Session 2 - Implementation and process evaluation

Session 2: Implementation and process
Neil Humphrey (Manchester)
Ann Lendrum (Manchester)
Process evaluation
Louise Tracey (IEE York)
Implementation: what is it, why is it
important and how can we assess it?
Neil Humphrey, Ann Lendrum and Michael
Manchester Institute of Education
University of Manchester, UK
[email protected]
What is implementation?
Why is studying implementation important?
How can we assess implementation?
Sources of further information and support
What is implementation?
• Implementation is the process by which an
intervention is put into practice
• If assessment of outcomes answers the question of
‘what works’, assessment of implementation helps
us to understand how and why
• Implementation science has grown dramatically in
recent years. For example, in the field of social and
emotional learning, an early review by Durlak (1997)
found that only 5 per cent of intervention studies
provided data on implementation. This figure had
risen to 57 per cent 14 years later (Durlak et al,
What is implementation?
Aspects of implementation
– Fidelity/adherence
– Dosage
– Quality
– Participant responsiveness
– Programme differentiation
– Programme reach
– Adaptation
– Monitoring of comparison conditions
Factors affecting implementation
– Preplanning and foundations
– Implementation support system
– Implementation environment
– Implementer factors
– Programme characteristics
See Durlak and DuPre (2008), Greenberg et al (2005), Forman et al (2009)
Why is studying implementation important?
• Domitrovich and Greenberg (2000)
So that we know what happened in an intervention
So that we can establish the internal validity of the intervention
and strengthen conclusions about its role in changing outcomes
To understand the intervention better – how different elements
fit together, how users interact etc
To provide ongoing feedback that can enhance subsequent
To advance knowledge on how best to replicate programme
effects in real world settings
• However, there are two very compelling additional reasons!
– Interventions are rarely, if ever, implemented as designed
– Variability in implementation has been consistently shown to
predict variability in outcomes
• So, implementation matters!
Why is studying implementation important?
Teacher-rated change: SDQ peer problems
Low PR
Moderate PR
High PR
Why is studying implementation important?
InCAS Reading
Low dosage
Moderate Dosage
High Dosage
How can we assess implementation?
Some choices
– Quantitative, qualitative or both?
– Using bespoke or generic tools?
– Implementer self-report or independent observations?
– Frequency of data collection?
– Which aspects to assess?
Some tensions
– Implementation provides natural variation – we cannot randomize people to be
good or poor implementers! (although some researchers are randomizing key
factors affecting implementation – such as coaching support)
– Assessment of implementation can be extremely time consuming and costly
– Fidelity and dosage have been the predominant aspects studied because they are
generally easier/simpler to quantify/measure. We therefore know a lot less about
the influence of programme differentiation, quality et cetera
– The nature of a given intervention can influence the relative ease with which we can
accurately assess implementation (for example, the assessment of fidelity is
relatively straightforward in highly prescriptive, manualised interventions)
• Think about a school-based intervention that you
are evaluating – whether for the EEF or another
• How are you assessing implementation?
• What choices did you make (see previous slide) and
• What difficulties have you experienced? How are
these being overcome?
• How do you plan to analyse your data?
• What improvements could be made to your
implementation assessment protocol?
Assessment of implementation – case study (PATHS
PATHS trial overview
Universal social-emotional learning curriculum delivered in twice-weekly lessons, augmented by generalisation techniques
and home-link work
45 schools randomly allocated to deliver PATHS or continue practice as usual for 2 years
c.5,000 children aged 7-9 at start of trial
Outcomes assessed: social-emotional skills, emotional and behavioural difficulties, health-related quality of life, various
school outcomes (attendance, attainment, exclusions)
Assessment of implementation
Independent observations
Teacher self-report
Teacher implementation survey developed following structure/sequence of observation schedule to promote comparability
Teachers asked to report on their implementation on each of the above factors over the course of the school year in addition
to providing information about the nature of adaptations made (surface vs deep)
School liaison report
Structured observation schedule developed, drawing upon previous existing tools
Piloted and refined using video footage of PATHS lessons; inter-rater reliability established
1 lesson observation per class; moderation by AL in 10% to promote continued inter-rater reliability
Provides quantitative ratings of fidelity/adherence, dosage, quality, participant responsiveness, reach and qualitative fieldnotes on each of these factors
Annual survey on usual practice in relation to social-emotional learning (both universal and targeted) to provide data on
programme differentiation
Analysis using 3-level multi-level models (School, class, pupil)
Plus! Lots of qualitative data derived from interviews with teachers and further quantitative data
on factors affecting implementation
Sources of further information and support
• Some reading
– Lendrum, A. & Humphrey, N. (2012). The importance of studying the
implementation of school-based interventions. Oxford Review of
Education, 38, 635-652.
– Durlak, J.A. & DuPre, E.P. (2008). Implementation matters: A review of
research on the influence of implementation on program outcomes and
the factors affecting implementation. American Journal of Community
Psychology, 41, 327-350.
– Kelly, B. & Perkins, D. (Eds.) (2012). Handbook of implementation science
for psychology in education. Cambridge: CUP
• Organisations
– Global Implementation Initiative:
– UK Implementation Network:
• Journals
– Implementation Science:
– Prevention Science:
Developing our approach to
process evaluation
Louise Tracey
Process Evaluation:
‘documents and analyses the development
and implementation of a programme,
assessing whether strategies were
implemented as planned and whether
expected output was actually produced’
(Bureau of Justice Assistance (1997)
(cf: EEF 2013)
Reasons for Process Evaluation
1. Formative
2. Implementation/Fidelity
3. Understanding Impact
Methods of Process Evaluation
Quantitative / Qualitative
 Observations
 Interviews
 Focus groups
 Surveys
 Instruments
 Programme data
Plymouth Parent Partnership: SPOKES
 Literacy programme for parents of
struggling readers in Year 1
 6 Cohorts
Impact Evaluation:
• Pre-test, post-test, 6-month & 12month follow-up
Plymouth Parent Partnership: SPOKES
Process Evaluation:
• Parent Telephone Interview
• Teacher SDQs
• Parent Questionnaire
• Attendance Records
• Parent Programme Evaluation Survey
SFA Primary Evaluation
Impact Evaluation:
• RCT of SFA in 40 schools
• Pre-test / post-test (Reception) & 12-month
follow-up (Year 1)
• National Data (KS1/2)
Process Evaluation:
• Observation
• Routine Data
Discussion Questions
1. What are the key features of your
process evaluation? Why did you
choose them?
2. What were the main challenges? How
have you overcome them?
Key Features?
Why chosen?
Key stakeholders
Inform impact evaluation
Main challenges?
How overcome?
1. Shared understanding with key
2. Reliability
3. Burden on schools
4. Control groups
5. Costs
Any Questions?
Thank you!
[email protected]

similar documents