24/05/2018, 21:32

Program Evaluation-Part 1 CITAR Workshop July 28-Aug 1, 2008 Hue City

Definition of program evaluation “…the systematic collection of information about the activities … and outcomes of programs to make judgments about the program, [ ...

Definition of program evaluation

  • “…the systematic collection of information about the activities … and outcomes of programs
  • tomakejudgments about the program,
  • [in order to]improve program effectiveness, and/or informdecisions about future program development” (Patton)

Part 1: Objectives Participants will be able to

  • Define program evaluation
  • Describe 2 sources of failure of a program to achieve a population impact
  • Distinguish the questions, methods, and standards of comparison of the 1st level of evaluation from those above it in the hierarchy:

1) concept and design evaluation

2) process evaluation

3) outcome evaluation

4) economic evaluation

Potential Sources of Failure to Achieve a Population Impact

Concept & Design Evaluation

  • Step 1, Describe the program or policy
    • Problem it is supposed to solve
    • “Target” population
    • Objectives
    • Program theory – Methods
    • Delivery strategies
    • Resources required
    • Stage of development
    • Organizational and political context
  • Step 2 – Assess the need for the program
    • What is the evidence that the problem is “real” and worthwhile trying to solve or reduce?
    • Are the appropriate population(s) being targeted?
    • What standards of comparison might be used here?
      • Expectations of funders
      • Expert opinion
      • Social justice
  • Step 3 - Assess the program theory
    • Do the change objectives reflect validassump-tionsaboutthecauses of the problem?
    • Are the interventionmethodsappropriate?
    • Will the planned deliverystrategies reach the target population?
    • Are the planned resourcesadequate to support the program and its delivery?

What standards of comparison might we use?

  • Expert opinion Past achievements Theoretical standards
  • Norms Professional standards

Use a logic model

  • A systematic, visual way to present and communicate an understanding of the relations among the resources for operating the program, the activities that comprise the program, and the intended outcomes
  • Shows how the program is supposed to reach the outcomes
  • Provides the basis (logic, theory, evidence) for expecting these outcomes

Basic Logic Model Template

Inputs

What resources are needed for this program to operate?

  • Funding
  • Personnel
  • Materials
  • Equipment
  • Relationships
  • Others?

Activities

  • What events or actions are planned to take place to create the desired results?

Outputs

Impact of the program

  • Short-term
  • Intermediate
  • Long-term

Tips for Writing Logic Models

  • Be specific

Knowledge of what? Which beliefs? skills?

  • Work both left and right

Even if you start with a program, specify the problem at the far left before you start to work back and forth between them.

  • Modify the form – columns

You might need more categories to convey the causal chain of hypotheses.

Approaches to developing a logic model

  • No right way
  • Clear definition of problem should be done before starting
  • Forward (left to right)
    • Ask the question “But why?”; or
    • If…. Then….
  • Reverse-
    • Starts with desired outcomes
    • Asks the question, “But How?”
  • Modify the form - rows Use rows & partial rows to separate causal chains; e.g., unintended pregnancy & STDs don’t belong on the same row if main intervention is oral contraceptives
  • Recognize differing perspectives

Usually there are >1 possible logic models for the same problem

Logic Model: Source of Causal and Intervention Hypotheses

  • Causal Hypothesis: What is the program’s underlying hypothesis about the “cause(s)” (or determinants, risk factors) of the problem?
  • Intervention Hypothesis: What are the program’s underlying hypotheses about the “cause(s)” of the behavior and what will change those “causes”?
    • The “Logic” of the Causal Hypothesis

The logic or rationale for a program’s causal hypothesis should be developed from:

  • Epidemiological evidence
  • Theoretical evidence
  • Evidence from other programs
  • Qualitative data

Causal Hypothesis: Examples

Unprotected sex among adolescent girls

What are some of the “causes” (i.e., risk factors or determinants)?

  • Too much sex in the media
  • Girls don’t know the risks
  • Condoms are not available
  • Girls want to get pregnant
  • Welfare supports pregnancy as a career
  • Poverty supports early pregnancy as normative

The “Logic” of the Intervention Hypothesis

What is the evidence for this intervention hypothesis?

  • Focus groups
  • Key informant assessment
  • Formal behavioral theory
  • Consensus-based theory
  • Previous evaluation results
  • Evidenced-based guidelines for practice
  • “Intervention Mapping” is one approach which uses theory and evidence to develop an intervention program

Intervention Hypothesis: Examples

  • Too much sex in the media
    • Movie ratings, other controls reduce media exposure
    • Promotion of role models who don’t have sex
  • Girls don’t know the risks
    • Provide information about the risks by peers
  • Condoms are not available
    • Make condoms more accessible to girls
    • Boys?
  • Girls want to get pregnant
    • Promote alternative norms for relationships
    • Provide experience with the responsibilities of parenting
  • Welfare policy supports pregnancy as a career
    • Reform welfare policy
    • Promote plausible “futures” for girls
    • Improve schools
  • Poverty causes girls to see early pregnancy as normative
    • Mixed income housing

This afternoon’s exercise

  1. What problem is addressed?
    • What evidence is presented that this problem affects the groups “targeted” in this study?
  2. What are stated or inferred causal Ho?
    • What evidence is presented?
  3. What factors/”determinants” are addressed?
    • What evidence is presented?
  4. What are stated or inferred intervention Ho?
    • What evidence is presented?
  1. How well would these causal hypotheses fit Vietnam?
    • What evidence would support your opinion?
  2. How well does the profile of factors or ‘determinents fit Vietnam?
    • What evidence would support your opinion?
  3. How well would these intervention hypotheses fit Vietnam?
    • What evidence would support your opinion?

Click here to download slide powerpoint

0