In April 2014, SNV and the UK Department for International Development (DFID) officially launched the multi-country Sustainable Sanitation & Hygiene for All (SSH4A) Results Programme, which will be conducted across nine countries in Africa and Asia from 2014 to 2018. DFID awarded SNV with a €28 million service contract to fund the SSH4A Results Programme in nine countries: Ethiopia, Ghana, Kenya, Mozambique, Nepal, South Sudan, Tanzania, Uganda and Zambia. It aims to improve the lives of around 2 million people with better sanitation. SNV approached Akvo to support it in the SSH4A programme, through various country training exercises and technical support in the use of Akvo FLOW, first during the “baseline” – setting the initial bar against which progress would be measured – and then the first mid-term monitoring household survey data collection exercises.
Akvo has just finished conducting training exercises in nine countries, within a timespan of two weeks. Here I explain what we did and some of the challenges each country team faced, and what the results are so far in terms of the various data collection efforts.
Caption: Screenshot of map within SSH4A Akvo FLOW Dashboard showing data collected in the nine countries.
SNV was on a rather tight schedule, with DFID expecting the report on the baseline to be done by August 2014. So we had to finish all of the training exercises by June, with data collection done in time by July for the subsequent data analysis. With locations spread around the world, a concerted effort needed to be made between the various SNV country teams and Akvo’s hub offices. Akvo’s Asia hub supported the training in Nepal (and indirectly also Bhutan), Akvo’s West Africa Hub supported training in Ghana, and the remaining seven countries in Eastern and Southern Africa were handled by Akvo’s East Africa hub. That meant careful planning of logistics was needed to ensure a successful execution of the activities. Two Akvo staff members were responsible for the facilitation of each of the country training sessions.
Training exercises Week 1 (9-13 June): Kenya, Tanzania, Uganda, and Nepal
The very first training for the SSH4A program was already held on the 27th of May in Lusaka, Zambia. Francis ((@fnwarui)) and Tim ((@timbjanssen)) were already in Lusaka for Akvo FLOW training with SNV in relation to the waterpoint mapping exercise taking place there. We had decided to conduct a Trainer of Trainer (ToT) session with the local consultant SNV had hired to take care of the data collection exercise for the SSH4A program. That proved to not have been very successful, which I will elaborate on later.
In Nairobi I had conducted another ToT training for SNV staff that were working within the Kenyan SSH4A team, key M&E staff responsible for analysing all of the data globally, and the supervisors with some enumerators for the data collection effort in Kenya. The ToT training was held on 11th and 12th June, followed by Akvo FLOW Dashboard training on the 13th. Lissy ((@lissyvn)) joined me on the 12th to help me out. All together we had a group of 17 people in Nairobi. Here’s the Flickr set of all the people that attended.
The SSH4A training for Tanzania took place in Mwanza on Monday the 9th of June. It took place at the local SNV Mwanza office, and was facilitated by Francis and Elma (@elmadentoom). The first day was spent mainly on going through the household survey and getting the translations done into Kiswahili. The second day the training was held involving a total of 16 enumerators, followed by a field visit in Mwanza. The enumerators each practiced on three households. In the evening a brief Akvo FLOW Dashboard training session was given to four key SNV staff and four enumerator supervisors. The last day was used for a final Q&A session. The team in Tanzania had also purchased solar chargers as backups. Here’s a Flickr set of the people trained. It went smoothly – no major issues arose.
The Akvo FLOW training for the Uganda SSH4A country team took place on the 11th and 12th of June in Fort Portal, Uganda. The training was facilitated by Andrew (@andrewmolo) and Tim (@timbjanssen). Due to the remote location of Fort Portal and the limited amount of time available, they were forced to do the training in one and a half days. Although we had communicated that no more than 25 people were to be invited, 44 participants showed up for the training. There was also a barely working internet connection. With all of these elements combined it made this probably the most challenging training workshop in the whole SSH4A program. We also found we had to provide much more support to the Uganda team during the data collection. The attendants Flickr set of the training held in Fort Portal can be seen here.
The SSH4A training in Nepal was held between 11th and 13th June, and was attended by 21 participants from the SNV regional team, and an external consultancy agency called RECID. Various pictures from the training can be viewed here. The training was facilitated by Joy (@joycarpediem) and Isha (@ishaparihar). Some issues did arise during this training exercise. Firstly, there was insufficient training time. A ToT training needs adequate time to enable the participants with the basic knowledge/understanding of the FLOW tool. In this case, the speed of the trainings slowed down significantly due to low and poor internet connection. The team had lost a lot of time on Day 1 of the training, which we attempted to make up by rushing through the remaining days. That meant that for the SSH4A Nepal team, a lot of extra support was needed post-training during the data collection process. The second major issue related to managing translations and unicodes. It’s essential to get the translations inputed into the FLOW system prior to the training. However, partners were having difficulty understanding how to get translations in Unicode format. The Lesson learned was that we need to come up with basic and easy-to-follow instructions related to unicode/translations. The last important issue is linked to the finalisation of the survey. This was an issue experienced by all country teams and in all workshops being held. The survey had been finalised by the SNV global team just a week before the initial training exercises were due to take place. That meant that a lot of queries and corrections were received during the training exercises and during the data collection efforts.
Joy Ghosh preparing for a field exercise with the Nepal team. Photo credit: Isha Parihar.
Training exercises Week 2 (16-20 June): South Sudan, Ghana, Mozambique, and Ethiopia.
The Akvo FLOW training for the SSH4A program in South Sudan started on Monday the 16th June and took place in a hotel in Juba. Elma and Andrew facilitated this one. The first day was spent going through the survey with the 18 enumerators, two SNV staff and one external consultant. A few changes were made to the survey and it was decided to not make any translations. The reasoning was that there are just too many dialects in South Sudan. All of the 18 enumerators that were trained are regional government officials. On the second day, the Akvo FLOW training took place. There was no field visit scheduled due to the volatile situation within the country. Instead the enumerators practiced outside the hotel. On the last day the dashboard training was given. The training went well and no major issues worth mentioning. The attendants Flickr set can be viewed here. This is actually the very first time that Akvo FLOW is being used to collect data in South Sudan.
Data collectors practising with Akvo FLOW in Ghana Photo credit: Giel Hendriks.
The training workshop for the SSH4A program in Ghana was held between 17th and 20th June 2014 in Tamale. From our West Africa Hub we had Emeline (@EmelineWASH) and Giel (@GAHenriks) who facilitated the training. In total they had 33 participants, with 26 of them being enumerators and seven SNV staff. The attendants Flickr set can be seen here. All the participants were very enthusiastic. During the training the different groups immediately started to translate the survey in various local languages. Some issues were encountered again in relation to the survey and setting up translations. Lots of comments were collected on the survey, and therefore also lots of improvements needed to be made. The SNV team in Ghana had decided to add six translations of local languages. That meant extra effort was needed to get these setup in the FLOW system before the actual data collection. This all worked out eventually, just in time.
The Akvo FLOW training exercise for the SSH4A program in Mozambique was held in Nampula between 16th and 19th of June. The training was originally meant to be given by both Tim (@timbjanssen) and Phylis (@phyliswebi), but immigration officials at Nampula airport had sent Phylis back on the same flight to Nairobi for unknown reasons. Tim had to therefore do this one on his own. In total 18 people were trained, and you can see here who attended. The phones arrived on the morning of the training, so some delays were encountered in having to set them all up. The training went smoothly after that. The group consisted of university students who, even though they only understood Portuguese, were quick to get to grips with the FLOW system.
All people trained in use of Akvo FLOW in Nampula, Mozambique. Photo credit: Tim Janssen.
The Akvo FLOW training for the SSH4A programme in Ethiopia was held in Bahir Dar on 17th and 18th of June. This training was facilitated by Francis and Lissy. In total 18 people attended the training, and you can see the pictures of attendees here. Fortunately all the Amharic translations were entered in time into the system. All enumerators and SNV staff were trained successfully with no real issues encountered.
Data collection results:
As mentioned earlier in this post, data collection started in all the nine countries very soon after the training workshops were finalised. On average data collection took between two to three weeks depending on the sample size and the distance needed for travelling. All of the data was therefore successfully collected in the originally perceived time schedule, except for in Zambia. Apparently, the consultant they had hired to undertake the work did a very poor job. To be sure that the data quality was of high standards they opted to start over completely. That process is on-going as we speak.
The total amount of household surveys collected per country are as follows: Kenya (2,117), Uganda (3,326), Tanzania (2,179), Ethiopia (2,186), Mozambique (2,139), South Sudan (2,139), Zambia (1,096, still on-going), Ghana (2,113), and Nepal (2,740). As this is household data, it’s not publicly available to showcase.
Luuk Diphoorn leads Akvo’s East Africa hub, based in Nairobi.