Evaluating Development Cooperation/Standard Evaluation Methods/Data Analysis Methods

From Wikibooks, open books for an open world
Jump to: navigation, search

Data Analysis methods
Standard Evaluation methods

It is very important to link evaluation of managerial processes with the evaluation of the programme/project outcomes. For a manager to learn means to understand the relationship between processes and outcomes since s/he can achieve outcomes only by improving the management of processes. Although there are different indicators and different information sources for monitoring and evaluating processes (efficiency, accountability and transparency , etc.) and outcomes (effectiveness, sustainability, impact), real understanding can be done only integrating the two dimension in a unitary vision of necessary relationship amongst resources, activities, results, objectives and impacts. The central element is always organization development and the empowerment of the human resources learn to perform actions in a better way so as to achieve better results and obtain better outcomes.

Vrinda here please re-write the following para to keep them logically related to the lines above also verify the coherence with the links above.

Questions to Ask of Intended Users to Establish an Evaluation’s Intended Influence on Decisions

  • What decision, if any, is the evaluation finding expected to influence?
  • When will decisions be made? By whom? When, then, must the evaluation findings be presented to be timely and influential?
  • What is at stake in the decisions? From whom? What controversies or issues surround the decision?
  • What’s the history and context of the decision-making process?
  • What other factors (values, politics, personalities, promises already made) will affect the decision-making? What might happen to make the decision irrelevant or keep it from being made? In other words, how volatile is the decision-making environment?
  • How much influence do you expect the evaluation to have – realistically?
  • To what extent has the outcome of the decision already been determined?
  • What data and findings are needed to support decision making?
  • What needs to be done to achieve that level of influence?
  • Who needs to be involved for the evaluation to have that level of influence?
  • How will we know afterward if the evaluation was used as intended?

In general, the overall approach and methods selected for an evaluation should involve beneficiaries and the other relevant stakeholders in the design and implementation of the evaluation. (see Stakeholders Identification - Involving Stakeholders ; Participation Methods and Tools)

The approach and methods used in data analysis will be determined by the reasons for which it is being undertaken, who is taking responsibility and who needs to be involved.


Analysis may be described as:

Seeing how things are related or what triggers what?

Makings sense of the sequence of events, information to make it applicable

The following key questions may be useful to research teams:

  • What does it mean? (For instance if a respondent says: “I am poorer than I was last year.” What does it mean? What was the situation before? What has changed? Why has the change affected him/her? Who else is in a similar situation?
  • Why? Why is the respondent saying that he is poorer this year than last year? Try to ask why seven times.
  • What are the consequences/ implications of the information collected? What is the perspective of the respondents on this? What is the researchers interpretation of the said consequences?
  • What are the unexpected? What in the findings is contrary to “normal” expectations? What are the contradictions between what is expected and the reality on the ground.
  • What is unique? Researchers should ask what makes the respondent, subcounty, district etc…different from the others. What are the unique features?
  • What is important? This refers to prioritization. What are the key issues that are coming out?

You can take ideas from the most utilzed approaches:

  • Outcome mapping - outcomes as behavioral change
  • The ‘Most Significant Change’ (MSC) Technique
  • Participatory poverty assessment Methodology - Analysing findings from semi-structured interviews
  • Methods for understanding the area around us (analysing space);
  • methods for analysing time
  • methods for prioritising and ranking
  • Livelihood analysis
  • Methods for gender analysis
  • Do the analysis as and when information becomes available. The longer you postpone doing the analysis the more difficult it becomes.
  • Do analysis collectively to guard against biases. Two heads are better than one. Putting heads together leads to better quality analysis.
  • Develop a working hypothesis as you go along and revise the checklist accordingly.
  • Use all opportunities for cross-checking. If necessary go back and reinterview the same respondent.
  • Have a holistic view. Link questions and answers. Have the research question in mind all the time. Look for the relationships between different responses.
  • Draw preliminary conclusions with the respondents. Use triangulation, observation and ask more than one question all the time on the same issue.