- Monitoring and Evaluation NEWS - https://mande.co.uk -

The Logical Framework: A list of useful documents

Tweet [1]

Contents: 1. Explanations of the Logical Framework [2] | 2. Wider discussions of Logic Models [3] | 3. Critiques of the Logical Framework [4]4. Alternative versions of the Logical Framework [5]5. The Editor’s concerns (about uses of the Logical Framework) [6] | 6. Online survey on views and usage of the Logical Framework [7]

Please feel free to suggest additions or corrections to this list, by using the Comment facility at the end of this post

New section: Software

1. Explanations of the Logical Framework

2. Wider discussions of Logic Models

3. Critiques of the Logical Framework

4. Alternative versions of the Logical Framework

5. The Editor’s concerns (about uses of the Logical Framework)

  1.  Long, complex, unreadable sentences, in the narrative column of the Logical Framework
    1. Often the result of compromises between many different parties who have been negotiating the contents of the Logical Framework. Net result: an unreadable document
    2. Sometimes the result of people not knowing that the whole story does not need to be told in one sentence. The row below should say what happens before (the cause) and the row above should say what happens next (the effects)
    3. Sometimes the result of people forgetting there is a column for indicators next door, where they can provide lots of interesting detail about what is expected to happen at this stage
  2. Narrative statements without people in them. E.g “Rice productivity increased”
    1. Another reason some many Logical Frameworks are so unreadable, and so boring when they are readable, is that somehow their authors have managed to leave out people. Instead we have lots of abstract and disembodied processes. And then we wonder why some people have difficulty understanding Logical Frameworks
  3. Means of Verification that refer to reports and surveys, but not who is responsible for generating and / or providing this information (and when it will be available)
    1. This problem is similar to the above, reflecting a continuing aversion to making references to real people in Logical Frameworks.
    2. One consequence is lack of clear ownership and responsibility for M&E of the changes being described at that level of the Logical Framework
  4. Insistence on there being only one Purpose level statement in a Logical Framework
    1. I have recent experience of colleagues insisting on this. For reasons I have not yet established, beyond the “it is not allowed” variety. Insisting on one Purpose and One Goal really is pushing a very linear model of reality. It does not even allow for any parallel but convergent events, such as those usually come through problem tree analyses that sometimes precede the design of a Logical framework
  5. Ambitous narrative statements coupled with modest indicators / Overly simple indicators used to describe complex developments.
    1. Such as “number of meetings held” as an indicator for the functioning of stakeholder’s advisory committee. For an alternative, see “Checklists as mini-theories of change [87]
  6. Lists of indicators in no apparent order
    1. “A (unsorted) list is not a strategy” A sorted list can convey relative importance (most important indicator at the top), or a sequence (starting from the bottom), or multiple alternative routes to the objective in the narrative column. If there is a list, the reader should be told what sort of list it is.
  7. Broad generalisations at the Goal level
    1. Sometimes arising from confusion of a temporal hierarchy (A leads to B which leads to C which leads to D) and a nested hierarchy (A is part of B which is part of C which is part of D). The Logical Framework is supposed to be a temporal hierarchy, that tells a story. Not a pile of increasingly broad statements about the same thing
  8. Confusion over the meaning of different levels in a Logical Framework. Between Activities and Outputs, Outputs and Purpose level outcomes, and outcomes at the Purpose and Goal level.
    1. Often cause by leaving people out of the picture, as above.
    2. A workable rule of thumb, for seperating levels of the Logical Framework
      1. Activities are things that “the project” can control. The boundary of a project being defined by the reach of its contracts (with staff, consultants, suppliers and sub-contractors)
      2. Outputs are the activities of the project (if services), or their results (if goods), that people and organisations outside the project can use e.g workshops, publications, trainings, etc. Ask here: What is available to who, and in what form?
      3. Purpose level changes (outcomes), are changes in those people or organisations who have used those goods or services. Normally the project would hope to influence these (and learn about how it can have influence) but it would not be expected to control events at this level
      4. Goal level changes (outcomes), are longer term changes in those same people or organisations, or others they have subsequently interacted with.
  9. Long lists of assumptions
    1. Apparently designed to cover people’s backsides
    2. Including many events that the project should be able to influence
      1. …which therefore should be listed as one of the outputs or outcomes. I.e. brought into the central narrative of the Logical Framework
  10. Things the Logical Framework cant do very well, even in the best of hands
    1. Represent multiple parallel processes, as distinct from a single process
      1. E.g. What people are doing at multiple project locations, within a single national project
        1. Representing their interactions is even more of a challenge
    2. Represent the interactions between multiple events at the same level of a Logical framework.
      1. E.g. How different project outputs (manuals, training events, newsletters, websites, etc) feed into each other
      2. Or, how different health outcomes (at Purpose level) feed into each other, before finally contributing to Goal level changes e.g. reduced mortality
    3. Represent the interactions between multiple outputs and the many users of those outputs
      1. E.g., the range of communications products used by a range of clients of a project . Many people will use multiple products, but their usage patterns will vary. Many products will be used by multiple users, but their user groups will vary.

All these processes can be represented by network models. See the new page on developing network models of development projects [88]. However network models are generally too complex to provide a substitute for the Logical Framework. One proposed alternative is the Social Framework, originally described here [74] and now updated here [89]. The Social Framework can be used to describes a pathway through a network, in a way that capable of being monitored and evaluated. Your comments are welcome.

6. Online survey into the uses of the Logical Framework

Please consider taking part in this survey [90]. You can access the cumulative results to date at the end of the survey form. It is not long.

thanks, rick davies