential for the virus to spread in matatus (minibuses) and boda bodas (motorbike taxis).
Respondents seem to be aware of this. When we conducted qualitative interviews with residents from Kibera in June, they reported that they had reduced their trips in public transport by half; partly because of an increase in fare prices, partly because of wanting to avoid contacts. Some respondents even stated that they would be prepared to wait longer to find a matatu that was following strict safety measures. Nevertheless, in this study, only 15% of respondents ticked the option avoiding public transport when we asked them about the most important behaviors to practice “good social distancing”.
Conclusion
The inhabitants of informal settlements are often faced with the choice of either satisfying their basic needs or adhering strictly to social distancing rules, and governments are often faced with the choice of which rules to impose on them. These decisions are complicated by two factors.
First, not nearly enough is known about the health impact of the virus on younger, poorer populations. Second, there is a lack of information on how informal residents are affected by the restrictions and how they cope with them.
The second research gap in particular is difficult to fill, as fieldwork is currently hardly possible. Remote studies such as this one can contribute to this task, and a few insights have emerged. Respondents seem to be slowly recovering from the income shocks experienced in March, their confidence in the social distancing measures seems to be declining, and men seem to experience social distancing more negatively than women. But remote methods cannot fully replace field research. On-site interviews or observations are necessary to nuance the results presented here and to find out more about the mechanisms behind them.
