EDI Ltd.pdf

EDI Limited

Expression of Interest to

Partner In The MCC-PEPFAR

Tanzania Data Collaborative (Co-Lab) Project

#MCC-16-RFA-0002

December 2015

Head Office

PO Box 393

Bukoba, Kagera, Tanzania

Tel: +255 (0)28 2220059 (General line)

+255 (0)28 2220078 (Finance)

Contact Person:

Dr Matt Wiseman, Operations Manager

E: m.wiseman@edi-africa.com

EDI Expression of Interest

1. INTRODUCTION

EDI Limited is submitting this Expression of Interest in response to the Millennium Challenge Corporation’s call for

Technical Assistance for the MCC-PEPFAR Tanzania Data Collaborative (Co-Lab) Project. EDI Limited is interested in joining consortia bidding for this exciting opportunity.

As a leading Tanzania based Research Company1 with significant expertise in the collection, processing and cleaning of large datasets combined with substantial experience in capacity building of data users and a strong track record in the analysis and dissemination of survey findings, EDI believes it is ideally placed to support the set up and implementation of the Co-Lab.

The following document provides an overview of our technical capabilities and includes:

.. An Executive Summary describing how EDI envisages contributing to the Co-Lab;

.. An introduction to EDI and Surveybe, highlighting the key skills and experiences of our team;

.. Examples of work undertaken by EDI’s Research Team; and, .. Examples of Publications prepared by the EDI team.

Any organisations interested in working with EDI Limited are encouraged to contact Dr Matt Wiseman

(m.wiseman@edi-africa.com) and Dr Johanna Choumert Nkolo (j.choumert.nkolo@surveybe.com) to initiate discussions.

1EDI is ranked as the number one private research institution in Tanzania - https://ideas.repec.org/top/top.africa.html mailto:m.wiseman@edi-africa.com mailto:j.choumert.nkolo@surveybe.com https://ideas.repec.org/top/top.africa.html

2. EXECUTIVE SUMMARY

EDI believes it is ideally placed to provide data related support to consortia bidding for the Tanzania Data Collaborative

(Co-Lab). In particular, EDI believes it can add value in the following areas:

Local Research Knowledge –With over 12 years research experience in Tanzania, EDI has an in-depth knowledge of the research landscape in Tanzania including a strong understanding of Tanzania’s current research priorities, significant experience of relevant legislation, research clearances and ethical approval requirements and of the practicalities of collecting data and accessing data sets;

Broad-based Partnership -Having conducted over 40 large scale research studies in Tanzania, EDI has built strong relationships with development partners including MCC, World Bank, IFPRI, Twaweza, IRRI, CHAI, DFID and CGD, as well as a wide range of academic institutions including Harvard, MIT, Oxford, Columbia and AIID.

EDI hopes to leverage our research networks to strengthen the effectiveness of the Co-Lab and help Co-Lab access a broader range of datasets;

Strong Cross Sector Knowledge–Having interviewed over 270,000 respondents and created datasets for a wide range of research programmes, EDI is able to bring to the table a research team with experience of data collection and analysis in multiple sectors including Health, HIV, Education, Agriculture, Infrastructure (Roads, Water, Electricity), Natural Resources, Social Inclusion, Gender and Governance;

Capacity Building of Data Users – EDI is able to offer a wide range of tailored capacity building programmes to support individual and organisational users of the Co-Lab. EDI’s team has substantial training experience for trainees at all backgrounds and education levels from grass-roots organisations to university level. Examples of the short courses that EDI could develop and deliver include:

o Basics of Data Collection –Giving data users context as to how data sets are collected in Tanzania covering research and ethical approvals, practicalities of field work in Tanzania, methods of data collection (CAPI, PAPI, CATI, CAWI, etc);

o Survey Design – To help users understand datasets this course would explore how surveys are designed from the development of high quality questionnaires, sampling methods, data labelling, data structure, meta/para data, administrative units, listing, replacements, interview best practice and overview of data management and cleaning protocols;

o Questionnaire Design – A more in-depth course on how questionnaires (Households, Schools, Health

Facilities, etc) are designed in the Tanzanian context;

o Data Cleaning – Available datasets in Tanzania are likely to be of varying quality, this course will look at how to assess the quality of a dataset and practical approaches to data cleaning;

o Software Training – Making the best use of datasets implies some degree of IT knowledge. EDI is able to provide introductory training courses in a number of software applications including STATA and

Surveybe;

o Basics of Data Analysis – Introductory course to the analysis of survey data and basic statistical principles;

o Advanced Data Analysis

Review and Cleaning of Datasets: As noted above, it is likely that the quality of the available datasets will vary somewhat. To ensure that Co-Lab Users have access to the highest quality data:

o EDI’s Researchers and Data Officers would review all datasets and supporting documents held by Co-

Lab, and speak to the data owners to establish and document the quality of each data sets, noting any known issues;

o Where issues are found EDI’s team, where possible, would reprocess data using STATA to conduct additional cleaning protocols;

o EDI researchers would also look to homogenise current data sets into a single standard format.

Data Analysis and Dissemination –EDI’s Research team have strong experience in the analysis of large datasets and the preparation of Data analysis and report writing using existing data sets of clear and accessible reports.

Driven by Co-Lab User demand, EDI researchers would:

o Conduct meta-analysis and systematic reviews of existing datasets on specific topic studies in Tanzania;

o Prepare user-friendly reports in English or Swahili for dissemination;

o Conducted dissemination workshops presenting research results to non-technical audiences as well as technical audiences.

Technological Innovation – EDI has been a leading innovator in the CAPI (Computer Assisted Personal Interview) software market, resulting in the development of the surveybe (www.surveybe.com) application. EDI’s UK based software programming team would be able to provide its expertise to the consortium for example, dealing effectively with data privacy, data security and encryption, linkage to other software applications ('plug-ins') to assist in better compatibility across software in difference phases of research cycle and overcoming offline restrictions.

Data Collection – Whilst not core to the Co-Lab project, EDI envisages that increased usage of data in Tanzania will result in user driven demand for additional data to fill gaps in existing data sets. Given our background, EDI will be ideally placed to assist in filling these gaps. EDI would be able to work with Co-Lab users to define demand, identify alternative sources of data and if the required data is not available then work with Co-Lab to develop proposals and budgets to raise funding to conduct fresh data collection.

http://www.surveybe.com/

3. INTRODUCTION TO EDI

EDI Limited2 is a Tanzania-based, but internationally focussed research firm. Since our foundation in 2002, EDI has grown to become Tanzania’s most respected research house and is now ranked as the top private research institution in Africa3.

EDI was borne out of the desire to provide assistance towards economic and social development.EDI has a strong reputation both within East Africa and globally for the design and implementation of a wide range of surveys providing our clients with high quality datasets.

We provide expert advice and support to a range of Tanzanian and International

Organisations in the Public and Civil Society Sectors including (but not limited to):

The World Bank, IFPRI, IRRI, ILRI, Millennium Challenge Corporation, Millennium

Challenge Account-Tanzania, Centre for Global Development, Columbia

University, Clinton Health Access Initiatives, Uganda Bureau of Statistics, DFID, DANIDA, Tanzania Social Action Fund, Rockwool Foundation, Prime Minister’s

Office (PMO-RALG), Harvard University, University of Oxford, University of

Copenhagen, University of Ghent, MIT-Zaragoza International Logistics

Programme, ICAP – Columbia University, Centre for Disease Control, Twaweza and the Amsterdam Institute for International Development. Surveybe has been deployed on over 1 million data collection instances.

Our sister company Economic Development Initiatives (EDI) Ltd4, is a UK based developer of our SurveybeTM, Computer Assisted Personal Interview (CAPI) software which is used by data collection firms in over 50 countries around the world by leading academic and research institutions including (but not limited to)

Stanford, Tsukuba, Passau, Oxford and Rhode Island Universities and ILRI, IFPRI, IRRI and IITA.

EDI’s Team has substantial experience covering the complete cycle of data collection: sampling, listing, design of questionnaires, configuration of questionnaires under Computer-assisted personal interviewing (CAPI), mobile phone surveys, field logistics, training of interviewers, quantitative data collection, qualitative data collection, data cleaning, data analysis, report writing, dissemination through blog posts, academic papers, conferences, seminars, workshops. In particular, EDI has:

In-depth knowledge of the research landscape in Tanzania including a strong understanding of Tanzania’s current research priorities, significant experience of relevant legislation, research clearances and ethical approval requirements and of the practicalities of collecting data and accessing data sets.

Over 12 years’ experience in the design and implementation and the production of high quality, clean, analysis ready datasets for over 40 large scale surveys in Tanzania including Household, Community, Agricultural, Health, Schools and Enterprise Surveys including interviews of close to 300,000 respondents, including:

o Experiments to Improve Learning Outcomes in Tanzania, TWAWEZA (2013-15);

2Registered in Tanzania, Registration Number 43600 3http://ideas.repec.org/top/top.africa.html 4Registered in the United Kingdom, Registration Number 06314477

Our Vision:

To create a positive impact on people's lives through the provision of world class research services, tools and innovations that enable better decision making from global to household level.

Our Mission:

To be the leading research company focussed on the provision of world class services, tools and innovations that enable the collection, analysis and use of high quality data. We aim to facilitate more accurate, informed and timely decisions to be made that can create a positive impact on people's lives around the world. We strive to push the frontiers of information provision by combining academic rigour, commercial acumen and flexible cutting-edge design and technology utilising methodologies that ensure value for money for our clients. The cornerstone of our work is our continued combination of international best practice with regional and local expertise. In everything we do we seek to provide opportunities for growth and enrichment for our employees, shareholders, clients and the communities in which we operate.

o Public Opinion Study of Natural Resource Management Revenues in Tanzania, Centre for Global

Development (2014);

o Geographically Concentrated Multi-Level HIV Prevention in Bukoba Urban District: Outcome Evaluation of a Combination Prevention Program for ICAP, Columbia University (2013-16);

o Improving Institutions for Pro-Poor Growth, University of Oxford (2012) o TASAF Community Based Conditional Cash Transfer Impact Evaluation, TASAF/World Bank (2008-2012);

and, o Baseline Data Collection for the Water Sector Impact Evaluation, Millennium Challenge Account –

Tanzania, 2013.

A proven quality focussed data management team with strong experience of processing, cleaning and managing both small and large data sets, using best international practices.

o Well established systems and procedures to ensure high quality data collection:

o Rigorous data confidentiality and privacy protocols in line with international standards;

o Tried and tested data quality control procedures including both automated validation checks in surveybe, as well as a series of manual checks, revisits, interview observation and feedback procedures;

o Expertise in surveybe™ data collection software, as well as CSPro, SPSS and Stata data analysis software;

and, o Proven training modules for Supervisors, Interviewers, Data Processing and other survey staff.

In house capacity to provide capacity building and training on international best practices by both our Tanzanian and International staff. EDI has trained hundreds of interviewers in interview techniques and best practice, anthropometric measurements, use of CAPI and quality control, as well as in survey specific skills and knowledge. In addition, EDI has staff members with senior teaching, training and lifelong learning experience and experience in curricula development in general and in Tanzania;

Experienced staff with a strong track record in proven experience of statistical analysis (statistics, econometrics, micro-econometrics, spatial econometrics, meta-analysis techniques) and in the preparation of clear and accessible reports. We have conducted all types of statistical analysis typically required in an impact evaluation, such as dif-in-dif regressions, propensity score matching and the like. EDI’s team have published in top field journals as well as preparing publications to non-technical audiences. Examples of publications can be found in

0 and at http://edi-global.com/publications/.

Experience in the designed and implementation of dissemination programmes. Examples include the the production of 28 district level CWIQ reports and the dissemination of the Core Welfare Indicator Questionnaire

(CWIQ) results to District Councils to promotes evidence-based policy debate and policy formulation, provide tools for district level M&E; and, increase accountability of LGA to citizens. This exercise involved the dissemination of the results of CWIQ to stakeholders in each district (councillors, district officials, NGOs, CBOs, Advocacy Groups, MPs, ‘interested citizens’, etc) through the facilitation of participative stakeholder workshops and the distribution of colourful, easy to read and informative posters and research briefs;

A proven track record in collecting data using electronic data collection enabling EDI to deliver high quality digitised data within a short time, usually within a week of collection in the field, with:

o The most experience in electronic data collection of any firm in Tanzania, having first commenced using

CAPI in 2007. EDI has since developed in-house two full electronic data collection software packages, http://edi-global.com/publications/

CWEST™ and surveybe™ (www.surveybe.com). Surveybe is used around the world by organisations such as the World Bank, IADB, ILRI, IFPRI, IRRI, Young Lives, as well as central statistics agencies in Uganda, Ethiopia, Bolivia and Georgia;

o A team of electronic data collection experts who not only have significant experience of designing and implementing CAPI surveys, but also played an integral role in the development of surveybe™ and therefore have an in-depth understanding of how to get the most out of our software;

Considerable experience in the collection and analysis of a wide range of supplementary survey data, including anthropometrics, GPS, Distance (pedometer), rainfall, water quality and biometric data.

Further information on EDI and the work it undertakes can be found at www.edi-global.com.

3.1. OUR RESEARCH

EDI has substantial experience in the design and implementation of surveys to provide data to understand poverty in all its dimensions and assist in formulating policies to fight it. At EDI academic rigor and policy relevance complement each other. From survey design, over respondent sampling, interviewing techniques and quality control, up to data analysis and documentation, EDI has collaborated with the world’s finest survey experts to established powerful best-practice techniques.

We understand that data collected locally, needs to be reported on globally. The uniqueness of EDI lies in its ability to bridge the gap between the local and the global. While EDI is a Tanzanian registered company, with excellent knowledge of the realities on the ground, its data abides to international quality norms. For example, data collected by EDI has been officially included in the World Bank’s LSMS (Living Standard Measurement Survey) list of data sets.

EDI’s website plays host to the Kagera Health Development Survey (KHDS) dataset which we believe is Africa’s longest running panel survey.

At EDI, by using a variety of quantitative and qualitative tools, we triangulate information and analyse it in light of the fieldwork and the local context. We are committed to providing strong analysis through statistical figures and we enrich it with qualitative information and our excellent understanding of the local context. We understand that poverty is multidimensional concept. We have an excellent knowledge on debates regarding its measurement, while we can provide simple and relevant poverty indexes. For instance, using principal component analysis we can provide an asset-based poverty profile of households which is in line with international standards and comprehensive for policy makers.

Surveys and research at EDI, while motivated from policy significance, are of sufficient quality to be published in top peer-reviewed journals. EDI collaborates with acknowledged researchers to achieve this excellence. Recent journals in which work from EDI has been published include: Review of Economics and Statistics, International Journal of

Epidemiology, AIDS, Journal of Development Studies, Demography and Journal of Development Economics. Our knowledge of multiple regression techniques and objectives allows us to lay the first stone of data analysis, by providing key statistics on context-relevant variables.

http://www.edi-global.com/

3.2. SURVEYBE

Surveybe is a new generation CAPI software that gives

Survey Managers’ direct control of survey design, accurate responses that have been validated and cleaned at the point of collection, time saving efficiencies in all phases of their projects and real-time, analysis ready, data. Surveybe has been designed by a team with a blend of backgrounds in survey management, economics and commerce along with expert software developers. Creating an easy to use, comprehensive and affordable software suite that can intuitively and simply handle all manner of complexities has been the driving ethos of our work.

The Surveybe Designer has an intuitive and easy to use drag-and-drop interface for users to design and build their own questionnaires. Surveybe allows users to design and implement both, simple, linear face-to-face surveys as well as large scale complex longitudinal or panel surveys. The resulting product allows survey managers to fully embrace the undoubted benefits of CAPI over PAPI. EDI has supported many multi-topic household panel surveys in Sub-Saharan Africa and increasingly across the globe, with users in over 50 countries, in using electronic methods for collecting data. For more detail about the functionality of surveybe and the services we are able to offer to our clients,visit www.surveybe.com.

4. RESEARCH TRACK RECORD

The following table provides examples of the research projects that EDI and its team have implemented:

Timing Client and Country of Assignment Project Description

ITAD Limited on behalf of DFID

Impact Evaluation of Improving Rural Access in Tanzania in Five Districts:

Combining quantitative and qualitative methods this study assessed changes in the socio economic characteristics of households affected by rural road improvements, as well as changes in access to selected health, education and community services.

2013 -

TWAWEZA

Experiments to Improve Learning Outcomes in Tanzania – Implementation and monitoring of interventions aimed to improve the channelling of resources to public primary schools for quality improvement of public education, and the restricting of incentives to improve learning outcomes of primary school students. The study involves a randomized control trial of Public Primary Schools in 10 Tanzania districts to evaluate and understand the effects of project interventions and included annual baseline, midline and endline school and household surveys to collect school and teacher information as well as student and household information from randomly selected students from standard 1 to 3.

Mathematica/

Millennium

Challenge

Corporation

Evaluation of Components of the MCC Energy Sector Project in Tanzania. The

Millennium Challenge Corporation (MCC) aims to improve living standards and reduce poverty through economic growth. The Government of Tanzania received a grant from MCC to advance economic growth and reduce poverty through investments in energy infrastructure. The short-term objectives of this project are to improve the quality and increase the quantity of electricity to enterprises and households in order to increase investment potential. The project components to be evaluated include the following: Transmission and distribution systems rehabilitation and extension activity (T&D activity); The financing scheme initiative

(FS initiative) and a Kigoma Solar initiative.

The T&D and FS initiatives were rolled out in 7 study regions: Dodoma, Iringa, Kigoma, Mbeya, Morogoro, Mwanza and Tange, while the Kigoma Solar Activity was exclusive to Kigoma. EDI is conducting a follow-up survey of 11,000 households in

7 regions to assess the impact of this project.

2014-

Centre for Global

Development

Public Opinion Study of Natural Resource Management Revenues in Tanzania - In

2010, Tanzania discovered natural gas reserves off its southern coast worth roughly fifteen times its annual GDP. This project assesses the ability of Tanzanian voters to formulate coherent views on a range of options for the use of gas revenues – including a sovereign wealth fund, revenue sharing with regional governments, and direct distribution in the form of cash transfers. The sample is 2000 nationally-representative respondents in 20 districts in Tanzania, who are divided into control and treatment groups. The intervention consists of a “deliberative polling” exercise to be held in Dar es Salaam in April 2015. This will be a series of public deliberations with a nationally representative sample of ordinary Tanzanians, where respondents will be exposed to expert information and public debate. Baseline surveys were conducted prior to the deliberative poll for both the control and treatment group.

Two rounds of follow-up phone surveys will be conducted after the poll to assess

Timing Client and Country of Assignment Project Description whether respondents have changed their opinions and their level of public engagement.

2011-

IFPRI – International

Food Policy Research

Institute

Guatemala

Malnutrition in Children Under Two: This 3.5 year study in Guatemala follows pregnant women through pregnancy, birth and then the addition of new children into mother/child pairings over the first 24 months of a child’s life. The survey involves regular interviewing and extensive anthropometric measurements. The surveybe team has provided configured survey tools translated into Spanish as well as support and training to the survey team in the use of surveybe software, data structure and field management. IFPRI will be configuring their own questionnaires for the 3 month follow-up survey onwards.

Harvard University/

Results for

Development

Institute, Tanga, Tanzania

Transparency for Development Project – This project aims to generate actionable evidence for practitioners, researchers, and other stakeholders working to improve health, accountability, and citizen participation. The randomized control trial will collect data and inform average citizens in selected villages about maternal and new-born health problems in their communities, and to empower communities to take action change health outcomes.

2013-

Amsterdam Institute for International

KNCU Community Health Insurance Fund Survey - The Health Insurance Fund (HIF) and its implementing partner PharmAccess facilitate access to comprehensive health care by making health insurance affordable to low and middle-income groups in Africa. The program aims to alleviate poverty among households, by building a health insurance financing and delivery system that functions effectively and efficiently over a long-term period. For more information see www.hifund.org.

TWAWEZA

Uwezo Follow Up Survey – Twaweza is a ten year citizen-centered initiative

(www.twaweza.org), focusing on large-scale change in East Africa that seeks to foster conditions and expand opportunities through which citizens can get information and make change happen in their own communities directly and by holding government to account. As part of the evaluation of Twaweza’s first strategic period (2009-2014) and in collaboration with the Amsterdam Institute of

International Development (AIID), Twaweza conducted a nationally-representative survey of 2,500 households throughout Tanzania. The survey collected data on an extensive set of indicators related to citizen agency and government responsiveness particularly in the education, health and water sectors (the highlights of the findings can be found here http://www.twaweza.org/uploads/files/AIIDBaseline-small.pdf).

World Bank

Measuring Household Labor in Smallholder Farming - In Sub-Saharan African farms, family labour is the major farm input and having a good measure of it is critical. This project assessed the degree of measurement error in reporting household labour inputs in farming and the analytical implications of this error. The methodological study collected information about labour for every member of farm households in a sample of villages in Tanzania during the 2013/2014 agricultural season. The experiment focused on three alternative survey designs: Control Group

(C): Standard agricultural labour module, with labour reported in the aggregate by recall for the entire season (End-line Survey). Treatment Group 1 (T1): Baseline

Survey and phone surveys for labour module for the duration of the main season

(Mobile phone survey). End-line Survey fielded to capture completed agricultural http://www.hifund.org/ http://www.twaweza.org/ http://www.twaweza.org/uploads/files/AIIDBaseline-small.pdf

Timing Client and Country of Assignment Project Description season yields and other inputs. Treatment Group 2 (T2): Baseline Survey and intensive interview labour module during the duration of the main season (Time

Use survey). End-line Survey fielded to capture completed agricultural season yields and other inputs. GPS location data was collected for all households and in addition, GPS data were used to calculate the area of farm plots.

IFPRI

Tanzania Africa RISING Baseline Evaluation Survey (TARBES) - As part of the US government’s Feed the Future initiative that aims to address global hunger and food security issues in sub-Saharan Africa, the US Agency for International Development is supporting a multi-stakeholder agricultural research projects under Africa

Research In Sustainable Intensification for the Next Generation (Africa RISING - AR) program. The overall aim of the program is to transform agricultural systems through sustainable intensification projects in three regions of Africa. In this project, EDI collected baseline household and community data in Babati, Kongwa, and Kiteto districts in Tanzania for the purpose of facilitating M&E of Africa Rising-Tanzania project. Baseline geo-coded household data was collected from ‘baby’ farmers in the intervention villages, with household survey data also collected from counterfactual villages. In addition to the household survey, a community survey was conducted using a short structured community questionnaire from key informants and/or village leaders.

2013-14

Columbia

University/ICAP

Geographically Concentrated Multi-Level HIV Prevention in Bukoba Urban

District: Outcome Evaluation of a Combination Prevention Program for ICAP. In this survey, a joint EDI/ICAP team interviewed over 5000 respondents in 2,500

Households. In addition, ICAP medical staff tested conducted rapid HIV tests on all consenting respondents. A cohort of over 380 respondents who were newly diagnosed as being HIV positive were then re-interviewed within two weeks of the diagnosis and again 6 months later to track their experiences.

2014-

Samuel Hall

Consulting - AFD –

French Agency for

Development

External Consultancy to Baseline for the impact evaluation of the Central Highlands rural development program in Bamyan and Wardak Provinces in Afghanistan Program operated by a consortium of three NGOs and funded by the AFD; €10 million in funding.

- Field visit conducted to Bamyan province to meet local communities and stakeholders of the project.

- Designed an evaluation plan in a challenging and demanding environment.

Prepared the tools (household survey, market prices survey and focus group discussions).

- Built and analyzed poverty and vulnerability profiles at the household level using principal component analysis (2000 households). Performed statistical analysis on the data collected.

- Wrote a report on poverty and vulnerability for the baseline.

Samuel Hall

Consulting - The

World Bank

External Consultancy to Community Perception Survey of Mining in Afghanistan

- External consultant (team of three) for Samuel Hall Consulting, Kabul (Afghanistan)

- Conducted fieldwork in the Kabul area to meet mine workers and communities living near mines and prepared the tools (household survey, market price survey and focus group discussions).

Timing Client and Country of Assignment Project Description

- Analyzed the perception of mining activities by local communities in Afghanistan.

- Performed statistical and econometric analysis of data (1000 households in five provinces) and wrote a report with recommendations for benefit-sharing schemes.

2008 -

World Bank/IFPRI/

Tanzania Social

Action Fund (TASAF)

TASAF Community Based Conditional Cash Transfer Impact Evaluation – Follow

Up Survey (CCT3) - Second Follow up Impact Evaluation of the TASAF Conditional

Cash Transfer project. Conditional Cash Transfer have been disbursed to households assessed vulnerable by both the community and a survey instrument administered to all households in the village. Eligible (vulnerable) households needed to abide by certain conditions, such as children's school and clinic attendance in order to remain in the transfer programme. The impacts are identified by a staggered introduction of the programme in 40 out of the 80 villages.

The outcome of this project provides a basis for the government to decide whether or not to scale this programme up to national level.

2007 -

Tanzania Social

Action Fund (TASAF)

Impact Assessment of the Tanzania Social Action Fund

This 3 year panel survey, commissioned to EDI by the World Bank and TASAF, follows a treatment and comparison groups of vulnerable people, who participate in a community driven development project. Impacts will be identified through the staggered introduction of the project in 50 of the 100 sampled communities. This project involves a complicated listing exercise to stratify the sample within the village and oversample vulnerable groups.

Jan –

Sept

Millennium Challenge

Account Tanzania

Baseline Data Collection for Water Sector Impact Evaluation – Baseline Survey of

MCA Tanzania’s Water Sector interventions which aim to 1. Expanding the capacity of the Lower Ruvu water treatment plant serving the Dar es Salaam areas, from about 180 million litres per day (MLD) to about 270 MLD and 2. Improving water supply in Morogoro through rehabilitating a water treatment plant and improving water transfer in the existing distribution network. These interventions are expected to increase production from 18 MLD to 33 MLD. EDI is undertook a baseline survey as a part of an impact evaluation to measure the effects of the above two interventions at the utility and household level, in the treatment areas.

The survey involved a range of quantitative data collection (including listing, a 5000 household survey, follow up phone survey, service provider survey and water quality survey), as well as focus group discussions and semi-structured interviews.

Feb 2013

- July

IFPRI

Baseline and Endline Paralegal Survey - Evaluation of community-based legal aid programme initiated in Kagera by Mama's Hope Organization for Legal Assistance

(MHOLA). The intervention is intended to improve the delivery of legal services through 1) the provision of training paralegals at the community level and 2) providing material incentives to encourage the transfer of knowledge to all individuals in the community. The survey involved a panel household survey (1800 households sampled from 150 clusters); a community panel survey; and, a paralegal panel survey (150 paralegals - 1assigned per cluster).

2011-

CSA – Central

Statistics Agency;

Ethiopia

Living Standards Measurement Study – Integrated Surveys on Agriculture (LSMS-

ISA): Ethiopia Rural Socioeconomic Survey. Under the LSMS-ISA initiative, the

World Bank is supporting governments in 7 Sub-Saharan African countries to generate household panel data with a strong focus on agriculture and modelled on

Timing Client and Country of Assignment Project Description the integrated household survey design of the LSMS. Surveybe developed a CAPI instrument surveybe software and to provide training and support during various stages of the project.

2011-

ILRI – International

Livestock Research

Herd Histories in Ethiopia: ILRI in collaboration with various partners is pursuing a comprehensive research agenda aimed at designing, developing and implementing market mediated index-based insurance products to protect livestock keepers from drought related asset losses they face, particularly those in the drought prone Arid and Semi-Arid Lands (ASAL). ILRI have now partnered with surveybe for the provision of electronic data collection tools which have been designed to assess the herd size and lineage of livestock for farmers in Ethiopia, and noting the key events or shocks that have direct impact upon the herd population. Collared Sub-set, Baseline and Repeat surveys: Following the Herd Histories project ILRI quickly expanded their use of surveybe software and consultative services on three other projects for a twelve month period. These are on-going projects located in Kenya and Ethiopia, which utilise different languages including Amharic.

University of Oxford

Tanzania

Improving Institutions for Pro-Poor Growth. As part of a project with the World

Bank, CSAE ran a business ideas competition in Tanzania during 2011. This survey follows those young entrepreneurs to learn about their experiences since the competition, and to measure what effect any prize money from the competition may have had.

2010-

Geostat – Georgia

Statistical Bureau

Technical Assistance Intervention for the Consumer Price Index Surveys. The

National Statistics Service of Georgia (Geostat) is currently receiving support from the Focus on Results: Enhancing Capacity Across Sectors in Transition (FORECAST) organizational development programme. The goal is to support capacity building at

Geostat for its field work as well as improving the reliability of data and surveys produced. Surveybe is working with Geostat to assist it in improving its data quality through the use of electronic data capture. Surveybe has provided configured surveybe software, as well as, training and piloting support for the monthly Price

Statistics Survey. The survey serves as bases in the computation of Consumer Price

Index (CPI) and inflation rate.

World Bank

Tanzania

Promotion of Women’s Entrepreneurship: Tanzania Virtual Business Incubator -

End line impact evaluation of a World Bank project - Promotion of Women’s

Entrepreneurship: Tanzania Virtual Business Incubator. A total of 820 women are part of the evaluation design. The project targeted women who are: (i) aged between 18 and 50 years old; (ii) functionally literate and display basic technical and managerial skills; and, (iii) running their own business at the start of the project and willing to turn their enterprises into growth-oriented businesses. The sectors they engaged in were handicrafts/artwork, textile and tailoring, poultry, food processing, food vending, and soap processing.

2011-12

World Bank

Water And Sanitation Programme - Post-Intervention Data Collection And

Analysis For The Impact Evaluation In Tanzania

Development and implementations of a post-intervention data collection and analysis for the impact evaluation of the Scaling up Sanitation and Hand-washing projects in Tanzania. The follow-up household survey was implemented in the 10 districts included in the baseline study, with a sample size of 3620 households, in a

Timing Client and Country of Assignment Project Description total of 181 wards and 362 Sole Consultant villages. Primary participants were mothers and caretakers of children under five years old.

2010-

Clinton Health Access

Initiative and MIT-

Zaragoza

International

Logistics Program

Assessing the Impact of a Supplier Incentive on Coverage and Uptake of

Subsidized ACTs in Remote Retail Shops in Tanzania. Our research team is undertaking data collection for the monitoring and evaluation of the Affordable

Medicines Facility-malaria (AMFm) initiative. The survey involves:

Over a 21 month period conducting a total of 3,400 basic retail audits, 800 comprehensive retail audits, and 9,800 exit interviews at a sample of 200 Accredited Drug Dispensing Outlets (ADDOs) in Rukwa, Lindi, and Mtwara regions of Tanzania.

Conducting baseline, midline, and endline surveys in 1,000 households in the communities served by the ADDOs.

2010-11

Centre for the Study of African Economies, University of Oxford

Tanzania

Baseline and Follow Up Survey of Manufacturing Firms in Dar es Salaam – The survey was implemented together with a randomised experiment on the effects of social networks on firm outcomes.

2010 -

World Bank/Uganda

Bureau of Statistics

Uganda National Panel Survey – Uganda Bureau of Statistics Responsible for training 9 Supervisors, 36 interviewers and 4 data processing staff at the Uganda

Bureau of Statistics in using CAPI methods and in implementing the UNPS on CWEST software.

July 08 –

June 11

UK Department for

International

Development (DFID)

Implementation of the Research into Use Programme - The Research Into Use

Programme (RIU) is intended to increase demand for the natural resource research findings that were produced under the DFID’s Renewable Natural Resources

Research Strategy (RNRRS), by promoting innovative strategies to facilitate the flow of technology and information between individuals, organisations and institutions.

Responsible for assisting in programme management including the design and implementation of systems to monitor the performance of grantees organisations in complying with the terms of their funding agreements.

Millennium Challenge

Account

Impact Evaluation of Rural Roads Up-Grading Project: 5-year assessment of the impact of the upgrading of rural road projects in Tanzania. The survey will involve the interviewing of 4200 households in 280 communities across mainland Tanzania and Zanzibar.

July 2009

– Dec

World Bank, Rockwoold

Foundation and

Agence française de développement

(AFD),

Assessing the Long-term Economic Impact of HIV/AIDS on Households and

Individuals in Kagera Region (KHDS-3) - KHDS is a Living Standards Measurement

Survey (LSMS) which will track 3,200 households that were interviewed initially by the World Bank in 1991 and tracked and re-interviewed by EDI in 2003/4. In this third round EDI tracked and interviewed the Households from the first (and second) rounds to collect data to assess the impacts of HIV and other health shocks on their well being.

Feb 2008

- 2009

PMO-RALG (Prime

Minister’s Office –

Regional

Administration and

Local Government)

Dissemination of the Core Welfare Indicator Questionnaire (CWIQ) results to

District Councils. Design and implementation of dissemination exercises to ensure that the findings of each district level CWIQ survey were disseminated to stakeholders from district Government all the way down to the lowest administrative units (i.e. villages). The dissemination of CWIQ data will:

Timing Client and Country of Assignment Project Description

(i) promote evidence-based policy debate;

(ii) promote evidence-based policy formulation;

(iii) provide tools for district level M&E; and,

(iv) increase accountability of LGA to citizens.

The exercise involved the dissemination of the results of CWIQ to stakeholders in each district (councillors, district officials, NGOs, CBOs, Advocacy Groups, MPs, ‘interested citizens’, etc) through the facilitation of participative stakeholder workshops and the distribution of colourful, easy to read and informative posters and research briefs.

Jun 07 -

World Bank (LSMS group)

Tanzania Consumption and Welfare Survey

The objective of this 4000-household year-long survey was to improve on current methods of collecting survey data on household and individual economic welfare.

This was achieved through the careful design and implementation of experiments in the collection of the primary measures found in practice for both objective and subjective welfare measurement. The analytical approach was to design several alternative consumption modules with variation in terms of collection method

(diary or recall), length of detail in consumption items, and length of recall period.

Subjective welfare measures were anchored using innovative anchoring vignettes.

Each of the 8 modules is assigned to 500 households across 7 districts in Tanzania.

One module was a careful individual diary that will serve as the benchmark of consumption. The relative performance of each module is compared to the relative costs of administration. The results of this study yielded important information concerning the relative accuracy and costs of various consumption measurement designs.

Apr 06-

May 07

PMO-RALG

Core Welfare Indicator Questionnaire Survey of 28 Tanzanian Districts

A team of 35 people conducted surveys in 28 districts in Tanzania. District level data was collected for about one quarter of all districts in the country. The work included listing households in 840 Enumeration Areas and organising household interviews of 12,600 households. Data entry happens through Teleform software, using scannable questionnaires.

The data collection and entry methods allow one district to be listed in 2 weeks, interviewed in 2 weeks (sample size = 450 households) and all its questionnaires to be scanned in about 3 or 4 days. Software was written to automatically create all tables to ensure that reports come out just a few weeks after the completion of the field work. The work entailed the administration of 28 different surveys and the formulation 28 individual reports. These reports are intended for district level policy makers who, due to Tanzania’s ongoing decentralisation process, are becoming key policy makers and are increasingly in need of data to base their policies on.

Feb 05 –

Sept 06

World Bank, PREM

Moving out of Poverty Study – Simultaneous Mixing of Qualitative and

Quantitative Research Methods

Design of an innovative strategy to simultaneously mix qualitative and quantitative research methods to understand movements out of poverty over the past 10 years in Kagera. Using the KHDS data set (see below) an econometric prediction was made of the growth path a respondent was expected to follow between 1993 and 2004, based on what was known about him/her in 1993. Life histories were then

Timing Client and Country of Assignment Project Description conducted in matched sets of people who had similar predictions, but different outcomes. Thus a treatment and comparison group set-up is mimicked. The life histories were complemented with other qualitative data collection techniques such as focus group discussions and key informant interviews. Contrary to many qualitative studies, we managed to maintain very rigid sampling procedures (using the KHDS data as our sampling frame), which will allow us to make firmer inferences. This study is part of a 16-country study to understand the determinants of poverty from the bottom up. The global effort is directed by Deepa Narayan at

PREM, World Bank.

May –

Nov 05 Europe Aid

Evaluation of European Community Support to the United Republic of Tanzania.

Review of European Development Fund (exceeding 0.5 Billion Euros) support to

Tanzania from 2000-2005 (EDF 7-9).

Nov 04 –

Sept 05

World Bank (DECRG)

Orphanhood and its Long-run Impact on Children Formulation of a policy research paper using the KHDS data set (see below) to determine whether orphanhood matters in the long-run for health and education outcomes in Kagera, an area deeply affected by HIV-AIDS in Africa. The use of long-term panel data (11-13 years) allows an assessment of whether the effects of orphanhood persist into adult life.

The tracking of children who have left their original area of residence also allows for more robust inferences. These two features make this a unique study.

Jan 05 –

Aug 05

SNV – Netherlands

Organisations –

Implementation and Analysis of the CWIQ Survey in Karatu, Kondoa, Mbulu and

Monduli Districts

Project director of a research project conducting an 1800 household CWIQ survey in Karatu, Kondoa, Mbulu and Monduli districts. The resulting reports provided detailed district level baseline data on poverty, nutrition and other welfare indicators as well as of service use and satisfaction. Poverty mapping techniques were used to determine, for the first time, district level poverty rates.

Oct 04 –

July 05

World Bank (Social

Protection Unit)

Measuring Risk Perceptions

This research started from the premise that economists study data on choices that people make and from this deduce people’s preferences and expectations. This identification process, as well as the predictions based on it, become flawed when multiple sets of preferences and expectations are consistent with the same data.

The research explored one possible way forward: to measure directly people’s expectations on future states of the world. The research established the theoretical merits and practical constraints of doing so. Data on risk perceptions seems particularly relevant for understanding savings and investment behaviour in the developing world, where risk is pervasive and often posited to have significant costs.

The project established a field instrument to collect risk perception data. This culminated in a paper in the Social Protection Unit Discussion Paper Series (peer-reviewed).

Jun-Aug

World Bank (TUDTR)

Perceptions of Time and Distance

This project was to assess the quality of data collected through household surveys as they relate to time and distance travelled. EDI organised over 500 trips in which respondents were asked to estimate time and distance to a destination (water

Timing Client and Country of Assignment Project Description source or school) and were subsequently requested to walk this same distance with a pedometer to measure the actual time and distance. Initial comparisons of estimated and actual times and distances shows that correlations are weak, but slightly better for time than for distance (25% and 50% correlation respectively).

Further work on whether certain socio-economic groups systematically over or underestimate will be done. If measurement error is systematic, then this could lead to erroneous inferences about the relation between access and poverty.

Mar -

Sept 05

University of

Copenhagen

The Determinants of Schooling in Kagera Region – Sequential Quan/Qual Work

This project designed and implemented a qualitative instrument to look at the determinants of schooling in the KHDS clusters (see below). Very specifically it was geared towards understanding an empirical fact observed in the data, namely that villages with high ethno-linguistic fractionalisation performed worse in terms of schooling than those with low ethno-linguistic fractionalisation. The qualitative work was conducted to feed back into the quantitative work: to assess the likelihood of several possible scenarios to explain the empirical observation and determine strategies to test them in the data.

Jan 03 –

Aug 04

World Bank/Danida

Kagera Health and Development Survey: Tracking Individuals after 10+ years

Managing a team of 28 Interviewers, 7 Data Entry Operators, 4 Supervisors and 6 consultants to undertake a study into the long-run impacts of HIV/AIDS and other health shocks on the wealth dynamics of households and individuals within North

West Tanzania. This study entails the resurvey of the panel of households, communities and service providers in the Kagera Health and Development Survey

(KHDS) undertaken in 1991-1994.

It involves retracing all individuals from the initial 919 households Collection of community and service provider data augments this analysis and enables the tracking of how service provision has developed in the region over the past ten years. The novel feature of this survey is its long time period (11-13 years) combined with locating even respondents who reside outside their original place of residence.

Teams were sent across the whole of Tanzania and into neighbouring countries to collect data.

Apr-Nov

World Bank

Tanzania

Mixed Methods Analysis of Long-term Welfare Mobility

Using qualitative research methods, including focus group discussions and life histories, to determine the primary factors relating to upward and downward income and…