eduardo@movimentar.eu

Last update: 25 July 2017

Since 15 July 2015, there is a new version of the EuropeAid-PRAG documents, which lays the procedures for actions funded by the European Union (EU) worldwide. This post is focused on the new logical framework template, which is already being requested in some of the calls for proposals that are currently open.

The new logframe makes it clearer that projects need to have baseline and target values disaggregated by sex for their indicators already during the submission stage. Changes in the column for the intervention logic may generate some confusion though … The logframe now opens the possibility of more than one specific objective, naming them also “outcomes”. Below that, one will see “outputs” and then “activities”. This may contradict the European Commission’s Project Cycle Management Manual but seems to be in line with the policies adopted by other development agencies (e.g., USAID, DFID and SDC) which also use the result chain:

inputs –> activities –> outputs –> outcomes –> impact

Suggestion of project structure

Looking at the section “definitions” in the logframe template, the specific objective is not even mentioned. Therefore, I can imagine that in many cases the specific objective will be just dropped in many grant proposals. This can be good for result orientation as one now needs to breakdown results into outcomes (consequence/effect of project deliverables such as increased weekly income) and ouputs (concrete deliverables from activity implementation such as number of trainings or training participants by sex). This forces applicants to reflect more on objectively verifiable results.

My recommendation in terms of project structure is as follows:

1) A single overall objective and long-term goal (impact) indicating the strategic orientation of the project. The convention here is to start with “To contribute to … “,

2) Two or three key outcome areas (No more than five to ensure maximium clarity to implementing staff and key stakeholders). The convention is to present outcomes and outputs in the past tense (e.g., Improved coping strategies of at least 2,200 vulnerable farmer households).

3) One to three key outputs for each outcome area. These should also be phrased in past tense and must concretely come from activity implementation. So, take a few minutes to brainstorm products/deliverables for each activity. This will give you elements for designing the goal, outcomes, outputs as well as their indicators. In the end, always cross check if your brainstorming results are well refected in your outputs, outcomes and indicator statements.

4) You should have as many activities as necessary. Quantify activities as much as possible and do not include tasks (sub-activities) in the logframe but only the key project activities. I recommend to plan activities by output (max. four per output for simplicity and clarity). You need to make sure that the activities are really delivering the outputs you plan. If there is one output for which there is not related activity do not hesitate to add the required activity (or delete the output).

How important is the logframe?

The logframe is a crucial document. It should be the main guiding document for the project staff. It can help a lot to ensure smooth implementation when well designed. The logframe is one of the annexes to the contract with the European Uninon. Hence it will be legally binding.

That does not mean that the logframe is written on stone. The EU Project Cycle Manual (PCM) emphasises the importance of undertaking regular project monitoring and evaluation (M&E). Assessments of indicator performance during project implementation may require logframe adjustments (e.g., increases or reductions in indicator targets). So, the general rule is:

• If changes do not affect the basic purpose of the action (e.g., replacing two nurses by a doctor or the other way round should be fine), you should inform changes without delay (e.g., in interim reports following yearly project implementation review and planning workshops).
• If a major change is necessary that may affect the basic purpose of the action (e.g., substantial changes in the intervention logic – 1st column of the logframe), the project applicant will need to ask for permission from the project officer in the EU. They can do that by sending a revised logframe with highlighted changes and a letter with detailed justifications.

It is important to make sure that everyone have these PCM principles and contract rules in mind when designing and adjusting the logframe.

Increased scientific rigour in monitoring and evaluation

Projects are policy experiments, which require scientifically valid and reliable M&E data. Credible data is critically important to demonstrate value and to justify potential adjustments in your initial plans. This also helps the EU to justify the use of funds towards taxpayers. All this requires professional scientific methods and principles such as randomisation and reproducibility. Using digital data-collection tools (e.g., Kobotoolbox / Open Data Kit) and computer syntax to draw reproducible random samples, process and analyse data have become increasingly important in our digital era of analytic dashboards, algorithms and automated reports. These are great tools to ensure quality and manage the substantial data load that your project will need to work with.

I highly recommend watching Ester Dufluo’s TED talk on how the lack of data on aid’s impact raises questions about how to provide it. And rightly so. There are still a few colleagues who are still reluctant about employing data science and statistics to project management in international development and humanitarian cooperation. Yes, all this rigour leads to increased time and result pressure on project implementers. However, it also helps increasing transparency, real learning and impact potential. Integrating mobile-data collection and programmatic data processing and analysis can help you to save time and resources by increasing efficiency of your M&E processes. Doing all this on paper forms, for example, increases the likelihood of data-quality problems and is too resource intensive compared with the existing data-collection, database and visualisation tools.

Final remarks

Goal/impact indicators (the highest level) can be focused on a broader geographic area or on forecasts for a longer period than the project implementation, for example. Please be aware that you should also monitor and report the situation for long-term indicators or indicators focusing on broader administrative territorial units (e.g., regions, country, state or municipality).

The separation of “results” into “outcomes” and “outputs” with baseline and target values disaggregated by sex is likely to be a challenge, specially when it comes to output baseline values … You will need to become clearer about your targets by sex, and commit to that.

Good luck to us all in our next project proposals!

More tips about logframe and M&E design:

• Post: Starting an ECHO or DIPECHO application under time pressure
• Post: Designing survey forms with evaluative scales
• External link: How to fill in the logframe (Punto.Sud/Helpdesk EuropeAid)

Written by: Eduardo W. Ferreira, PhD / Consultant, data scientist, trainer and facilitator. Eduardo supports designing, managing and evaluating projects and programmes for consultancy firms, non-governmental organisations, governments, research institutions and international organisations (Additional information).

M&E-data tidying can help increasing your team’s productivity, saving resources and making more time available for the data-analysis stage and, of course, for your beneficiaries. If your project M&E data is “messy”, you will probably need to spend quite some time until you can begin any comprehensive analysis. Understanding well what tidy is data and ensuring that the output of your project’s M&E system is tidy can save you a great deal of trouble.

In the video below, you will find some useful tips and functions for tidying data with R, presented by Hadley Wickham, creator of many popular R packages.

Wickham’s paper on Tidy Data is another very useful resource and its abstract is presented below. You might also want to read his complete article on Tidy Data.

A huge amount of effort is spent cleaning data to get it ready for analysis, but there has been little research on how to make data cleaning as easy and effective as possible. This paper tackles a small, but important, component of data cleaning: data tidying. Tidy datasets are easy to manipulate, model and visualise, and have a specific structure: each variable is a column, each observation is a row, and each type of observational unit is a table. This framework makes it easy to tidy messy datasets because only a small set of tools are needed to deal with a wide range of un-tidy datasets. This structure also makes it easier to develop tidy tools for data analysis, tools that both input and output tidy datasets. The advantages of a consistent data structure and matching tools are demonstrated with a case study free from mundane data manipulation chores.