One of the most frequent lost opportunities when conducting an evaluation is not setting detailed enough expectations and targets when you begin. By setting quantitative goals ahead of time, all parties will enter the evaluation with the same outcome expectations and therefore when the results come in, there will not be disagreement with the findings. I made this mistake with my first evaluation.
© Olivier Le Moal – Fotolia.com
I conducted that evaluation at my first job after one of our lay leaders was interested in what the results of our work. I arranged a conference call and we decided on overall goals which were later emailed around for approval. However, as I began to conduct evaluation, I then had to turn the goals into benchmarks. That is where the challenge began. Would success be at 75% of 90%? Was 3 positive interactions a year good or should that number be higher? All of these were questions I was faced with.
I remember presenting the data to the department and lay leaders. There was a lot of discussion during the presentation about what the data meant and one of the lay leaders told me that my expectations were different than hers. While we set goals as a group because I then assigned the benchmarks on my own the findings were able to be ignored because benchmarks were not agreed upon while the underlying data was still accurate and findings could be adjusted.
There is another reason to set goals before the analysis begins. Once you see data, any prediction you make may be biased as a results of the results. Take a second now to think about the work you are involved in. Do you think you are doing a good job? If the answer is yes, then unconsciously you will alter you expectations to align with the results if they aren’t documented beforehand so the evaluation proves what you believed going into it.
Lessons:
- When discussing goals before a survey, be sure to set quantitative benchmarks you would like to meet and then use the evaluation to see where you currently are.
- If you design new metrics which were not initially discussed, before you present the data you should have a discussion and set a benchmark then and then show the result. As the facilitator, you should not influence their decision since you likely have likely seen the result.