In-person & Synchronous Online, Spring, Second 5 weeks, 1 Credit, Tu/Th 11:30am - 12:45pm

Instructor: Carlo Salvinelli

Course Information

This course will review the characteristics of the study designs that can be used to assess the impact of international development interventions. Students will be introduced to the main types of study design, including observational and experimental studies. Emphasis will be placed on Randomized Controlled Trials and Interrupted Time Series analysis through the review of two case studies. Ethical consideration faced when conducting research on human subjects and the compliance process to obtain approval from the Institutional Review Board will be presented.


Learning Goals

  1. Be familiar with the concept of theory of change and the indicators used to measure progress towards an intervention's outcome achievement.
  2. Be able to explain most commonly used impact evaluation designs and understand the conditions under which each may be used.
  3. Be aware of the ethical issues that need to be considered when running impact evaluations.

Textbooks and Materials

There is no required text. Readings will consist of textbook chapters, peer-reviewed journal articles and reports. All required readings will be provided through Canvas. Check regularly on Canvas for information about lecture content, readings and assignments.

Course Calendar (subject to change)




Introduction to Impact Evaluation

Study Designs


How to implement an Impact evaluation

Ethics / IRB


Sample size calculation

Colorado WASH Symposium


Randomized Controlled Trials - Rwanda Case Study

Quasi-experimental designs - Sierra Leone Case Study


Impact Evaluation: A Cautionary Tale



Attendance/Class participation (20% of final grade): Attendance is required. Class participation is an essential component of the requirements for successful completion of this course. Points for class participation are not simply guaranteed by attendance; participation consists of being present in class, reading all assignments prior to the beginning of class, and being an active member in class discussions and presentations.

Study design assignment (20% of final grade): Students will analyze four different scenarios related to the USAID/MCC funded BRIGHT program and identify the best study design for each scenario.

Ethics/IRB assignment (20% of final grade): Complete the CITI Program "Social Behavioral Research Investigators and Key Personnel" course.

WASH Symposium assignment (20% of final grade): Students will attend at least 2 sessions of the CU WASH Symposium that discuss a topic related to this class content (study design, impact evaluation, data, evidence based decision making...). and will write an essay about the presentation/discussion attended.

Impact Evaluation assignment (20% of final grade): Students will analyze a real impact evaluation and will assess the appropriateness of the methodology and the effectiveness of the study.

Late Policy

The grade for late assignments will be reduced by 10 percentage points (on a 100 point scale) per day that the assignment is late. We may agree to make exceptions to this policy on a very limited basis, provided that the reason is very compelling and provided that you ask us in advance.

Extra Credit (TBD points)

Extra credit assignments are not anticipated but may arise as the course progresses. 


There will be no exams in this course. 

Grading Scheme

Grades will be recorded in Canvas throughout the semester. At the end of the term, the scores on all assignments are weighted by the percentages given above to determine a semester score. Student grades will be determined as follows based on their semester score rounded to the nearest single decimal place:

A+: 100 to 97.0 percent

A: 100.0 to 93.0 percent

A-: 92.9 to 90.0 percent

B+: 89.9 to 87.0 percent

B: 86.9 to 83.0 percent

B-: 82.9 to 80.0 percent

C+: 79.9 to 77.0 percent

B: 76.9 to 73.0 percent

C-: 72.9 to 70.0 percent

D+: 69.9 to 67.0 percent

B: 66.9 to 63.0 percent

D-: 62.9 to 60.0 percent

F: 60.0 to 0 percent