Key Concepts
Anonymity – condition in which the person or item can never be identified by anyone by name.
Baseline, midline, endline – time periods in which data collection can take place to enable you to make comparisons.
- Baseline: time point before the beginning of the program or very early on in its implementation before any changes might have taken place. This is often referred to as pre-test or pre program data because you are measuring how things were before the program.
- Midline: time point during program implementation; after it is well underway but far enough from the end of the program or designated project time period so that any necessary “mid-course” corrections can reasonably be expected to have an impact on the program¡¯s results.
- Endline: time point either at the end of a program or specific time period, for example when funding has ended. This is often called post-test or post-program because you are measuring how things are after the program or intervention. It is also referred to as follow-up if done weeks or months after the program or exposure to the program has ended.
Biases – non-program factors that systematically alter the data you collect.
Confidentiality – condition in which the person or item can never be identified by people other than the data collector, including other members of the evaluation and program teams.
Comparison group – a group of people, documents, events that have not been exposed to the intervention or program but are similar to those exposed to the program in all essential characteristics.
Control group – a group of people who do not receive the intervention but who are similar in every important way to the group receiving the intervention. This is the ideal comparison group- and very difficult to achieve because it is very resource intensive. Some believe that it is unethical to have a control group- to withhold services from people who need them. If you have some evidence that your intervention provides a benefit. If this is the case, then you may consider having the control group receive the intervention at another time.
Data source – the kind of people, events and documents from which you will collect data (not the methods - such as a focus group - or the specific instrument - such as your interview guide) that you use to obtain the information.
Privacy – condition in which the person who is providing the data is not overheard by others or otherwise made uncomfortable about providing honest responses.
Population – all the people, events or documents in a particular group.
Reliability – the regularity with which a measurement will yield the same result if applied by different people or by the same person at close but different times.
Rigor – how an evaluation is designed so as to ensure greater validity and avoid or correct biases. Greater rigor produces more accurate and complete findings but almost always costs more in terms of time, personnel, data collection skills, and resources.
Sample – a subgroup of data sources that is similar to the entire population of possible data sources. It needs to be similar to the entire population on key characteristics.
Triangulation – collecting information on the same indicator from three or more sources to increase confidence in findings and understanding of the phenomenon.
Validity – accurate measurement of what you really and truly want to measure.
