Researching on sanitation behaviour and practices in low- income settlements: the need for sensitivity, skill, and creativity

ABSTRACT

Due to inadequate sanitation services, residents of low-income settlements practice open defecation or use other methods for faecal waste disposal. Such residents may not easily disclose their sanitation practices or behaviour since it requires them to disclose information that may be considered personal. Research on sanitation practices in these settings requires skill, creativity, and appropriate data collection methods. This article draws insights from research conducted in low-income settlements of Kisumu city in Kenya and explains how to skilfully and creatively collect data on sanitation practices. The research began with an exploratory study, followed by a quantitative and a qualitative phase. An initial exploratory stage led to an understanding of the context, the quantitative stage identified the sanitation practices, and a subsequent qualitative stage explained why residents engaged in some sanitation practices. Probing, use of skits, and group discussions encouraged respondents to openly talk about their practices. I posit that understanding the local context using appropriate data collection methods and flexibility to learn and unlearn are necessary when researching on sanitation practices, especially in low-income settlements. Documentation of the methods and experiences would further guide other researchers and practitioners in the sanitation field.

KEYWORDS:

• Low-income settlements
• sanitation
• sensitive research
• skits

Background

Low-income settlements are characterized by poverty, tenure insecurity, poor/substandard housing, overcrowding, and lack of basic services like safe water and sanitation (Cranby 2012, Nuissl and Heinrichs 2013, Harris 2014, UN-Habitat 2014). Due to inadequate sanitation facilities in these settlements, residents make do with practices such as open defecation or sharing the available facilities, which are often in deplorable conditions.

Figure 1. Map of Kenya showing the location of Kisumu city.

Source: https://www.worldatlas.com/img/locator/city/063/13063-kisumu-locator-map.jpg

Within such an environment of inadequacy of sanitation services, sanitation behaviour and practice can be challenging and unpleasant for residents to ‘truthfully’ talk about. Researching on sanitation practices in such an environment can be considered as research that is ‘sensitive’. Sensitive research is that which requires respondents to disclose opinions, behaviours, or attitudes that they would normally keep private (Liamputtong 2011, p. 108). The term ‘sensitive research’ is often used to denote research topics/questions that are socially unpopular and which people believe threaten their presentation of themselves (Neuman 2011, p. 320). A sensitive topic may influence the information that respondents provide as they may misreport or not provide any information.

Researchers of such topics as sanitation practices in low-income settlements should, therefore, adopt strategies to ensure that their research question is fully answered while being mindful of the challenging living conditions of their respondents. Such strategies may involve using appropriate and suitable data collection methods. These methods should adequately capture sanitation behaviours and practices, while ensuring that respondents provide accurate information.

Several sanitation studies have been carried out in low-income settlements in Africa and Asia, each of them using various designs and data collection methods (Schouten and Mathenge 2010, Isunju et al. 2011, Heijnen et al. 2015, Alam et al. 2017, Tsinda et al. 2017). However, few of these studies have documented their insights, reflections, and lessons learnt while researching on sanitation practices in these settlements. Drawing from sanitation research I conducted in the low-income settlements of Kisumu city in Kenya, I share the strategies used as well as the insights gained when researching on such a sensitive topic as sanitation behaviour and practices.

Kisumu city’s low-income settlements

Kisumu city is situated in Kenya’s western region by Lake Victoria (Figure 1). The city had a population of approximately 500,000 people as per the 2019 population and housing census (Kenya National Bureau of Statistics 2019). It is estimated that approximately 60% of the residents in the city live in informal settlements characterized by poverty, substandard housing, and lack of services such as sanitation (County Government of Kisumu 2018). These settlements are spread around the city, as shown in Figure 2, and they include Nyalenda, Bandani, Obunga, and Manyatta.

Figure 2. Map showing the low-income settlements in Kisumu city. Source: Karanja 2010.

Most of the land in the settlements is freehold, mainly obtained through inheritance (Smith 2017, County Government of Kisumu 2018). Due to population growth and demand for housing, some landowners have constructed rental housing units on their land thereby living with tenants, while others live elsewhere and have rented out the housing units to tenants (commonly referred to as absentee landlords). These absentee landlords usually appoint one of the tenants (commonly referred to as a caretaker) to be in charge of any tenant matters within the plot.

Approximately half of the plots in the settlements lack sanitation facilities, and ‘flying toilets’ (the practice of defecating in a bag and flinging it away) is a common practice (Karanja 2010, Simiyu 2015). Available sanitation facilities are mostly traditional pit latrines (Letema et al. 2014) which are shared among households living on the plot (Karanja 2010, Simiyu 2015, Simiyu et al. 2017b) with a typical example of such pit latrines shown in Figure 3.

Figure 3. An example of pit latrines in Kisumu’s low-income settlements. Photo credits: Author’s own.

Findings from the settlements shows that these shared sanitation facilities are oftentimes not hygienically clean (Simiyu 2015, Simiyu et al. 2017b).

Research design and data collection methods

This article is based on research conducted in Kisumu’s low-income settlements between 2013 and 2016. The research covered aspects such as

• An overall understanding the sanitation situation in the settlements (stakeholders involved in sanitation provision, available sanitation facilities, socio-cultural dynamics influencing sanitation provision and use).

• Socio-economic factors influencing access to sanitation facilities at the household and community level.

• Hygienic quality of the available sanitation facilities and factors contributing to the prevailing conditions.

• Decision making of users regarding the sanitation facilities in the settlement.

The research was carried out in a number of sequential stages.

A review of literature on sanitation in the settlements was first carried out, leading to identifying research that had been carried out and stakeholders who were involved in sanitation activities within the settlements. Documents that were reviewed included peer-reviewed journal articles, project and government reports, books, and conference proceedings. This review highlighted scanty information on sanitation in Kisumu, therefore, revealing the need for an exploratory study to understand the study context, assess the sanitation situation in the settlements, and further refine the research question. I applied a number of methods during the exploratory stage, including review of project documents, transect walks, interviews, observation of sanitation facilities, and focus group discussions. This stage lasted for approximately one month and findings were useful in refining the research design in view of the context. A total of 51 IDIs and 8 FGDs were conducted.

A robust investigation of sanitation practices required the use of multiple methods of data collection with a continuous reflection of the results and the research process. A mixed-methods approach was deemed appropriate as it would ensure a more comprehensive investigation of the research problem (Creswell 2015:15; Creswell and Clark 2011, p. 11). An explanatory sequential mixed-methods design was selected; beginning with a quantitative phase followed by a qualitative phase (Creswell 2015:38; 2014:224, Klassen et al. 2012).

Quantitative data were collected through a survey that lasted approximately 3 months (January to March 2014) with a sample of 180 compounds to assess socio-economic conditions related to sanitation access. Activities preceding the surveys included recruiting and training research assistants, contacting community leaders, and piloting of data collection tools. Research assistants used structured questionnaires to collect information through face to face interviews. The assistants worked in pairs, with one person engaging the respondent and maintaining contact with the respondent, while the other person recorded the responses. Assistants inspected sanitation facilities of each respondent (if available).

After the survey, I carried out an initial analysis to summarize the data and identify areas that required follow-up and in-depth probing. The next stage entailed conducting 40 in-depth interviews and lasted for approximately 1 month. The interviews were guided by an interview tool that had open-ended questions which were posed to the respondents. If there were any issues that were not clear, research assistants probed further for clarification. Researchers also observed sanitation practices and took note of hygienic conditions of shared sanitation facilities.

After the survey and in-depth interviews, we held group discussions in order to gain an understanding of the social aspects of sanitation in the settlements. Experiences from the earlier stages informed the criteria for selecting participants for group discussions. Together with community leaders, we selected participants based on residence status (landlords or tenants) and availability of sanitation facilities; thus, discussions were held with several groups, e.g. landlords who lacked sanitation facilities, or tenants who had sanitation facilities.

During respondent selection, more participants were identified to cater for respondents who would decline to participate (Cohen et al. 2011, p. 437, Green and Thorogood 2014, p. 143). The selection process and the group discussions were done concurrently over a number of days until all the discussions were done. Having the selection and the discussions being done concurrently ensured that invited participants had enough time, i.e. neither too early an invitation for them to forget nor too late for them to decline the invitation due to other competing commitments.

The number of participants during the discussions varied between 8 and 15. The meeting began with an ‘ice breaker’, which was a skit of approximately 5 minutes that depicted sanitation experiences commonly encountered by residents in the settlements (see box 1 for a synopsis of one of the skits). The skits were tailored for the participants in each FGD, depicting the absence/presence of a toilet, type of residence, practices, and dilemma over the solution.

The moderator initiated the discussion by making reference to the play, and from this introduction, the rest of the discussion followed; being guided by questions in the discussion guide. The note taker recorded the response/interaction on an audio device and took notes in a field notebook. This phase lasted for approximately one month with a total of eight FGDs. A summary of these research phases is summarized in Figure 4.

Figure 4. Flow process during data collection in Kisumu’s low-income settlements.

In addition to these methods, other strategies were used to ensure the validity of the results. Four research assistants were recruited based on their understanding of the local language and cultural context, their gender (two men and two ladies), their residence (they needed to be from the community), and previous experience in conducting similar research.

Box 1: A synopsis of one of the skits

Table

Respondents: Landowners (without tenants) and lacking toilets A man (Baba Lupita) enters the stage carrying a portable radio while singing and dancing along to the music. Suddenly, he puts the radio down while holding his stomach as though in pain. He calls out for his wife, Mama Lupita, who responds but does not show up immediately. Desperate, Baba Lupita calls out for her again, asking her to hurry up. She finally comes on to the stage saying she delayed because she was getting dressed. He is furious that she took long to respond to his call yet he is in pain. He asks her to bring some toilet paper as he is having a stomach upset. She, however, covers her nose because of the smell from her husband’s fart. Surprised, she asks him if he is asking for toilet paper because they do not have a toilet, and the norm is to dig a hole behind the house and bury the contents. Besides, they use leaves instead of toilet paper, and so his request, according to Mama Lupita, must have been because they had guests (in reference to the audience). He asks her not to embarrass him before his guests, but rather to bring him toilet paper because it was urgent. She walks out, while he paces about in pain, and calls out to her asking her to hurry up. She soon comes back with some papers for him to use, and puts them in his pocket. She then directs him to the neighbouring plot, where there is a toilet that is not locked, and asks him to sneak in through the fence and use the toilet. She further advises him to be very cautious and follow her instructions to the latter so that he is not caught sneaking in. In the meantime, she leaves for work. Baba Lupita is seen sneaking in to their neighbour’s plot, but just before he lets himself into the toilet, he is discovered by the landowner who starts yelling at him. Baba Lupita starts explaining that he is in urgent need of a toilet and there was nothing he could do other than use his neighbour’s toilet. He begs her (the owner) to allow him to use the toilet. The owner continues yelling at him, saying that he looks down on her because she is a widow, and that he has been sneaking in to her toilet for the last three years. Baba Lupita explains that he does not have a toilet where he lives, hence the reason for his asking to use her toilet. He begs, saying this would be the last he requests to use her toilet. The landowner continues yelling, saying they (households from Baba Lupita’s plot) should pay for using her toilet. Baba Lupita continues begging her, saying it would be the last time. She continues lamenting that the toilet is almost collapsing and she does not have anyone to help her in its maintenance. Baba Lupita again begs that this would be the last time; but she responds that he has been saying the same thing for the last three years and it seems he is playing hide and seek. She then leaves saying she is on her way to report him to the chief. He begs her not to, saying he would not sneak into her toilet again; but she is long gone. Baba Lupita is now alone, and he laments that he has had enough insults from his neighbour because of always sneaking into her toilet. It was time he constructed his own toilet. He wonders (aloud) why he has not constructed his own toilet, if he has enough space within the plot to install a toilet, if he has people who could assist, who he could consult, and how he will keep the toilet clean. He leaves the stage, still wondering aloud...

These assistants were recruited at the beginning of the study and were engaged throughout the study. After recruitment, the assistants were thoroughly trained on several aspects including the data collection tools and methods, ethics in research, skills needed to conduct research on sanitation practices in the settlements, engaging respondents in individual or group discussions, and on inspection of sanitation facilities. The training sessions had several sessions of role play where the assistants practiced how to approach, interview, and engage respondents. The interviews, discussions, and skits were conducted in a language that was suitable for respondents/participants. During the survey, the assistants worked in pairs comprising a male and female researcher with one assistant interviewing the respondent and another inspecting the toilets. Debriefing sessions were held at the end of each day to discuss any unusual events or happenings that needed a change of strategy or clarification.

The research was approved by Stellenbosch University Research Ethics Committee and a permit was received from the Kenya National Commission of Science and Technology (NACOSTI) (No.: NACOSTI/P/14/5546/781). All participants were given full information about the study, and allowed to consent to participate in the study. Participants preferred to give a verbal rather than written consent (out of concerns of being implicated or identified later, and not having any direct benefits from the study). More details on sampling, sample size, and results from all the phases have been detailed in peer-reviewed journal articles (Simiyu 2015, 2017, Simiyu et al. 2017a, 2017b, 2017c, 2018).

Outcomes from the research design and data collection methods

Exploratory study

The exploratory study led to an understanding of the settlements, including the geographical expanse, layout of housing, residents living in the settlements, availability and types of sanitation facilities, community norms, and identification of sanitation practices and behaviours. I noted that researching on sanitation practice and behaviour in the settlements needed careful thought and planning, and the use of multiple methods that complement each other. Findings from the study led to refining the research questions as well as the data collection methods for the main research.

Surveys

During the survey, I noted that some respondents did not freely discuss their sanitation practices, especially if they lacked sanitation facilities or had poor quality sanitation facilities. Using face to face interviews enabled the researchers to ‘connect’ with the respondents, capture non-verbal communication and emotions, and to physically inspect the available sanitation facilities. A few respondents sometimes provided ‘acceptable’ responses about their sanitation facilities or practices; for example, some respondents mentioned that they had sanitation facilities within their compounds when they did not have any, others mentioned that they lacked sanitation facilities when they actually had sanitation facilities (although they were full pit latrines), and others showed discomfort in responding to the sanitation questions especially if their latrines were full. This discrepancy could be picked out because the tool had follow-up questions to counter-check the responses given by respondents. Any extra information given by the respondents was noted by researchers in notebooks, and discussed during the debriefing sessions. The debrief meetings revealed challenges that the research assistants and the respondents faced with regard to providing truthful information, and research assistants were encouraged to always seek clarification from respondents. This stage led to the identification of some sanitation practices answering the ‘what’ questions of the research. The stage also highlighted the ‘sensitivity’ of the research question, justifying the need to use different data collection methods/approaches in subsequent stages that would provide further details to the ‘what’ questions.

Interviews

Interviews allowed the respondents to provide more details of their sanitation practices. Experience from the previous stages enabled the researchers to understand the respondent’s ‘language’ in reference to sanitation practices; for example, as noted from the survey, sometimes respondents mentioned that they had latrines when in actual sense they did not have latrines-perhaps to avoid the embarrassment of not having a latrine. We, for instance, interviewed a lady who initially reported that she had a latrine within their plot. Upon further probing and discussion, she confessed ‘… we really do not have toilets in this plot… we all pay to use that toilet. She clarified that she and her neighbours all used a toilet from the neighbouring plot which they considered their toilet because they paid to use it. The lady further narrated the challenges she went through while using another of her neighbour’s toilet, saying ‘… I borrowed from my neighbours, but sometimes they chased us away …. my children were chased from using the toilet … they would even soil their clothes because they were chased from the toilet …’.

It was possible to elicit such in-depth information because the tool had open-ended questions that allowed probing when responses were inadequate and through such probing, respondents provided more detail about their sanitation practices. Thus, in addition to answering the ‘what’ questions, the interviews enabled answering the ‘how’ and ‘why’ aspects of the research. Results provided insight into how and why residents engaged in some sanitation practices. During the debrief sessions, we still discussed our experiences with regards to interviewing the respondents, and their responses to the questions. By this time, however, we the researchers had learnt how to engage the respondents who felt embarrassed and it was easier to elicit responses from the respondents. Additionally, we began exploring the possibility of using creative methods during data collection and we all proposed trying out skits.

Use of skits/theatre

During the skits, sanitation aspects that would normally not be discussed were highlighted in an entertaining manner to the participants. Participants openly shared their ‘embarrassing’/sensitive/personal experiences, which they would not easily have spoken about. They identified with the issues portrayed in the skits, and made mention of experiences they had encountered that were similar to those depicted in the skit. Relating to the poor state of pit latrines, one landlord related to the skit by saying

My toilet is also just like that [in the skit], it is almost collapsing so I am not using it

One tenant admitted that she had experienced an embarrassing situation where she was found sneaking into a toilet in a neighbouring plot:

The same [in relation to the skit] happened to me. I went to use the toilet at our neighbouring plot … . but the owner found me out.

A tenant also used the skit to express her experience in using her neighbour’s latrine:

‘Just like we saw in the skit … . We ask for permission, but they [their neighbours] refuse because we are always begging … we wait until they [the neighbours] are indoors; then we sneak in and use the toilet’

Additionally, participants shared the lessons they had learnt admitting that the plays were educative. One landlord stated

‘I have learnt that everybody should make an effort and construct their own toilet, even just an ordinary pit latrine’

Similarly, another landlord said

‘I have learnt that as a landlord, I should be aware that my tenants need a toilet, and tenants should ensure that the toilet is clean’

Speaking of sharing sanitation facilities, a tenant shared her lessons saying

‘I have learnt that people living in the same plot should ensure that the toilet they use is clean’

Another tenant shared his insights concerning health and improving conditions in their community:

‘The play has enlightened me about health issues … . for example, Baba Lupita was complaining of stomach upset … . maybe because of not having a toilet … . we need to have toilets in our community’

Use of group discussions

Within the groups, participants shared similarities such as residence type, gender, and sanitation status. These similarities within the groups enabled participants to interact openly among themselves, while giving reasons for some of their practices. Participants communicated with, and learnt from each other; realized the importance of sanitation from a community perspective, and further proposed actions that they needed to take to improve sanitation and hygiene practices in their community.

A highlight during the research, however, was that although the topic seemed embarrassing, through engagement and discussions, the researchers and respondents learnt how to handle the subject matter by making fun of some practices. Discussants in the group discussions and interviews narrated their challenges and experiences in a laughable manner. One respondent, for instance, narrated how lack of toilets can lead to using plastic bags during the night. He went on to mimic how to dispose the waste in such a plastic bag saying ‘… in the morning, you have to be tactful…you do not want to be seen carrying the bag by anyone, lest they ask you what the contents of the bag are…’. Another lady further described that they locked a stranger in their toilet for a whole day as punishment for sneaking into their toilet; and yet another discussant in one of the group discussions explained that they chased a stranger who had sneaked into their toilet by hurling stones at him. Such narratives often made the discussion easier, interesting and engaging to us the researchers as well as the respondents.

Discussion

This article has detailed the process I took to research on sanitation practices in low-income settlements of Kisumu. Key highlights in this reflection are the use of appropriate data collection methods, the importance of understanding the local context, the need to be adaptive and use creative ways of data collection, and the need for skills in engaging respondents especially when the subject is sensitive or embarrassing. Sanitation practice (including defecation) is generally regarded as private behaviour and in many communities, such practices are rarely discussed in public. Residents of low-income settlements have to contend with the shame of not having access to sanitation facilities and using alternative options-such as open defecation-which may be embarrassing to disclose since nobody would want to be associated with such practices. I encountered this situation in Kisumu’s low-income settlements, as residents were often embarrassed to openly talk about their sanitation practices.

This sensitivity of sanitation practices is alluded to by McGranahan (2015) who correctly notes that sanitation entails private behaviour which people do not talk about publicly yet it has public impacts which people have little they can do about on an individual capacity. Sensitive topics such as sanitation behaviour require that researchers be adequately prepared before embarking on research. Preparation entails knowledge of the socio-cultural, economic, and political background of the study area and the research subjects, which determines the data collection process. In this study, preparation included an exploratory study that provided first-hand experience of sanitation practices in the settlements. The importance of such a study is that it enables the researcher to explore areas which he/she has little knowledge about, and thus to make decisions on the need for a detailed research (Kumar 2011, p. 11). Researchers have made use of such approaches from studies in Kisumu, Kampala, and Kigali (Tsinda et al. 2013, Okurut and Charles 2014, Okurut et al. 2015). Such studies provide information that aids in making decisions on research design, sampling, and the use of appropriate data collection methods.

It is a good practice for researchers to make use of designs and data collection methods that answer the study objectives/questions (Guest 2012, Sandelowski 2014). This research entailed using qualitative methods that complemented and explained results from an initial stage of quantitative methods. A similar approach has been applied in sanitation studies in Kenya, Uganda, and Rwanda (Tsinda et al. 2013, Okurut and Charles 2014, Okurut et al. 2015, Simiyu 2016), others have used purely quantitative approaches (Heijnen et al. 2015, Nelson et al. 2014; Innocent Kamara Tumwebaze et al. 2013; Innocent K Tumwebaze and Mosler 2014, 2015; Innocent K Tumwebaze et al. 2014), concurrent quantitative and qualitative methods (Kwiringira et al. 2016), and yet others began with qualitative studies that informed a quantitative survey (Lagerkvist et al. 2014). It is posited that qualitative methods are more suited for sensitive research since they are flexible and allow people to express their own reality (Dickson-swift et al. 2008, Aldridge 2014, Lambrechts 2014). In this research, in-depth interviews and group discussions were used during data collection. Other studies on sensitive research have also alluded to the benefits of combining interviews and group discussions (Lambert and Loiselle 2008, Pool et al. 2010, Lambrechts 2014, Zeldenryk et al. 2014), including several studies on sanitation practices (Murungi and van Dijk 2014, Kwiringira et al. 2014a, 2014b, Bisung et al. 2015, Sahoo et al. 2015, Shiras et al. 2018). The advantage I encountered with the study design and the combination of qualitative methods, which is also alluded to in other studies is that there is more openness among respondents, which results in greater understanding, improvement of data quality through validation, and data that is richer due to the ability to probe for clarification.

One main highlight in this research was the need to be skilful and creative during data collection (Kumar 2011, p. 138, Neuman 2011, p. 47). Although I combined various data collection methods, I still integrated skits during data collection to encourage openness. Participants related to the experiences depicted in the skits which encouraged them to freely share their experiences; possibly, because they realized that other members of the group also had similar experiences. The use of drama/skits in data collection, referred to as Theatre for Development (TfD), Role play, Participatory theatre, Theatre for education or Popular Theatre; and has been used to investigate delicate health and social matters (Cox et al. 2009) such as sexual practices (Mbachaga 2010), Female Genital Mutilation (FGM) (Ugwu and Ashaver 2014), and sexual violence (Christensen, 2014). In sanitation research, theatre has been used mainly to raise awareness and achieve community participation on sanitation issues (Mushangwe and Chivandikwa 2014, Trivedi and Patel 2016). The benefits of using theatre also experienced in this research are that they facilitate and promote dialogue, raise awareness, motivate community members to think of appropriate solutions, and stimulate social and behaviour change (Mbachaga 2010, Sloman 2011). Aside from using theatre, other studies on sanitation behaviour and practices among women have explored interactive methods such as sanitation walks (Cristine et al. 2019) and the use of cards that depict sanitation behaviours (Hulland et al. 2015).

Furthermore, this research highlighted the need to be skilful in understanding the responses given by respondents within the research context. Skilfulness may entail taking note of the respondents’ body language and non-verbal communication. In this research, for example, respondents oftentimes responded with their body language, which required probing to understand the response. Whereas some body language is clear (such as nodding one’s head), some of it requires attention and engagement with the respondents. We particularly learnt to listen to women, who due to embarrassment, did not freely admit that they did not have sanitation facilities. Their responses, for example, were through their facial expressions, such as a shy smile or a situation where the respondents looked away and did not respond to the questions. In this case, skills were needed to further engage the respondents, for example, by using various probing approaches.

Finally, the research process in this study required that we (the researcher and the assistants) continuously think through and evaluate the process as well as the data collection methods to assess if there was a need for modification. Since the research was carried out in phases, results from each phase informed the next phase and confirmed the results from the previous phase. Regular reflection and debrief sessions made us more familiar with the study context and research subjects, enabled identification of unforeseen gaps, and led to devising appropriate data collection methods. The use of skits as an icebreaker, for example, was conceived during the research as a way of stimulating open talk. In addition, the use of local language was conceived during the early stages of the research and it helped respondents to express themselves better, especially as pertains sanitation practices. This process of thinking through the research process often involves being aware of research methods, and the impacts that these methods have on participants (Gavin 2008, p. 54). Continuous reflection leads to using strategies that yield better results and to drawing lessons that are beneficial to the researchers and the research process.

Conclusions

In low-income settlements where sanitation facilities are either lacking or inadequate, researching on sanitation requires careful thought and consideration. Residents in such settlements may be engaged in sanitation practices that are embarrassing or shameful, and they may not disclose their practices due to the fear of being associated with shameful sanitation practices. By drawing from sanitation research conducted in Kisumu city’s low-income settlements, I have shared my research journey, and illustrated that researching on sanitation practices in low-income settlements requires attention to detail, adequate planning, flexibility, skilfulness, and continuous thinking through the research processes and the results. Selected methods need to complement each other, so that what one method is incapacitated in achieving is captured by the other method(s). Such skills and techniques create an enabling environment for participants to openly discuss otherwise ‘unpleasant’ or ‘sensitive’ topics. Although these lessons may be applicable to Kisumu and other similar contexts, it is important that researchers design their strategies according to the local context and document their experiences, so that other researchers and stakeholders draw lessons from the insights shared.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This article was made possible with U.K. Aid from the Department of International Development (DFID) as part of the Sanitation and Hygiene Applied Research for Equity (SHARE) Research Consortium (www.shareresearch.org). However, the views expressed do not necessarily reflect the Department’s official policies.

Notes on contributors

Sheillah Simiyu

Sheillah Simiyu is a public health researcher in water, sanitation and hygiene (WASH) issues, and how these link to health and development. Her work and research interests in WASH cut across child health, gender, community engagement, psycho-social well-being, and institutional WASH.