Logic Model: A Program Performance Framework

1
Logic Model A Program Performance Framework

• Madison Workshop
• December 5, 2001
• Ellen Taylor-Powell, Ph.D.
• Evaluation Specialist
• UW- Extension-Cooperative Extension

2

• A New Era
• What gets measured gets done
• If you dont measure results, you cant tell
  success from failure
• If you cant see success, you cant reward it
• If you cant reward success, youre probably
  rewarding failure

3

• If you cant see success, you cant learn from it
• If you cant recognize failure, you cant correct
  it
• If you can demonstrate results, you can win
  public support
• Osborne and Gaebler, 1992 in MQ Patton, 199714

4
Logic Model is

• Picture of a program
• Graphic representation of the program theory or
  action what it invests, what it does and what
  results
• Logical chain of if-then relationships if
  x occurs, then y will occur
• Core of program planning and evaluation

5

• LOGIC
• Reasonable
• To be expected
• MODEL
• Represents reality, isnt reality

6
Why Logic Models--Why the Hype?

• Shows difference between what we do and impact we
  are having
• Provides a common vocabulary
• Focus on quality and continuous improvement

7
Logic Model Origins

• Private Sector Total quality management
• Public Sector GPRA, performance budgeting
• Non-Profit Sector Outcomes measurement, eg.
  United Way
• International Arena Results Framework of USAID,
  etc.
• Evaluators Evaluability assessment, bennett
  hierarchy

8
Logic Model
S I T U A T I O N
OUTCOMES
INPUTS
OUTPUTS
9
Everyday Logic Model
H E A D A C H E
Feel better
Get pills
Take pills
10
An Extension Example Business Counseling
Extension invests time and resources
A variety of educational activities are provided
to business owners who participate
These owners gain knowledge and change practices
resulting in
Improved business performance
11
LOGIC MODEL Program Performance Framework
12
Logical Linkages Series of If-Then Relationships

• IF THEN IF THEN
  IF THEN IF THEN
• INPUTS OUTPUT OUTCOMES

13

• Activity (p 2-12)
• Completing the causal linkages

14
LOGIC MODEL Program Performance
INPUTS
OUTPUTS
OUTCOMES
Activities
Participation
Short
Medium
Long-term
Program investments
What we invest
What we do
Who we reach
What results
SO WHAT??
15
Outcomes vs. Activities

• BE OUTCOME DRIVEN,
• NOT ACTIVITY DRIVEN

16

• Activity (p 2-13)
• Program Performance Levels

17
Situation

• Situational analysis
• Need
• Asset
• Problem analysis
• Priority setting
• Engaging others

18
INPUTS
Faculty Staff Money Time Volunteers Partners Equipment Technology
19
OUTPUTS What we do Who we reach OUTPUTS What we do Who we reach
ACTIVITIES Workshops Counseling On-farm research Facilitation Product development Curriculum design Trainings Field days Conferences Media PARTICIPATION Participants Clients Customers Users
20
OUTCOMES What results for individuals, businesses, communities.. OUTCOMES What results for individuals, businesses, communities.. OUTCOMES What results for individuals, businesses, communities..
SHORT Learning Awareness Knowledge Attitudes Skills Opinion Aspirations Motivation MEDIUM Action Behavior Practice Decisions Policies Social action LONG-TERM Conditions Human Economic Civic Environment
21
INPUTS
OUTPUTS
OUTCOMES
Parents increase knowledge of child dev
Design parent ed curriculum
Staff
Parents use improved parenting skills
Reduced rates of child abuse neglect
Targeted parents attend
Money
Provide 6 training session
Parents learn new ways to discipline
Partners
22

• Activity (p 2-24)
• Which are the Outcomes-Impacts?

23
Chain of Outcomes
24
Focus of Outcomes

• Child is prepared to enter school teen uses
  savings/spending plan
• Community group has inclusive membership policy
  family increases its savings
• Communications are more open agency adopted
  smoke-free policy
• Family serving agencies share resources
• Shared community responsibility has increased
  youth are valued as contributing members

• Individual
• Child, parent, client, resident
• Group
• family, team, community
• group
• Agency, organization
• System
• Community

25
How far out the outcome chain do we go?

• What is logical?
• What is realistic?
• What is meaningful?

26
INPUTS
OUTPUTS
OUTCOMES
Parents increase knowledge of child dev
Develop parent ed curriculum
Staff
Parents use improved parenting skills
Reduced rates of child abuse neglect
Targeted parents attend
Money
Deliver 6 interactivesessions
Parents learn new ways to discipline
Partners
Research
WHICH OUTCOMES???
27
Outcome of Interest?

• Inherently valued outcome (Mohr, 1995)
• higher level outcome is immaterial
• we are willing to assume that a higher outcome
  will also be attained if we achieve the outcome
  of interest
• Participant valued outcome if participants
  experience a change or benefit that makes a a
  real difference to them (United Way, 1999)

28
Assumptions

• Beliefs about the program
• the participants
• the way the program will operate
• how resources, staff will be engaged
• the theory of action

29
Assumptions, cont.

• Faulty assumptions are often the reason for poor
  results
• Check and test assumptions
• Identify potential barriers for each
  if-then sequence

30
EnvironmentInfluential Factors

• Extension program does not exist in a vacuum
• Context of the program
• politics, family circumstances, cultural milieu,
  demographics, economics, values, biophysical
  environment, policies, services
• What affects the program over which you have
  little control?

31
What does a logic model look like?

• Graphic display of boxes and arrows
• Any shape possible
• Circular, dynamic
• Relationships, linkages
• Level of detail
• simple
• complex
• Multiple models

32

• Compare examples

33
Logic Model Limitations

• Represents reality, not reality
• Programs are not linear
• Focuses on expected outcomes
• Challenge of causal attribution
• Many factors influence outcomes
• Doesnt address Are we doing the right thing?

34
Benefits

• Brings detail to broad goals
• Shows the chain of events that link inputs to
  results
• Builds understanding and consensus
• Identifies gaps in logic and uncertain
  assumptions
• Signals what to evaluate and when
• Summarizes complex program to communicate with
  externals

35
Building a Logic Model

• New program
• Existing program
• Team organization
• Involvement of others
• Keep it dynamic

36
Logic Model WORKSHEET

Program Title
Situation/Problem
37
Check Your Logic Model

• Are the outcomes really outcomes?
• Is the longest-term outcome
• meaningful?
• logical?
• realistic?
• Are the connections between inputs, outputs, and
  outcomes clear and reasonable?
• Does it represent research and best practice?
• Does it represent the programs purpose response
  to the situation?

38
OUTCOMES
INPUTS
OUTPUTS
Programmatic investments i
Short
Medium
Long term
Activities
Participation
39
Where does evaluation fit?
Providing Leadership for Program Evaluation
40
Evaluation Plan
41
Evaluation Questions What do you want to know?
42
Indicators - How will you know it?

• The evidence or measures that indicates what you
  wish to know or see
• often multiple indicators are necessary
• may be quantitative or qualitative
• culturally appropriate

43
Logic model with indicators for Outputs and
Outcomes
Outputs
Outcomes
Farmers practice new techniques
Farm profitability increases
Program implemented
Targeted farmers
Farmers learn
Number and percent of farmers attending
Number and percent who learned content
Number and percent who practice the
recommendations
Number of workshops held
Number and percent show farm profits increase
amount of increase
44
INDICATORS Examples
How would I know it?
45
Evaluating Your Program
, parents who increased knowledge ,
parents who learned new ways
Quality of curriculum sessions fully delivered
Decrease in rates of abuse neglect among these
parents
and parents who attended each session
, parents using improved skills (specify
skills)
46
Source and Method of Data Collection

• Source of information
• Participants
• Parents
• Teachers
• Key informants

• Method of collecting the information
• Survey
• Interview
• Observation
• End-of-program questionnaire
• Focus group
• Records

47
Evaluation Plan cont.
How will results be
Who will do what, when,
How will the data be
analyzed and
shared? To whom,
what resources?
interpreted?
how, when?
48
Enfin

• View evaluation as learning - integrate into the
  way we work
• Build evaluation in upfront
• Ask tough questions
• Make measurement meaningful
• Be accountable for highest professional standards