Hi All,

I work with Hana on this project and I love the discussion it has spurred. These discussions are are how we continue to improve our methods (and sustainability) in sanitation.

I couldn't agree more that sample size is very important particularly when looking at effectiveness of interventions and I too believe WaSH in general can continue to better by making sure data is statistically valid. I think part of this discussion highlights the difference between qualitative and quantitative research. Typically quantitative research's goal is to have empirical generalization to many verse qualitative research is to gain an in-depth understanding.

Our goal of the focus groups was not to empirically generalize sanitation across all of Kakuma (or even measure impact) but rather gain an in-depth understanding of people's experiences to make improvements.

From the social scientist realm, a focus group can be a great tool for this and the advice for sample sizes fall under a different rule of thumb than surveys for confidence levels. You'll also see the same for key-informant interviews. For example you wouldn't expect someone to conduct 300+ key informant interviews to have an in-depth understanding of sanitation for a particular demographic. The reason for this is the concept of redundancy where after a period of time you start to hear the same thing from subjects of a particular category. For example you'll see in a very widely cited paper titled Focus Groups, Morgan notes the concept of saturation, where each following focus group is not providing new information. www.annualreviews.org/doi/abs/10.1146/an....129?journalCode=soc For refined participant categories and topics, three-four groups per category can be sufficient.

I hope that clarifies some things and happy to continue this discussion as it's an important one. The biggest thing is when looking to gain an in-depth understanding of behaviors and preferences sample sizes differ from quantitative research. I think as we all know quantitative research often lacks this important in-depth understanding and is one reasons we selected a human-centered design approach, as it excels at getting to the root of human experience.

Also, as Hana alludes to this is not our only measurement tool. Since we are doing sanitation as as service it is very important for us to constantly monitor people's us a and satisfaction. We have instituted many small feedback loops to quickly asses and react to strengths and weaknesses. If the system were to scale to other parts of the camp we would continue to use these measurement and feedback loops to asses the service and make real-time changes. I hope that helps in explaining our approach.

I'd love to continue to hear people's thoughts on the advantages and disadvantages of using a qualitative verse a quantitative approach. My previous role was an analyst at a metrics firm for nonprofits so I really enjoy talking about these things.

+++++++
Note by moderator:
Further updates about Sanivation's work continued in this thread: forum.susana.org/169-production-of-bioch...-recording-available

I attached this paper to a post elsewhere on the forum today but just realized it may also be relevant in the context of this thread, particularly as it concerns evidence and reporting of the impact of measures.
all the best,
Patrick

Dear Hana and Patrick,

If I may just interject here, maybe on a bit of a tangent to what you are writing about directly Hana, but it is something I feel is generally not very well described when projects are presented. It concerns the issue of the representativity of survey and focus group results and their validity when applied to the larger population. For me this didn't really come out clearly from your brief presentation of the methodology you are using in the Kakuma Refugee camp.

Like Patrick, I am not a social scientist either but this is an issue that has bothered me quite a bit over the years and I think he has made a fair point in his questions.
The table and text on this web-page give a pretty good overview for determining sample size I think:
www.checkmarket.com/2013/02/how-to-estim...-survey-sample-size/

From this I guess that if you wanted to be 95% confident that the results of the FGD were representative of the broader population you would need to have consulted around 375 individuals.

The question of bias and it having a direct impact on findings is probably vey relevant here as no control measures appear to have been in place to assure random selection of FGD participants. Methods such as the random route or random walk method for household participation may help minimise such bias which could result from community leaders choosing which households participate by themselves.

best regards,
Patrick

Hey Patrick,
Thanks for your interest and I am sorry about the delayed response.

You have some good questions that I would like to respond to here.

1) "Is the sample size large enough?"
It depends on what you mean by "enough". We believe this size will give a good understanding of user acceptability for Somali and Dinka refugees. We cannot say this will reflect all refugees or all Africans but the variant preferences from these two cultures mean that they can perhaps reflect representative practices for more people. This sample size gives us many different household sizes, women and men or different ages and time in the refugee camp, and mobility level. in short we think it is enough to show something significant. Of course we would have loved to have a larger sample but it just wasn't feasible.

2) "Could there be bias in the selection of the participants of the FGDs?"
There should be no significant bias in the participants of the FGDs. Within Kakuma Camp 1, community leaders chose participants without discretion, by going to door to door in areas where the FGD language was prevalent. I suppose one bias could be that families who work more and declined to participant because they did not have time are not represented.

3) "How confident are you on the reliability of the result?"
This largely relates to the first question I think. But in short, we are confident. Beyond the FGDs we have other measurements to show what we did, how well it worked, and what people thought about it. Those will be published with a full report when the pilot is complete.

We are learning new things with every step of this project and would love to hear suggestions or advise from others' experiences.

-Sanivation

Interesting project!

I have some questions about the methodology though:

Is the sample size large enough?
Could there be bias in the selection of the participants of the FDG?
How confident are you on the reliability of the result?

I mean, in the case of an upscale, you want the pilot project to be representative for the 50.000 beneficiaries.

Keep in mind I am not a social scientist (or in particular proficient in statistics) so I apologize if my questions are bad.

Keep us posted!

Sanivation is currently working in Kakuma refugee camp with partners Centers for Disease Control and Prevention (CDC) and Norwegian Refugee Council (NRC), to pilot sanitation as a service including: new toilets, waste collection, solar waste treatment, and reuse among a select group of refugees.

I just recently returned from Kakuma after conducting focus group discussions (FGD) with refugees that will help inform the design of a new pilot sanitation system in the camp. While we already have some design ideas for toilets, it is very important to us that we hear from the users themselves what they need and like. I wanted to share breifly with you some of the methodology we have been using and lessons we have learned using human dentered design in this project in Kakuma.

We are still translating and coding the data from the discussions to pull out the most information possible, but just from scraping the surface we have already learned some interesting things. First, virtually all of the participants, including the Muslim Somalis, said if they had a choice of a house hold toilet, they preferred sitting to squatting. This goes against conventional thinking that Muslims always prefer squatting toilets. Second, we have also learned that the majority of participants were comfortable with the idea of using a cooking fuel or fertilizer made from treated human feces! This is huge, again in cultures where waste is generally seen as extremely taboo, that people are open or even excited about reuse products. [To learn more about our human waste briquetting production visit our Waste to Briquettes page: sanivation.com/our-projects/human-waste-to-briquettes/]

Conducting these focus groups is considered human research, and especially since we are looking at refugees, a vulnerable population, it was important to gain approval from the Kenya Medical Research Institute (KEMRI) Institutional Review Board. If you would like to view our protocol to see the details of the project or use it as a guide for IRB approval for your own project, don’t hesitate to contact This email address is being protected from spambots. You need JavaScript enabled to view it..
A summary of the entire project and the FGD procedures are included below.

Project Summary
The current sanitation norm within Kakuma refugee camp of pit latrines not only enables the transmission of diarrheal disease, but is expensive and requires land. As pits fill every 2-3 years, a new pit must be dug; with the growing refugee population and limited space in Kakuma refugee camp, this method is not sustainable. With the camp already housing over 120,000 refugees and a current influx pouring in from South Sudan, land is growing even more precious within the camp.
Our team is conducting a 6 week pilot of an alternative sanitation system that includes new culturally acceptable toilets, waste collection, waste treatment using Sanivation’s innovative solar concentrator to inactivate pathogens, and finally transforming the treated waste into solid cooking fuel.

The household toilet design will be based on information collected through 12 FGDs held in the camp in March to understand people’s sanitation behaviors and preferences. The pilot is specifically looking at Dinka and Somali speaking refugees residing in Kakuma Camp 1. Somali and South Sudanese refugees not only make up the majority of the camp, but also have distinctive sanitation practices from each other. Thus, piloting with these particular population will give our team generalizable feedback that would be essential in the possible expansion of the sanitation system in Kakuma or other refugee camps.

The project has 3 main sections:
1. Initial focus group discussions with Dinka and Somali refugees -- March
2. Six-weeks piloting the new complete sanitation with 32 households in the camp, using data from the focus group discussions -- May-July
3. Follow-up focus groups with users to understand satisfaction with the system -- July

NRC will be helping in the implementation of the pilot. As indicated above, this pilot will implement with 32 households. If the pilot demonstrates that system design is successful, it has the potential, with funding and resources, to be expanded to 50,000+ people in Kakuma and even other refugee camps in Sub-Saharan Africa.

FGD Procedures
The 12 initial focus groups were divided by language and gender resulting in 3 groups of Dinka women, 3 groups of Dinka men, 3 groups of Somali women, and 3 groups of Somali men. Each discussion held 6-10 participates.

Recruitment and Informed Consent
Recruitment for the focus group discussions was conducted by community (block) leaders with guidance by NRC and Sanivation. Recruitment followed a purposive sampling method centered around language and gender and required that participants be 18 years or older. Ten people were recruited for each group. The individuals were told when to gather in groups of same language and gender to hear the details of the project, give informed consent, fill out a brief demographic questionnaire, and be scheduled for a specific group slot—all with the help of local translators. The day of the focus group discussion participants were provided with transportation, water, and a bar of soap to serve as an incentive for participation.

FGD Process:
The moderator used a moderator guide translated into the discussion language, Dinka or Somali, to guide the discussion. Before the discussion began, the moderator explain ground rules such as respecting other’s views, and the participants right to leave at any time. Participants were told that the discussion would be audio recorded but that the recording would remain confidential. The moderator also explained that participants had the option of using the prepared name badges with common names as a pysudonem for the discussion, or writing in their own names if they were more comfortable that way. Each discussion lasted between 1 and 2 hours. Following the discussions the recordings are to be translated and transcribed for analysis.

Sample questions from the moderator’s guide:
1) Would you prefer to have a toilet located inside your home, just outside your home, or 20 meters away? [Possible probe: Explain your preference]

2) What was your impression of the first picture? [after being shown an image of a toilet design]
a. Was it obvious how you would use it?
b. What do you like about this design?
c. What would you change?

3) How would you feel cooking with a fuel made from treated human waste?

I hope you find this useful and we encourage other groups to conduct in-depth market research and use the human centered design process. As we progress, we will be posting more lessons from our research and recommendations for future human-centered design projects for sanitation and refugees. Finally, IDEO also has a great toolkit which is available here: www.ideo.com/work/human-centered-design-toolkit/.

Hi all,

We've been very busy working on toilets, treatment, and transformation. Thank you to everyone who has given us inputs along the way.

We're very exited about our upcoming pilot in Kakuma refugee camp where we get an opportunity to pilot an entire alternative sanitation system in the camp from mobile toilets, to solar treatment, and our re-use method of making briquettes from treated human waste. With support we plan on scaling our work and bio-briquette production in Naivsaha.

Below is a link to our recent newsletter that goes into more detail (including pretty pictures!) if you're interested. Thanks for all your support!

us4.campaign-archive2.com/?u=d78bf2d0118...bcb21b&id=b7b2e12730

Pictures such as these two:

I had seen a picture of a solar drier constructed by NGO FIN in the coastal areas of southern Indian in their presentation (see slide 7) www.friend-in-need.org/__media/friend-in-need-may-2009.pdf

Contact: Shyama Ramani (see email address in the presentation) for more information.

Yesterday I visited such treatment plant at Meherpur Pourashava, Meherpur, Bangladesh. There was no research, but as a rumor of poisonous bacteria, the plant is going to stopped. The farmers do not use it as manure.

Hello Andrew,

Thanks for your reply. I am interested in the fecal sludge drying by using conventional sludge drying bed with solar water heating system. The concept is that I will use hot water pipe in the sludge drying bed not only to help the evaoporation rate increased( as it is the main operation step in sludge drying bed) but also to disinfect the sludge with the temperature (around 50-60C).

I would like to set up lab scale experimental unit(drying bed with heated water pipe) in my university, and try to analyze the finished biosolid. But the problem what I am thinking is how I can get the solar water heater- I have to design myself or I have to buy (it will be costly for my experiment). I am not quite sure, if the energy department from our university could provide me with that solar collecter because our univeristy was flooded last year.

I would wonder if you could suggest me on this idea. Looking forward to you!

Best regards,
Su Su

Ah fair enough, I fell into that one didn't I?

I think the idea is reasonable, although I'm wondering if it would require too much equipment and knowledge to operate on a microscale, and I don't know enough about solar cooking to know how big these things can be constructed.

On a mesoscale, it would be nice to have some kind of continuous belt system avoiding a lot of handling. I am no engineer though, I'm not sure how that could be constructed.

Incidentally, my view is that you ideally want to mechanically dry sludge rather than trying to airdry it as you'll lose a lot of the nitrogen during the first few hours, particularly when it is really wet.

Hi all,

Thank you for all the replies and suggestions! Su Su, I have not come across anything else besides what I have listed earlier in the thread. We are working on finalizing one our first papers. Also Joeturner, Sanivation seems cool, although I am a bit biased since I am one of the cofounders.

Scalability is a great question and is something we are working through right now with a small pilot of our technology in Kenya. Drying beds also seem very attractive, particularly with the approach that www.waste-enterprisers.com/our-businesses/ is taking.

SuSu, I'd love to help advise you on your project. Is there a specific question you are looking to tackle or a resource I can help you find? The more research done on solar sanitation the better!

all the best,
Andrew