Hi Lindsay
Welcome to the RAMESES discussion list and thanks for your question(s).
I'll have a go and then some others can share their opinions too.  My
responses would go like this:

a) whether individual characteristics need to be accounted for in program
theory depends in part on the nature of the program, and in part on whether
you mean 'theory of action' or 'theory of change'.  If you intend 'realist
program theory' in the Pawson and Tilley sense, then you can't develop it
without individuals - see next point for why - but the issue is the level of
abstraction at which one works ('each individual' or 'aggregations of
individuals with something in common')

b) The Pawson and Tilley sense of 'mechanism' involves an interaction
between the resource(s) or opportunities that the program provides and the
REASONING (a catch-all term for anything that happens inside their heads -
logic-in-use, affect, etc) of those whose decision-making the changed
outcome depends upon.  I suspect you've misread Mark Pearson slightly
because realists are generally very clear that mechanism is NOT the program
itself, nor a particular action or strategy within a program.  A mechanism
involves INTERACTION between program and target.  I describe 'description of
a mechanism' as requiring three elements: the resource or opportunity the
program provides, the changed reasoning on the part of the target, and the
process by which the changed reasoning leads to changed behaviours which
generate different outcomes.  Note here however that mechanisms might be
fired at many stages of an intervention process (for example, the first
people whose decision-making is changed may be service providers rather than
clients), and that they are likely, according to realist theory, to be
different in different contexts or for different sub-groups. 

c) You're absolutely right that individual factors can be context in one
situation, mechanism in another, and outcome in a third.  The trick here is
to work out the level at which a particular piece of analysis is being done,
and to work 'backwards from outcome'.  That is: a mechanism is, by
definition, whatever generated the outcome, so you can't really identify
mechanisms OTHER than in relation to the outcome they generate. Mechanisms
only operate in particular contexts, so you can't work out 'in which
contexts' they work without knowing which mechanism you're talking about.
Whichever level your outcome is on, (internal to the individual, behaviour
of individuals, change of outcome as a result of changed behaviour,
behaviour of organisation or system - whatever), the mechanisms will be at
either a higher or lower level of the system(s) - causation works both
upwards and downwards.  Anyway - the 'logic' of working out whether the
individual factor is C, M or O is to first decide what the "O of interest
for this bit of the analysis" is, and then work back from that.

Over to others!
Cheers
Gill

-----Original Message-----
From: Realist and Meta-narrative Evidence Synthesis: Evolving Standards
[mailto:[log in to unmask]] On Behalf Of Lindsey Pike
Sent: Tuesday, 25 October 2011 5:46 AM
To: [log in to unmask]
Subject: Where do individuals fit in the C-M-O model?

Dear RAMESES group, 

I'm a PhD student  investigating how to maximise the effectiveness of
safeguarding adults training. I am attempting (perhaps foolishly) to carry
out a realist synthesis to help answer this question.

I've already carried out an integrative review of the training transfer
literature, which supports the notion that 3 main factors- individual
characteristics, training design and delivery, and workplace factors, affect
training transfer (how much learning is applied on the job) generally. I
want to specify the review to health and social care training (I can only
find 1 paper evaluating safeguarding adults training, so can't go that
specific) using realist synthesis.

My question is: where does the individual fit in the context- mechanism-
outcome model? The transfer literature states that individual
characteristics (things like motivation, self efficacy, locus of control,
personality, cognitive ability) are an important factor to transfer, and
presumably this means characteristics of the individual must be accounted
for in programme theories (or does it?). Additionally McCrae and Banerjee,
on evaluating mental health interventions, state "Programmes do not 'work';
people make them work" and I think this rightly applies to training; as the
saying goes you can lead a horse to water but you can't make it learn and
subsequently apply its learning to practice.  

Policy, on the other hand, makes no mention of individual characteristics
other than, to paraphrase, "staff must be trained".

The other argument, articulated by Mark Pearson recently on this thread
states that the mechanism is not the reaction to the intervention but the
intervention itself*; this implies (if I've understood it correctly) that
individual characteristics, which influence reactions to interventions,
can't be part of programme theories. So something like "motivation" cannot
be a mechanism, as it could be interpreted as an individual characteristic
and a reaction to training. 
This brings me to another issue, which is that factors such as motivation,
which are hailed as important in the training transfer literature, could be
seen as context, mechanisms, or outcomes; e.g. 

Context: whether delegates are motivated to learn, attend training, and
transfer training or not- affected by workplace systems, how training is
framed etc (accordingly maybe the workplace systems that affect motivation
are the context, rather than motivation itself?)
Mechanism: training results in better services by increasing staff
motivation to perform, which leads to improved (e.g.) prevention and
detection of abuse
Outcome: One purpose of training may be to increase motivation, through
raising awareness of safeguarding related issues/ valuing staff by providing
them with continuing professional development opportunities

What I'm trying to ask is,
- is possible/ sensible to account for individual characteristics in
programme theories?
- do you have any thoughts about whether they could or should be entered as
context, mechanism or outcomes?
Or would it be better to focus on the training design and delivery, and
workplace factors?

Thanks very much in advance for your thoughts.

*In context- Mark said in reference to the Access to health workers in rural
and remote areas review, "Conceptualisation of mechanisms- people's
reactions to programmes are conceptualised as 'mechanisms' throughout the
review (x leads to y in context z) rather than mechanisms being something
internal to people that is caused to 'fire' (in certain contexts..) by the
programme. So, in my understanding, just as a bullet fired from a gun is not
a mechanism, neither are people's reactions to a programme."

Best wishes,
Lindsey

Lindsey Pike
PhD student
School of Social Science and Social Work, University of Plymouth
[log in to unmask]
07814 843903
________________________________________