Propositional Evaluation & Outcomes Assurance by Andrew Hawkins is licensed under CC BY 4.0


Propositional Evaluation

What makes doing that a good idea?

Politicians, public servants, and citizens need practical tools to design and evaluate action intended for the public good. This means focusing on what is useful to know and what can be known with any degree of certainty about the value of a proposed or current course of action. 

This website is concerned with the design, management, and evaluation of reasoned public policy and programs. It is aligned to systems thinking and situations in which experimental, realist or other approaches to evaluation for causal inference may not be necessary, possible, or represent the best use of resources at a time when decision making and design, or redesign is required. 

Propositional Evaluation is dedicated to evidence-based policy but shifts the focus of evaluation from a concern with 'what happened?', and even 'what works?' - towards 'what makes doing that a good idea?' 

This approach is based on an ancient tool for deductive logic combined with a specific theory of causality supported by scientific methods for gathering data. The first step is to treat a policy or program as a proposition or argument that sets out how it is intended to work (i.e how the premises in the argument support the conclusion). People can discuss (or evaluate) if this proposition makes sense 'on paper' before we attempt to see if it makes sense 'in reality'. People familiar with program logic will recognise the concept - Propositional Design Logic (PDL) - is a more explicitly logical form of the outcomes hierarchy with an explicit theory of causality focused on necessary and sufficient conditions. Yes, this is like program logic - but an explicitly logical and powerful version that speaks to anyone regardless of their technical expertise.

Propositional Evaluation is an approach to evaluation that is inclusive, credible, useful, and cost-effective. Outcomes Assurance is the associated practice for making it happen. An end-to-end process to manage risks to program failure - providing confidence that it will work, and assurance that it is working, or a rational means of choosing something else if it cannot, will not, or is not working.


Core concepts

parliament house.jpg

What is a proposition?

Making claims

What is Propositional Evaluation?

Testing claims

How do I get started?

An idea to a good idea in under two hours

Outcomes Assurance

Managing the risk of program failure

Three rings_edited.png

Strategy, Risk Management and Evaluation

Propositional Evaluation and Outcomes Assurance consider design, delivery, and evaluation of public policy as different ways of thinking about the same thing: ACTION

  • Strategy concerned with the future

  • Risk management with the present

  • Evaluation, traditionally with the past.

Of course, boundaries are fuzzy, arrayed as they are on a temporal dimension. Strategy is about developing a plan, risk management is about managing threats to achieving the objectives of that plan, and evaluation is traditionally about what happened or is happening, as a result of that plan. This can be described in some disciplines as 'Praxeology' - the study of purposeful human action.

Three key questions for this unified approach align with those of management theorists Drucker and Ackoff (their questions are in parenthesis and are rooted in the present tense*). 

  • What should be done? (and, are we doing the right things? i.e., effectiveness)

  • What is being done? (and, are we doing them right? i.e., efficiency) 

  • What has been done? (and, what makes these the right things to do? i.e., evaluation).

The theory of Propositional Evaluation and the practice of Outcomes Assurance sit in the overlap between past, present, and future. It is a unified approach to answering these great questions about any current or proposed course of action.

*English has two tenses for the present and six for the past, but none specifically for the future. So to say 'it works' or ask 'what works?' is ambiguous. Such a statement implies the future and refers to the present as though all causally relevant present conditions and relations will hold indefinitely. This is safe to assume in some situations, but less often in dynamic systems. It may also be confusing because physical science tends to search for time-invariant answers to causal questions, so tense is not that important in its language, but human affairs and social sciences are contextual.

It is possible that our lack of a future tense and inferences and habits of language used in the physical sciences has blinded us to the difference between 'what happened' or 'what is working', and 'what will work'.

If we state that 'it is going to work' or ask 'what is going to work?' it is very difficult to be credible without feeling the need to provide some reasons that start with an analysis of the current situation. Outside social policy it is hard to think of any realm of human endeavor where people claim a particular strategy 'works' (because something similar did in a past time and/ or place) without providing specific reasons related to current context: to provide reasons to support a current or proposed course of action.