They are asking you how evaluation of a program already in existence can lead toward that program's improvement.
Here's an example:
In 2000, I worked at a high school that had been given a state grant to greatly expand the "technology" in the school. In writing the grant application, we had written that there was not enough computer access for students. (Plenty of computer access for office people, but very little for teachers and almost none for students.)
We ended up putting a teacher station and at least two student stations in each classroom, depending on what was being taught in that classroom. (Drafting/CAD classrooms received more computers than English classrooms, for example.) We also ordered two mobile labs, each of which consisted of 30 laptops, a printer, and a huge cart on which these could all be stored and then moved to different classrooms on a sign-up basis. There were many other elements in our plan, too.
After a year, we held a formal evaluation of the system so far, starting with a written survey from all the teachers on campus and moving on to individual interviews and faculty meetings. What we discovered was that just about everything was being used well except for the mobile labs. It turned out that only about 4 or 5 teachers (out of a faculty of about 130) figured out how to use them effectively in their classrooms, and the rest didn't want the hassle.
What do you think the conclusion was about just that part of the program?
Evaluation includes whatever ways you can find to think about what you've done, what is going wrong, what could be better, what needs to be fixed now, what can wait, etc.