The LFA Debate: A summary of the theory behind the Logical Framework Approach

-- Greg Armstrong--


[Edited to update links July 2019]

While SIDA has recently updated its publications on the Logical Framework, some of the earlier publications are still useful.  The first of three early SIDA papers describes the strengths, and the sequence in using the LFA in results-based management.

SIDA's LFA Papers: 1-The Logical Framework Approach: A Summary of the theory behind the LFA method


Level of difficulty: Moderate
Primarily useful for: Project planners
Limitations: Some ambiguity in the definition of results
Length: 35 pages
Most useful sections: Relating problem analysis to activity design (p. 9-12)


Background: Evolution of a debate on the LFA

The Logical Framework Approach - SIDA

The Swedish International Development Agency published three papers online between 2004-2006 that illustrate in some ways the evolution of a debate that exists over the utility of the Logical Framework Approach for results-based project planning, management and evaluation of international development projects. While Results-Based Management is not invariably tied to the LFA, the general approach of a sequence of activities, short-term and long term results, with or without a formal framework, is common to most RBM approaches. In this, and two separate posts, I will review these three documents.

• The first was The Logical Framework Approach, A summary of the theory behind the LFA method, which emphasises the strengths of the logical framework approach for RBM,
• The second is a paper probing in some detail the weaknesses of the LFA, and
• The third a proposal to link more qualitative methods with the LFA.

[Update: More recent publications have been added since this original review, and they are listed at the end of this post]

A Logical approach to development planning

Kari Ortengren's 2004 paper, "The Logical Framework Approach: A Summary of the Theory behind the LFA Method" is a straight-forward explanation of nine steps in using the logical framework approach for designing a project, or programme.

The general presentation is unexciting, in comparison, to say, the 2005 Philip Dearden guide on using the LFA for multi-agency planning which I reviewed earlier, but that, I assume, is a function of the donor's format. While this paper is not quite as easy to read, it does make some good points about the sequence of project and programme design. And although it is not expressly noted here, in a subsequent paper in 2006 on applying the general approach to environmental project design, the author made the clear distinction between the Logical Framework Approach and the Logical Framework Matrix.

The matrix, of course, is the grid - the approach is the process by which the information in the grid can be developed -- and much more important than the matrix.

Example of the Logical Framework Matrix for a project on drinking water
-Click to enlarge-

Problem identification - by stakeholders

The most useful part of this paper is its focus on problem identification. The paper notes - and many readers will recognise this from their own experience, that many projects appear to be started with a solution in search of a clear problem. By "clear" the writer appears to mean a problem defined in a way in which not just the bold problem statement is made, but in which there is a clear and detailed discussion of both the multiple causes, and multiple downstream effects of the problem.

In particular the paper is useful when it urges planners to avoid stating problems in terms of pre-ordained solutions. The writer urges us to avoid defining problems in terms of "absence" -- "absence of funding", or "absence of pesticides" for example, which more or less ignore what might be revealing causes of a problem, and limit discussion of alternative solutions.


"Lack of funding" as a problem statement, she notes, does not explore the possible details of the problem, which may be inadequate organizational capacity for financial control, planning, or other issues which could be directly susceptible to project intervention; and "Lack of pesticides" pre-empts discussion of alternative solutions by assuming pesticides are the only solution to crop failures.


The paper makes it clear that donors are not the best placed to conduct a problem analysis: "The problem analysis has to be made by the relevant stakeholders, including the owners of the problem, the people who know the situation, not by consultants or financing agencies." The paper does suggest, however, that donors and consultants can facilitate the discussion and exploration of problems.

Activity design

It is true that some projects become obsessed with activities, rather than with how activities relate to a problem, and prescription here - a reasonable one - is that...

"...the causes of the problem shall be treated by the activities, which are implemented within the framework of the project. The effects are handled automatically by treating the causes of the focal problem. Hence, no separate activities are needed for handling the effects."

This paper includes a short introduction to planning a project design workshop. The discussion lists seven major topics for discussion, the physical layout required and the time that should be allocated.



Limitations: Assumptions


There are two limitations here, although neither are serious, and both may be a product of the donor agency's prescriptions:

1. Results are defined rather amorphously here, using the usual donor jaron - objectives, purposes, outputs, and this never makes life any simpler.
2. Risks are usually discussed in guides or papers about results-based management or the Logical Framework Approach, but there is rarely any detailed discussion of assumptions.

Comment:  This paper mentions both identification of risks and discussion of assumptions in two pages. We could read between the lines here in the more detailed discussion of problem identification, to see the possibility for discussion of assumptions, but it would be useful if this were explored in more detail.  

If assumptions are so central to project success - to sustainability but also to examination of whether interventions make sense in development terms, then much more detailed discussion of assumptions should be a priority for everybody - donors, implementing agencies and other participants.  [[Update:  Later SIDA publications including a 2016 update by the same author, do explore this in more detail]


In the rbm training workshops I have conducted I have found several times that clarifying assumptions about a problem, and about implicit theories of what works (and what does not work) in project interventions, often reveals fundamental differences of opinion among stakeholders about what a project or programme should be doing, why it should be doing it, and how we will know if it makes progress. Yet this paper, like most, really concentrates on assumptions about a situation - what else is going on in the environment, that might impede project implementation.


Without a discussion of assumptions about the development theory in project interventions, projects are unlikely to be learning events, unlikely to learn from failure or success, and unlikely to advance our understanding of development and what works.

[Update 2019:] Fortunately, SIDA has updated the resources it makes available on results-based management and these now include the 2016 update on the use of the LFA, referenced above, a 2014 handbook on the use of RBM in SIDA research cooperation projects, and a long, but interesting 2018 YouTube video discussion on how SIDA uses Results-Based Management.


The bottom line: This paper is moderately useful in its stated purpose - providing an outline of the Logical Framework Approach. It is most useful in discussions of problem analysis but there are more accessible guides to the process available.

_____________________________________________________________

GREG ARMSTRONG

Greg Armstrong is a Results-Based Management specialist who focuses on the use of clear language in RBM training, and in the creation of usable planning, monitoring and reporting frameworks.  For links to more Results-Based Management Handbooks and Guides, go to the RBM Training website. 

A Comprehensive Monitoring and Evaluation Website

--Greg Armstrong--

The MaNDE website provides a host of resources on training, evaluation methods, and results-based management.

[Edited to update links July 2018]

The Monitoring and Evaluation News

Level of difficulty:  Moderate to complex 
Primarily useful for: Field managers, planners, monitoring & evaluation specialists 
Limitations: A large, complex site, which will take some time to explore 
Length: Roughly 2,000 web pages and documents
Most useful sections: The "search" function, and the archives


The Monitoring and Evaluation News website (MandE) established by Rick Davies, in 1997 is one of the most useful resources available to development professionals interested in results-based management or other approaches to planning, monitoring and evaluation.  The site could be a useful professional resource regardless of readers' level of expertise - for complete beginners, experienced managers, or those interested in the theoretical underpinnings of monitoring and evaluation; whether readers work in the field or in an academic institution or donor agency headquarters.  


This website offers a wide range of materials of potential interest, including practical guides to results-based management and other forms of evaluation and monitoring, articles on the theory of performance measurement, reports from practitioners working in the field, notices of monitoring and evaluation training events, commentaries on donor policies and guidelines, and a wide range of discussions between development professionals on planning, monitoring and evaluation issues.


The Monitoring and Evaluation website is, itself, a blog -- something like Slate for evaluators -- with the wide variety of news and opinion it includes.  But readers can also get to Rick Davies' blog and website from there.

Organization of the site

The home page for the site may mislead a casual visitor.  The most prominent feature is a list of the most recent additions to the site, but on the right, there is a search box, and a long list of other available topics, both of which deserve a careful reader's attention.


The material on the Monitoring and Evaluation News is organized in archives, with 73 categories, when I last visited, including posts related to:

• training workshops  
• new books on monitoring and evaluation  
• a list of Monitoring and evaluation blogs  
• issues related to governance assessments 
• online training 
• impact evaluation  
• the Most Significant Change approach to monitoring and evaluation  
• quantitative and qualitative research methods   
• and a number of other topics, including, of course, results-based management and the Logical Framework Approach.

Most of the web pages the reader will initially find for these categories have been developed since 2008, and the reader can get a complete list of the contents of categories since 2008 using the "search" function.  But using the search box on the web site will also lead to a large number of useful tools produced before 2008.

Narrowing the Search for results-based management material

There are over 1,800 pages on the Mande site, and it can take some time to go through the material that is here. Not all of the material may suit your own needs, and some may be slightly outdated, but what this site does is to focus on performance assessment, monitoring, and evaluation, something that can narrow considerably the time we spend on searching for results-based management tools on the internet.


In the first post on this blog in early December of 2009, on how to search for RBM tools, I noted how many web pages you might get if you did a search for  "RBM" on Google, Yahoo or Bing.  At the time, the number of hits on Google was 296,000 pages - including a lot  not even remotely related to results-based management.  And if you were not careful, you could get over 100 million pages returned on a general search for results-based management.  But using the Mande search, you will get about 190 pages returned, all of them more or less relevant to what I, at least, am looking for.

The Logical Framework Approach: Documents on MandE

I will take one example, from one category here, to illustrate the range of materials available on the Monitoring and Evaluation news website.  If you search for "LFA" on Google, you will get close to 4 million pages listed - and not one in the first 100, when I looked today, was related to the Logical Framework Approach. Doing the same search at MandE, you will get 8 pages produced since 2008, and another 25 produced prior to 2008, all directly relevant to the Logical Framework.  


One of those is particularly relevant - The Logical Framework: A list of Useful Documents, produced in January of 2008.  This page links itself to 40 sites, in four categories:

• Explanations of the Logical Framework - containing a number of donor references and guides

• A Wider Discussion of Logic Models - some fairly complex pages discussing alternative approaches to the Logical Framework, and some online courses

• Critiques of the Logical Framework

• Alternative Versions of the Logical Framework - including discussions of Outcome Mapping, Appreciative Inquiry and other topics

• The Editor's Concerns (about uses of the Logical Framework) - Rick Davies' thoughts on the problems of using the LFA - and links to some alternatives.

The bottom line: You can spend a long time exploring the Monitoring and Evaluation News but it will be time well spent.


_____________________________________________________________

GREG ARMSTRONG

Greg Armstrong is a Results-Based Management specialist who focuses on the use of clear language in RBM training, and in the creation of usable planning, monitoring and reporting frameworks.  For links to more Results-Based Management Handbooks and Guides, go to the RBM Training website. 

Project Cycle Management, the LFA and RBM

An Introduction to Multi-Agency Planning Using the Logical Framework Approach

A practical guide to using the LFA in project planning and management.,
-Reviewed by Greg Armstrong-

[Review Updated March 2016, July 2018]

Level of Difficulty: Easy-to-moderate

Primary Useful for: Project field managers

Limitations: More detail on baseline would have been useful

Length: 58 pages

Most useful sections: p. 20-24 (risks and assumptions); p. 49-58 (LFA examples)

Problem Tree

This document, prepared in 2005 by Philip Dearden, Head of the University of Wolverhampton's Centre for International Development and Training, is a practical, hands-on guide to the use of the Logical Framework Approach in project planning. It is placed in the context of Project Cycle Management, but it is clearly relevant to results-based planning and management, which also uses the Logical Framework -- even if in reality many RBM practitioners stop with the "Framework", and forget the "Approach". In fact, there is not much difference between the Project Cycle Management and many forms of results-based management, which incorporate the Logical Framework -- except that Project Cycle Management emphasises stakeholder involvement in a way that RBM does in theory, but often ignores in practice.

Who the Guide is for

While I think anybody generally interested in results-based management could get some benefit from this guide, it will be most useful for project field managers and others trying to design a practical consultation process aimed at clarifying results, results chains and indicators.

Clear Language RBM

Most of the guide is easy to understand, using clear language to set out the basic steps in using the logical framework approach - something that should be central to the use of RBM. Where the UNIFEM guide, reviewed earlier, was essentially an introduction to RBM terms, this guide goes further, walking the reader through seven "core questions" beginning with a stakeholder analysis, and ending with a discussion of indicators.

Basic RBM Concepts

Philip Dearden and his colleagues have a long history of working in the field with development practitioners, and this is reflected in the way this document is written. It uses a simple, largely jargon-free approach to explain seven necessary stages in the Logical Framework Approach (LFA).

Participatory RBM Processes

The first section ("Who are we?") establishes the participatory nature of the Logical Framework Approach, something that really should apply to results-based management overall. Many of the criticisms I have seen and heard of RBM are that it is too top-down and technocratic in nature, but in fact it need not be this way. 

Stakeholder analysis

Bringing stakeholders into the process, from problem identification, through testing of assumptions, identifying potential results, and focusing on the most useful activities to reach results should be an essential component of RBM . In practice, however, because it takes time, effort, and money, participatory processes are often viewed as annoying distractions by many implementing agencies and donors. Talk about participation is cheap, but practice takes commitment.

Problem Identification

In a section called "Where are we now?", the guide discusses how to identify both the problem that the project or programme will address, putting it into context through a problem tree, and the potential strengths in the situation, those on which the project can build.

Maintaining a focus on the underlying problem is important. Many projects I have seen, bilateral and multilateral, essentially abandon the problem as the underlying foundation of the intervention once the activities are designed, and the projects become essentially activity-focused enterprises.

Clarifying Results-Chains

The section of the document "Where do we want to be?" focuses on choosing among broad results, then finding the short-term and mid-term results likely to contribute to what we want in the long run.

The term "Outputs" has been used differently in the past among different donor agencies - for some meaning essentially "completed activities" and for others "short-term results". In this section, Outputs are essentially defined as short term results - for example -- improvements in community capacity to manage activities and resources -- although later in the discussion of indicators, it labels them as "deliverables".

Defining Activities to reach results

The section called "How do we Get There?" ties the development of activities to the earlier analysis of problems.

And it really is important to go through the first three stages before identifying activities, because keeping in mind a clear idea of what the problem is, and what the logical sequence of possible results may be to the problem, tests the logic and soundness of any proposed activity. Unfortunately, in practice, again, the fact is that many organizations start with activities and then try to find a problem that might possibly be used as a justification for funding what they have already decided to do.

Risk Assessment and Assumptions

The section called "What may stop us getting there?" explains risk analysis -- discussing what potential problems could derail the project -- and the need to redesign activities to minimise those risks. This section also deals with the testing of assumptions -- how these assumptions need to be clarified and explicitly addressed in design of activities, and identifying the assumptions underlying the logic between short, mid-term and long-term results.

I found this section of the document very helpful. In my experience the great under-valued component of results-based planning has been the casual and obscure manner in which assumptions are often handled in the logical frameworks and, more importantly, in the discussion process which should precede the completion of the Framework. Seriously focusing on what stakeholders and implementing partners assume to be the relationships between activities and results, and about the underlying conditions necessary for solving problems, can reveal profound differences of opinion among these groups, not just about the political, social or economic conditions necessary to make a project work, but also about what types of interventions are likely to be most effective, about what our underlying, and often unspoken theories of learning and development are, and even assumptions about what the basic problems are, that the development activity will purportedly address.

Spending more time on clarifying these assumptions at the project design stage can prevent a lot of problems, and save considerable time and money during implementation, as, invariably, the different perspectives slowly start to surface. Of course, it is also important to reassess these initial assumptions at regular intervals as a programme or project evolves. When this is encouraged by donor agencies it makes it easier to make constructive adjustments to project management, perhaps even to the project's design, while always keeping an eye on the problem to be addressed. But this is something on which many donors, and most implementing agencies really don't want to spend time.

Indicators and Data Collection

Sections six and seven of the Guide ("How Will we Know if We've Got There" and "How do we Prove it?") deal with indicators and data collection. The indicator development discussion makes a useful distinction between indicators for completion of activities, and indicators for results.

The section on data collection deals with sources of data, and means of data collection. I think more space could usefully have been given to this, because the big problem in most indicator development is the impracticality of collecting data for many proposed indicators.

LFA Checklist

The concluding section provides a checklist of 29 issues against which to assess the utility of the Logical Framework for the project. Using the checklist without having reviewed the earlier text is probably feasible, but is likely to be considerably less useful than applying it after taking the time to review the rest of the document.

The appendices (p. 37-58) provide a glossary of terms, a list of advantages and disadvantages of using the Logical Framework for project management, a description of the Project Management Cycle approach, and a brief discussion of the purpose of monitoring and evaluation. But the most useful of the appendices may be the nine pages of examples of Logical Frameworks for the three children's projects in Sheffield, showing in detail how the indicators for these projects related to assumptions, results, and activities.

Limitations of the Guide

While the data collection discussion in the Guide does mention baseline data, it does not discuss it. This is the one area of the guide in which I think more detail would have been useful, even for beginners. It is something that all projects, and all donor agencies need to spend more time on at the beginning of the indicator development process.

Collecting baseline data is not necessary just for telling us if we have results --whether anything has changed -- but also for testing whether we can actually collect the data for the indicators we have agreed on. Yet, I have rarely seen international development projects, funded by any donor, where baseline information is actually collected even within the first year. Often, and this is no exaggeration - the baseline information is never collected - or it is collected retrospectively, or simply faked, three, four or five years after the project begins. This makes a mockery not just of the whole concept of "results-based" management, but also means that in a very substantial number of cases, only after the project has been implemented for years, does the realization hit home that many of the indicators are completely useless. If donors really took RBM seriously, and not just as window-dressing for their management committees or Auditors-General, they would insist that genuine baseline data -- and the consequent redesign of indicators, be completed before project activities are funded.

The bottom line: Overall, this is a practical and reasonably straight-forward Guide to the development of Logical Frameworks as part of the planning, monitoring and evaluation process. It is likely to be useful for many project field managers, in government, private sector or civil society implementing organizations.

More resources:

The Centre for International Development and Training now also offers an online course on RBM. 

_____________________________________________________________

GREG ARMSTRONG

Greg Armstrong is a Results-Based Management specialist who focuses on the use of clear language in RBM training, and in the creation of usable planning, monitoring and reporting frameworks.  For links to more Results-Based Management Handbooks and Guides, go to the RBM Training website.