MERL Tech News

MERL Tech 101: Google forms

by Daniel Ramirez-Raftree, MERL Tech volunteer

In his MERL Tech DC session on Google Forms, Samhir Vesdev from IREX led a hands-on workshop on Google Forms and laid out some of the software’s capabilities and limitations. Much of the session focused on Google Forms’ central concepts and the practicality of building a form.

At its most fundamental level, a form is made up of several sections, and each section is designed to contain a question or prompt. The centerpiece of a section is the question cell, which is, as one would imagine, the cell dedicated to the question. Next to the question cell there is a drop down menu that allows one to select the format of the question, which ranges from multiple-choice to short answer.


At the bottom right hand corner of the section you will find three dots arranged vertically. When you click this toggle, a drop-down menu will appear. The options in this menu vary depending on the format of the question. One common option is to include a few lines of description, which is useful in case the question needs further elaboration or instruction. Another is the data validation option, which restricts the kinds of text that a respondent can input. This is useful in the case that, for example, the question is in a short answer format but the form administrators need the responses to be limited numerals for the sake of analysis.

The session also covered functions available in the “response” tab, which sits at the top of the page. Here one can find a toggle labeled “accepting responses” that can be turned off or on depending on the needs for the form.

Additionally, in the top right corner this tab, there are three dots arranged vertically, and this is the options menu for this tab. Here you will find options such as enabling email notifications for each new response, which can be used in case you want to be alerted when someone responds to the form. Also in this drop down, you can click “select response destination” to link the Google Form with Google Sheets, which simplifies later analysis. The green sheets icon next to the options drop-down will take you to the sheet that contains the collected data.

Other capabilities in Google Forms include the option for changing the color scheme, which you can access by clicking the palette icon at the top of the screen. Also, by clicking the settings button at the top of the screen you can limit the response amount to restrict people’s ability to skew the data by submitting multiple responses, or you can enable response editing after submission to allow respondents to go in and correct their response after submitting it.

Branching is another important tool in Google Forms. It can be used in the case that you want a particular response to a question (say, a multiple choice question) to lead the respondent to another related question only if they respond in a certain way.

For example, if in one section you ask “did you like the workshop?” with the answer options being “yes” and “no,” and if you want to know what they didn’t like about the workshop only if they answer “no,” you can design the sheet to take the respondent to a section with the question “what didn’t you like about the workshop?” only in the case that they answer “no,” and then you can design the sheet to bring the respondent back to the main workflow after they’ve answered this additional question.

To do this, create at least two new sections (by clicking “add section” in the small menu to the right of the sections), one for each path that a person’s response will lead them down. Then, in the options menu on the lower right hand side select “go to section based on answer” and using the menu that appears, set the path that you desire.

These are just some of the tools that Google Forms offers, but with just these it is possible to build an effective form to collect the data you need. Samhir ended with a word of caution that Google has been known to shut down popular apps, so you should be wary about building an organization strategy around Google Forms.

Gender-Based Violence Information System Design

by Stacey Berlow of Project Balance. Stacey co-facilitated a session on a Gender Based Violence Information System in Zambia at MERL Tech DC.

John and Stacey croppedA big thank you to our client, World Vision, and to Yeva Avakyan, Head of Gender and Inclusion at World Vision USA, for inviting Project Balance to participate in the recent panel sessions at MERL Tech and InterAction.

We participated on the panels with World Vision colleagues Holta Trandafili, Program Quality Specialist and John Manda, Senior Monitoring and Evaluation Technical Advisor Zambia last week (September 6-8th). The sessions received a lot of great feedback and participation. Yeva and her team put together this impactful video about the prevalence of gender based violence in Zambia and how One Stop Centers provide needed services to women and men who experience violence in their lives. World Vision works in close collaboration with the Zambian government to roll out gender based violence support services.

The Zambian 2014 statistic are compelling:

  • 37% experienced physical violence within the 12 months prior to the survey.
  • 43% of women age 15-49 have experienced physical violence at least once since age 15
  • 47% of ever-married women age 15-49 report ever having experienced physical, sexual, and/or emotional violence from their current or most recent partner
  • 31% report having experienced such violence in the past 12 months.
  • Among ever-married women who had experienced IPV in the past 12 months, 43% reported experiencing physical injuries.
  • 10% of women reported experiencing violence during pregnancy.
  • 9% of women who have experienced violence have never sought help and never told anyone about the violence.

The drivers of GBV include:

 

  • Norms that teach women to accept and tolerate physical violence, and teach men that it is normal to beat his wife.
  • Extreme poverty, high levels of unemployment
  • Women’s extreme economic dependence on men
  • Socialization practices of boys and girls in schools and the community
  • Sexual cleansing practices
  • Belief that having sex with a child who is a virgin will cure HIV/AIDS
  • Initiation ceremonies that encourage young women to be submissive
  • Forced early & child marriage

An occasionally connected system was built that allows facilities to enter and save data as well as run reports locally and when there is an internet connection, the data automatically synchronizes to a central server where data from across all facilities is available for reporting.

GBVIMS_System_Setup

A participant asked John how data collected was used and if technology positively impacted the program. John gave some concrete examples of how the data showed differences in the number of people receiving certain types of services between facilities and regions. The Zambian team asked “Why the differences?”. This led to analysis of processes which have been adjusted so that survivors can receive medical and psychological help as soon as possible. The technology allows trends to be identified earlier through automated reporting rather than having to hand calculate indicators at the end of each reporting period. As the technology provider, we were excited to hear that the GBVIMS is being actively implemented and program participants and managers are using the data for decision making. It’s wonderful to be part of such a passionate and professional team.

M&E software – 8 Tips on How to Talk to IT Folks

.

You want to take your M&E system one step further and introduce a proper M&E software? That’s great, because a software has the potential of making the monitoring process more efficient and transparent, reducing errors and getting more accurate data. But how to go about it? You have three options:

  1. You build your own system, for example in Microsoft Excel;
  2. You purchase an M&E software package off-the-shelf;
  3. You hire an IT consultant to set up a customized M&E system according to your organization’s specific requirements.

If options one and two do not work out for you, you can hire consultants to develop a solution for you. You will probably start a public tender to find the most suitable IT company to entrust with this task. While there a lot of things to pay attention to when formulating the Terms of Reference (TOR), I would like to give you some tips specifically about the communication with the hired IT consultants. These insights come from years of experience of being on both sides: The party who wants a tool and needs to describe it to the implementing programmers and being the IT guy (or rather lady) who implements Excel and web-based database tools for M&E.

To be on the safe side, I recommend you to work with this assumption: IT consultants have no clue about M&E. There are few IT companies who come from the development sector, like energypedia consult does, and are familiar with M&E concepts such as indicators, logframes and impact chains. To still get what you need, you should pay attention to the following communication tips:

  1. Take your time explaining what you need: Writing TOR takes time – but it takes even longer and becomes more costly when you hire somebody for something that is not thought through. If you don’t know all the details right from the start, get some expert assistance in formulating terms – it’s worthwhile.
  2. Use graphs: Instead of using words to describe your monitoring logic and the system you need, it is much easier to make graphs to depict the structure, user groups, linking of information, flow of monitoring data etc.
  3. Give examples: When unsure about how to put a feature into words, send a link or a screenshot of the function that you might have come across elsewhere and wish to have in your tool.
  4. Explain concepts and terminology: Many results frameworks work with the terms “input” and “output”. Most IT guys, however, will not have equipment and finished schools in mind, but rather data flows that consist of inputs and outputs. Make sure you clarify this. Also, the term web-based or web monitoring itself is a source of misunderstanding. In the IT world, web monitoring refers to monitoring activity in the internet, for example website visits or monitoring a server. That is probably not what you want when building up an M&E system for e.g. a good governance programme.
  5. Meet in person: In your budget calculation, allow for at least one workshop where you meet in person, for example a kick-off workshop in which you clarify your requirements. This is not only a possibility to ask each other questions, but also to get a feeling of the other party’s language and way of thinking.
  6. Maintain a dialogue: During the implementation phase, make sure to stay in regular touch with the programmers. Ask them to show you updates every once in a while to allow you to give feedback. When you detect that the programmers are heading into the wrong direction, you want to find out rather sooner than later.
  7. Document communication: When we implement web-based systems, we typically create a page within the web platform itself that outlines all the agreed steps. This list serves as a to-do list and an implementation protocol at the same time. It facilitates communication, particularly when on both sides multiple persons are involved that are not always present in all meetings or phone calls.
  8. Be prepared for misunderstandings: They happen. It’s normal. Plan for some buffer days before launching the final tool.

In general, the implementation phase should allow for some flexibility. As both parties learn from each other during the process, you should not be afraid to adjust initial plans, because the final tool will benefit greatly from it (if the contract has some flexibility). Big customized IT projects take some time.

If you need more advice on this matter and some more insights on setting up IT-based M&E systems, please feel free to contact me any time! In the past we supported some clients by setting up a prototype for their web-based M&E system with our flexible WebMo approach. During the prototype process the client learnt a lot and afterwards it was quite easy for other developers to copy the prototype and migrate it to their e.g. Microsoft Share Point environment (in case your IT guys don’t believe in Open Source or don’t want to host third-party software on their server).

Please leave your comments, if you think that I have missed an important communication rule.

Good luck!

M&E Squared: Evaluating M&E Technologies

by Roger Nathanial Ashby, Co-Founder & Principal Consultant, OpenWise

The universe of MERL Tech solutions has grown exponentially. In 2008 monitoring and evaluating tech within global development could mostly be confined to mobile data collection tools like Open Data Kit (ODK), and Excel spreadsheets to analyze and visualize survey data. In the intervening decade a myriad of tools, companies and NGOs have been created to advance the efficiency and effectiveness of monitoring, evaluation, research and learning (MERL) through the use of technology. Whether it’s M&E platforms or suites, satellite imagery, remote sensors, or chatbots, new innovations are being deployed every day in the field.

However, how do we evaluate the impact when MERL Tech is the intervention itself? That was the question and task put to participants of the “M&E Squared” workshop at MERL Tech 2017.

Workshop participants were separated into three groups that were each given a case study to discuss and analyze. One group was given a case about improving the learning efficiency of health workers in Liberia through the mHero Health Information System (HIS). The system was deployed as a possible remedy to some of the information communication challenges identified during the 2014 West African Ebola outbreak. A second group was given a case about the use of RapidPro to remind women to attend antenatal care (ANC) for preventive malaria medicine in Guinea. The USAID StopPalu project goal was to improve the health of infants by increasing the percent of women attending ANC visits. The final group was given a case about using remote images to assist East African pastoralists. The Satellite Assisted Pastoral Resource Management System (SAPARM) informs pastoralists of vegetation through remote sensing imagery so they can make better decisions about migrating their livestock.

After familiarizing ourselves with the particulars of the case studies, each group was tasked to present their findings to all participants after pondering a series of questions. Some of the issues under discussion included

(1) “How would you assess your MERL Tech’s relevance?”

(2) “How would you evaluate the effectiveness of your MERL Tech?”

(3) “How would you measure efficiency?” and

(4) “How will you access sustainability?”.

Each group came up with some innovative answers to the questions posed and our facilitators and session leads (Alexandra Robinson & Sutyajeet Soneja from USAID and Molly Chen from RTI) will soon synthesize the workshop findings and notes into a concise written brief for the MERL Tech community.

Before the workshop closed we were all introduced to the great work done by SIMLab (Social Impact Lab) in this area through their SIMLab Monitoring and Evaluation Framework. The framework identifies key criteria for evaluating M&E including:

  1. Relevance – The extent to which the technology choice is appropriately suited to the priorities and capacities of the context of the target group or organization.
  2. Effectiveness – A measure of the extent to which an information and communication channel, technology tool, technology platform, or a combination of these attains its objectives.
  3. Efficiency – Measure of the outputs (qualitative and quantitative) in relation to the inputs.
  4. Impact – The positive and negative changed produced by technology introduction, change in a technology tool, or platform on the overall development intervention (directly or indirectly; intended or unintended).
  5. Sustainability – Measure of whether the benefits of a technology tool or platform are likely to continue after donor funding has been withdrawn.
  6. Coherence – How related is the technology to the broader policy context (development, market, communication networks, data standards & interoperability mandates, and national & international law) within which the technology was developed and implemented.

While it’s unfortunate that SIMLab stopped most operations in early September 2017, their exceptional work in this and other areas lives on and you can access the full framework here.

I learned a great deal in this session from the facilitators and my colleagues attending the workshop. I would encourage everyone in the MERL Tech community to take the ideas generated during this workshop and the great work done by SIMLab into their development practice. We certainly intend to integrate much of these insights into our work at OpenWise. Read more about “The Evidence Agenda” here on SIMLab’s blog. 

 

 

 

Making (some) sense of data storage and presentation in Excel

By Anna Vasylytsya. Anna is in the process of completing her Master’s in Public Policy with an emphasis on research methods. She is excited about the role that data can play in improving people’s lives!

At the MERL Tech Conference, I attended a session called “The 20 skills that solve 80% of M&E problems” presented by Dr. Leslie Sage of DevResults. I was struck by the practical recommendations Leslie shared that can benefit anyone that uses Excel to store and/or present data.

I boiled down the 20 skills presented in the session into three key takeaways, below.

1. Discerning between data storage and data presentation

Data storage and data presentation serve two different functions and never the two shall meet. In other words, data storage is never data presentation.

Proper data storage should not contain merged cells, subheadings, color used to denote information, different data types within cells (numbers and letters), more than one piece of data in a cell (such as disaggregations). Additionally, in proper data storage, columns should be the variables and rows as the observations or vice versa. Poor data storage practices need to be avoided because they mean that you cannot use Excel’s features to present the data.

A common example of poor data storage:

Excel 1

 

One of the reasons that this is not good data storage is because you are not able to manipulate this data using Excel’s features. If you needed this data in a different format or you wanted to visualize it, you would have to do this manually, which would be time consuming.

Here is the same data presented in a “good” storage format:

2Good_Data_Storage

 

Data stored this way may not look as pretty, but it is not meant to be presented or read in within the sheet. This is an example of good data storage because each unique observation gets a new row in the spreadsheet. When you properly store data, it is easy for Excel to aggregate the data and summarize it in a pivot table, for example.

2. Use Excel’s features to organize and clean data

You do not have to use precious time to organize or clean data manually. Here are a few recommendations on Excel’s data organization and cleaning features:

  • To join to cells that have text into one cell, use the concatenate function.
  • To split text from one cell into different cells, use the text to columns
  • To clean text data, use Excel’s functions: trim, lower, upper, proper, right, left, and len.
  • To move data from rows into columns or columns into rows, use Excel’s transpose feature.
  • There is a feature to remove duplicates from the data.
  • Create a macro to automate simple repetitive steps in Excel.
  • Insert data validation in an excel spreadsheet if you are sending a data spreadsheet to implementers or partners to fill out.
    • This restricts the type of data or values that can be entered in certain parts of the spreadsheet.
    • It also saves you time from having to clean the data after you receive it.
  • Use the vlookup function in Excel in your offline version to look up a Unique ID
    • Funders or donors normally require that data is anonymized if it is made public. While not the best option for anonymizing data, you can use Excel if you haven’t been provided with specific tools or processes.
    • You can create an “online” anonymized version that contains a Unique ID and an “offline version” (not public) containing the ID and Personally Identifiable Information (PII). Then, if you needed to answer a question about a Unique ID, for example, your survey was missing data and you needed to go back and collect it, you can use vlookup to find a particular record.

3. Use Excel’s features to visualize data

One of the reasons to organize data properly so that you can use Excel’s Pivot Table feature.

Here is an example of a pivot table made from the data in the good data storage example above (which took about a minute to make):

3Pivot_Table

Using the pivot table, you can then use Excel’s Create a Chart Feature to quickly make a bar graph:

4BarGraph

In the Future

I have fallen prey to poor data storage practices in the past. Now that I have learned these best practices and features of Excel, I know I will improve my data storage and presentation practices. Also, now that I have shared them with you; I hope that you will too!

Please note that in this post I did not discuss how Excel’s functions or features work or how to use them. There are plenty of resources online to help you discover and explore them. Some helpful links have been included as a start. Additionally, the data presented here are fictional and created purely for demonstration purposes.

Qualitative Coding: From Low Tech to High Tech Options

by Daniel Ramirez-Raftree, MERL Tech volunteer

In their MERL Tech DC session on qualitative coding, Charles Guedenet and Anne Laesecke from IREX together with Danielle de Garcia of Social Impact offered an introduction to the qualitative coding process followed by a hands-on demonstration on using Excel and Dedoose for coding and analyzing text.

They began by defining content analysis as any effort to make sense of qualitative data that takes a volume of qualitative material and attempts to identify core consistencies and meanings. More concretely, it is a research method that uses a set of procedures to make valid inferences from text. They also shared their thoughts on what makes for a good qualitative coding method.

Their belief is that: it should

  • consider what is already known about the topic being explored
  • be logically grounded in this existing knowledge
  • use existing knowledge as a basis for looking for evidence in the text being analyzed

With this definition laid out, they moved to a discussion about the coding process where they elaborated on four general steps:

  1. develop codes and a codebook
  2. decide on a sampling plan
  3. code your data
  4. go back and do it again!
  5. test for reliability

Developing codes and a codebook is important for establishing consistency in the coding process, especially if there will be multiple coders working on the data. A good way to start developing these codes is to consider what is already known. For example, you can think about literature that exists on the subject you’re studying. Alternatively, you can simply turn to the research questions the project seeks to answer and use them as a guide for creating your codes. Beyond this, it is also useful to go through the content and think about what you notice as you read. Once a codebook is created, it will lend stability and some measure of objectivity to the project.

The next important issue is the question of sampling. When determining sample size, though a larger sample will yield more robust results, one must of course consider the practical constraints of time, cost and effort. Does the benefit of higher quality results justify the additional investment? Fortunately, the type of data will often inform sampling. For example, if there is a huge volume of data, it may be impossible to analyze it all, but it would be prudent to sample at least 30% of it. On the other hand, usually interview and focus group data will all be analyzed, because otherwise the effort of obtaining the data would have gone to waste.

Regarding sampling method, session leads highlighted two strategies that produce sound results. One is systematic random sampling and the other is quota sampling–a method employed to ensure that the proportions of demographic group data are fairly represented.

Once these key decisions have been made, the actual coding can begin. Here, all coders should work from the same codebook and apply the codes to the same unit of analysis. Typical units of analysis are: single words, themes, sentences, paragraphs, and items (such as articles, images, books, or programs). Consistency is essential. A way to test the level of consistency is to have a 10% overlap in the content each coder analyzes and aim for 80% agreement between their coding of that content. If the coders are not applying the same codes to the same units this could either mean that they are not trained properly or that the code book needs to be altered.

Along a similar vein, the fourth step in the coding process is to test for reliability. Challenges in producing stable and consistent results in coding could include: using a unit of analysis that is too large for a simple code to be reliably applied, coding themes or concepts that are ambiguous, and coding nonverbal items. For each of these, the central problem is that the units of analysis leave too much room for subjective interpretation that can introduce bias. Having a detailed codebook can help to mitigate against this.

After giving an overview of the coding process, the session leads suggested a few possible strategies for data visualization. One is to use a word tree, which helps one look at the context in which a word appears. Another is a bubble chart, which is useful if one has descriptive data and demographic information. Thirdly, correlation maps are good for showing what sorts of relationships exist among the data. The leads suggested visiting the website stephanieevergreen.com/blog for more ideas about data visualization.

Finally, the leads covered low-tech and high-tech options for coding. On the low-tech end of the spectrum, paper and pen get the job done. They are useful when there are few data sources to analyze, when the coding is simple, and when there is limited tech literacy among the coders. Next up the scale is Excel, which works when there are few data sources and when the coders are familiar with Excel. Then the session leads closed their presentation with a demonstration of Dedoose, which is a qualitative coding tool with advanced capabilities like the capacity to code audio and video files and specialized visualization tools. In addition to Dedoose, the presenters mentioned Nvivo and Atlas as other available qualitative coding software.

Despite the range of qualitative content available for analysis, there are a few core principles that can help ensure that it is analyzed well, these include consistency and disciplined methodology. And if qualitative coding will be an ongoing part of your organization’s operations, there are several options for specialized software that are available for you to explore. [Click here for links and additional resources from the session.]

Being data driven… can it be more than a utopia?

by Emily Tomkys Valteri, the ICT in Program Accountability Project Manager at Oxfam GB. In her role, Emily drives Oxfam’s thinking on the use of Information and Communications Technologies (ICT) for accountability and supports staff with applications of ICTs within their work. 

Every day the human race generates enough data to fill 10 million blu-ray discs and if you stacked them up it would be four times the height of the Eiffel tower. Although the data we process at Oxfam is tiny in comparison, sometimes the journey towards being “data driven” feels like following the yellow brick road to The Emerald City. It seems like a grand ideal, but for anyone who knows the film, inflated expectations are set to be dashed. Does data actually help organisations like Oxfam better understand the needs of communities affected by disaster or poverty? Or do we need to pull back the curtain and manage our expectations about getting the basics right? When there are no ruby slippers, we need to understand what it is we can do with data and improve the way data is managed and analysed across countries and projects.

The problem

Oxfam works in over 90 countries using a variety of different data management and analysis tools that are developed or purchased in country. In the past, we have experimented with software licenses and database expertise, but we have started aiming for a more joined up approach. It’s our belief that good systems which build in privacy by design can help us stay true to values in our rights based Responsible Program Data Policy and Information Systems Data Security guidelines – which are about treating those people whom data is about with dignity and respect.

One of our most intractable challenges is that Oxfam’s data is analysed in system silos. Data is usually collected and viewed through a project level lens. Different formats and data standards make it difficult to compare across countries, regions or even globally. When data remains in source systems, trying to analyse between different systems is long and manual, meaning that any meta analysis is rarely done. One of the key tenants of Responsible Data is to only collect data you can use and to make the most of that information to effectively meet people’s needs. Oxfam collects a lot of valuable data and we think we need to do more with it: analyse more efficiently, effectively, at national and beyond level to drive our decision making in our programmes.

The solution

In response, Oxfam has begun creating the DataHub: a system which integrates programme data into a standard set of databases and presents it to a reporting layer for analysis. It bakes in principles of privacy and compliance with new data protection laws by design. Working with our in-house agile software development team we conducted four tech sprints, each lasting two weeks. Now we have the foundations. One of our standard data collection tools, SurveyCTO, is being pushed via a webhook into our unstructured database, Azure Cosmos DB. Within this database, the data is organised into collections, currently set up by country. From here, the data can be queried using Power BI and presented to programme teams for analysis. Although we only have one source system into quantitative analysis for now, the bigger picture will have lots of source systems and a variety of analysis options available.

To get to where we are today, Oxfam’s ICT in Programme team worked closely with the Information Systems teams to develop a solution that was in line with strategy and future trends. Despite the technology being new to Oxfam, the solution is relatively simple and we ensured good process, interoperability and that tools available to us were fit for purpose. This collaborative approach gave us the organisational support to prioritise these activities as well as the resources required to carry them out.

This journey wasn’t without its challenges, some of which are still being worked on. The EU General Data Protection Regulation (GDPR) coming into force in May 2018, and Oxfam has had to design the DataHub with this in mind. At this stage, data is anonymised during integration and so no Personally Identifiable Information (PII) enters the DataHub due to a series of configurations and processes we have put in place. Training and capacity is another challenge, we need to encourage a culture of valuing the data. This will only be of benefit to teams and the organisation if they make use of the system, investing time and resources to learning it.

We are excited about the potential of the DataHub and the success we have already had in setting up the infrastructure to enable more efficient data analysis and more responsive programming as well as save resources. We are keen to work with and share ideas with others. We know there is a lot of work ahead to foster a data driven organisation but we’re starting to feel, with the right balance of technology, process and culture it’s more realistic than we might have first hoped.

 

 

You can’t have Aid…without AI: How artificial intelligence may reshape M&E

by Jacob Korenblum, CEO of Souktel Digital Solutions

Photo: wikipedia.org/

Potential—And Risk

The rapid growth of Artificial Intelligence—computers behaving like humans, and performing tasks which people usually carry out—promises to transform everything from car travel to personal finance. But how will it affect the equally vital field of M&E? As evaluators, most of us hate paper-based data collection—and we know that automation can help us process data more efficiently. At the same time, we’re afraid to remove the human element from monitoring and evaluation: What if the machines screw up?

Over the past year, Souktel has worked on three areas of AI-related M&E, to determine where new technology can best support project appraisals. Here are our key takeaways on what works, what doesn’t, and what might be possible down the road.

Natural Language Processing

For anyone who’s sifted through thousands of Excel entries, natural language processing sounds like a silver bullet: This application of AI interprets text responses rapidly, often matching them against existing data sets to find trends. No need for humans to review each entry by hand! But currently, it has two main limitations: First, natural language processing works best for sentences with simple syntax. Throw in more complex phrases, or longer text strings, and the power of AI to grasp open-ended responses goes downhill. Second, natural language processing only works for a limited number of (mostly European) languages—at least for now. English and Spanish AI applications? Yes. Chichewa or Pashto M&E bots? Not yet. Given these constraints, we’ve found that AI apps are strongest at interpreting basic misspelled answer text during mobile data collection campaigns (in languages like English or French). They’re less good at categorizing open-ended responses by qualitative category (positive, negative, neutral). Yet despite these limitations, AI can still help evaluators save time.

Object Differentiation

AI does a decent job of telling objects apart; we’ve leveraged this to build mobile applications which track supply delivery more quickly & cheaply. If a field staff member submits a photo of syringes and a photo of bandages from their mobile, we don’t need a human to check “syringes” and “bandages” off a list of delivered items. The AI-based app will do that automatically—saving huge amounts of time and expense, especially during crisis events. Still, there are limitations here too: While AI apps can distinguish between a needle and a BandAid, they can’t yet tell us whether the needle is broken, or whether the BandAid is the exact same one we shipped. These constraints need to be considered carefully when using AI for inventory monitoring.

Comparative Facial Recognition

This may be the most exciting—and controversial—application of AI. The potential is huge: “Qualitative evaluation” takes on a whole new meaning when facial expressions can be captured by cameras on mobile devices. On a more basic level, we’ve been focusing on solutions for better attendance tracking: AI is fairly good at determining whether the people in a photo at Time A are the same people in a photo at Time B. Snap a group pic at the end of each community meeting or training, and you can track longitudinal participation automatically. Take a photo of a larger crowd, and you can rapidly estimate the number of attendees at an event.

However, AI applications in this field have been notoriously bad at recognizing diversity—possibly because they draw on databases of existing images, and most of those images contain…white men. New MIT research has suggested that “since a majority of the photos used to train [AI applications] contain few minorities, [they] often have trouble picking out those minority faces”. For the communities where many of us work (and come from), that’s a major problem.

Do’s and Don’ts

So, how should M&E experts navigate this imperfect world? Our work has yielded a few “quick wins”—areas where Artificial Intelligence can definitely make our lives easier: Tagging and sorting quantitative data (or basic open-ended text), simple differentiation between images and objects, and broad-based identification of people and groups. These applications, by themselves, can be game-changers for our work as evaluators—despite their drawbacks. And as AI keeps evolving, its relevance to M&E will likely grow as well. We may never reach the era of robot focus group facilitators—but if robo-assistants help us process our focus group data more quickly, we won’t be complaining.

MERL Tech London session ideas due this Friday, Nov 10th!

MERL Tech London is coming up on March 19-20, 2018. Session ideas are due by Friday, November 10th, so be sure to get yours in this week!!

Submission Deadline: Friday, November 10, 2017.

Session leads receive priority for the available seats at MERL Tech and a discounted registration fee. You will hear back from us in early December and, if selected, you will be asked to submit an updated and final session title, summary and outline by January 19th, 2018.

Topics we’re looking for:

  • Case studies: Sharing end-to-end experiences/learning from a MERL Tech process
  • MERL Tech 101: How-to use a MERL Tech tool or approach
  • Methods & Frameworks: Sharing/developing/discussing methods and frameworks for MERL Tech
  • Data: Big, large, small, quant, qual, real-time, online-offline, approaches, quality, etc.
  • Innovations: Brand new, untested technologies or approaches and their application to MERL(Tech)
  • Debates: Lively discussions, big picture conundrums, thorny questions, contentious topics related to MERL Tech
  • Management: People, organizations, partners, capacity strengthening, adaptive management, change processes related to MERL Tech
  • Evaluating MERL Tech: comparisons or learnings about MERL Tech tools/approaches and technology in development processes
  • Failures: What hasn’t worked and why, and what can be learned from this?
  • Demo Tables: to share MERL Tech approaches, tools, and technologies
  • Other topics we may have missed!

To get you thinking — take a look at past agendas from MERL Tech LondonMERL Tech DC and MERL Tech News.

Submit your session idea now!

We’re actively seeking a diverse (in every way) set of MERL Tech practitioners to facilitate every session. We encourage organizations to open this opportunity to colleagues and partners working outside of headquarters and to support their participation. (And please, no all-male panels!)

MERL Tech is dedicated to creating a safe, inclusive, welcoming and harassment-free experience for everyone. Please review our Code of Conduct. Session submissions are reviewed by our steering committee.

Submit your session ideas by November 10th!

If you have any questions about your submission idea, please contact Linda Raftree.

(Registration is also open!)

Visualizing what connects us: Social Network Analysis (SNA) in M&E

by Anne Laesecke (IREX) and Danielle de García (Social Impact). This post also appears on the Social Impact blog and  the IREX blog.

SNA, or Social Network Analysis, continues to gain momentum in the M&E space. This year at MERL Tech, we held an SNA 101 session, giving a quick-and-dirty overview of what it is, how it can contribute to M&E, and useful tips and tools for conducting an SNA. If you missed it, here’s what you need to know:

What is SNA?

SNA is a way to analyze social systems through relationships. Analyzing and visualizing networks can reveal critical insights for understanding relationships between organizations, supply chains; social movements; and/or between individuals. It’s a very versatile tool which can be used throughout the program cycle to measure things like trust and social capital, information flows, resources, collaboration, and disease spread, among other things.

SNA uses a different vocabulary than other types of analyses. For example, the relationships we measure are called ties or links, and the entities that make up a network are called nodes or actors. These can be organizations, people, or even whole networks themselves. We can study nodes more closely by looking at their attributes – things that characterize them (like demographic information), and we can learn more about how nodes interact and cluster by studying communities or modalities within networks. Various measures of the roles nodes play in a network, as well as measures that characterize the networks themselves, can reveal a lot about the systems and hidden relationships at play. For example, we can determine who has the most ties with other actors; who is relatively cut off from the network, or who is connected to the most well-connected actors.

Why would you use SNA for M&E?

The term “social network analysis” often triggers associations with social media, but SNA uses data from a variety of platforms (including but not limited to social media!). For instance, SNA can identify key influencers in systems – important for programs that rely on thinking politically. SNA can also be a useful tool in processing big data with applications for cybersecurity as well as creating biological and epidemiological projections. Beyond looking at networks of individuals, SNA can explore relationships with concepts through analysis of qualitative data and concept mapping. It can also look at organizational risks and processes (think about comparing an organizational chart with who people actually go to within an organization for information).

How do you do SNA?

Conducting SNA mostly follows the same procedure as other analysis.

  1. Determine your purpose and questions. What decisions do you need to make based on the data you get? Who is your audience and what do they need to know? Answering these questions can help you decided what you are trying to measure and how.
  2. Collect your data. SNA can incorporate lots of different data forms, including customized surveys and interviews asking actors in your network about the links they have, external data such as census information or other public records to further inform attributes or triangulate your custom data; and mapping of key locations or concepts. One thing to consider while conducting an SNA – data cleaning is usually a heavier lift than for other types of analysis.
  3. Crunch those numbers. SNA uses matrices to calculate various measures – from types of centrality to network density and beyond. Lucky for us, there are plenty of tools that take on both the analysis and visualization portions of SNA. However, another consideration as you analyze your data is that network data is often not generalizable in the same way as some statistical analysis. If you miss a key node in the network, you may miss an entire portion that is only linked through that node.
  4. Visualize the network. Network visualizations are one of the most distinctive features of SNA and can be incredibly useful as tools to engage partners about your findings. There is a wealth of analysis and visualization tools that can help you do this. We created a worksheet that outlines several, but a few of the most popular are UCINet, Gephi, and NodeXL.
  5. Interpret your results. You now have a beautiful graph that shows what nodes are important in your network. So what? How does it relate to your program? Your interpretation should answer the questions around the purpose of your analysis, but beyond interpretation can serve to improve your programming. Often, SNA results can help make projections for program sustainability based on who key players are and who can continue championing work, or projecting where trends seem to be going and anticipating activities around those areas.

Conclusions and resources

We barely scratched the surface of what SNA can do and there are so many more applications! Some great resources to learn more are the SNA TIG of the American Evaluation Association, Stephen Borgatti’s course website on SNA, and a site of his dedicated completely to designing surveys for SNA.