Recap: “We soon discovered that we did not have enough data to do a quantitative evaluation and thus had to focus on a more qualitative one. The next few days involved writing up questionnaires and contacting as many stakeholders as we possibly could in order to set up interviews to ask them questions about the organization and their relationship with it.”
There are a number of different ways to do evaluations. Utilization evaluation is an approach that constructs evaluations based on its usefulness to its intended users. This is what we aimed to do with our evaluation of RealStart – we wanted to ensure that our evaluation proved useful to the organization. In order to do this, we focused on RealStart’s Program Theory: is the organization and operation of RealStart conducive to achieving its intended mission and vision? To do this, we created a series of surveys designed to target this central question.
We assembled the contact information for everyone involved with RealStart: donors, board members, staff, former students and current students. While we administered similar survey questions to everyone, we also tailored certain questions designed to measure a specific aspect of the program for the different surveyed parties. In order to assess RealStart’s Program Theory we needed to construct survey questions that could “compare” responses among the different groups to determine the cohesiveness of the organization. Through a comparative analysis we also sought to determine whether or not the program was having the intended effect on its recipients that its mission articulated. To do this, we asked staff, former and current students similar questions regarding specific aspects of the program. We then compared staff responses with student responses as well as analyzing the difference in responses among former and current students.
Qualitative Analysis taught us a lot to say the least. Given the lack of feasibility of mass communication for some of the former students, we literally had to physically track them down in the city. While interviewing some of these students, we also had to jump over a few communication hurdles. The students we interviewed came from either Xhosa or Afrikaans backgrounds. While they spoke fluent English, we found that certain ideas weren’t communicated through our survey questions. As these survey questions were administered through oral interviews, we had the opportunity of re-wording our questions when they didn’t understand what we were asking. We realized, though, that produced bias in our interviews. If we didn’t ask all of our interviewees the same questions verbatim, could we expect to generate unbiased responses?



