In my last post I said that I would use this one to explain how to avoid common pitfalls when undertaking evaluations. So here goes …

1. Identify your sponsor. This is the person asking for the evaluation and has the authority to implement any recommendations. If the person you are working to is not the person who can implement your findings, you haven’t found your sponsor. What’s the point in carrying out work specified by a person who cannot do anything with it?
2. The list of ‘10 key questions to ask’ that I published in my last post have to be discussed and agreed with your sponsor. Without clarity on these answers you will not be able to create an appropriate methodology. You may want to produce lots of statistical data, but if that’s not what he or she requires, it’s pointless creating it.  This helps you get your methodology spot on.
3. When you are asked to evaluate something, remember that there will be a reason for it – and sometimes that reason will not be shared with you. Your sponsor may want the evaluation to prove something, they might want to use it to justify something or they may have already decided what your findings should be. So, what happens when your findings are different or you come up with the ‘wrong’ answers? What can happen is that the evaluator gets blamed for problem. You have used an inappropriate methodology or interpreted the data incorrectly or just not done what the sponsor asked. The focus is turned on the evaluator. So whenever you agree something with your sponsor get it signed – it is one of the many strands of life where having an insurance policy can be useful even though you don’t expect to need one. And this one costs nothing. And if the sponsor is reluctant to acknowledge his or her agreement to it, you perhaps need to be concerned at that early stage. Why are they reluctant? Do they actually know what they want?
4. Once the evaluator has chosen an appropriate methodology, it can be very easy to stray off the planned route. At regular intervals check the methodology still fits with the agreement you have with the sponsor. This will help you stay on track, and to appropriately amend the methodology should it be warranted.
5. Triangulation. Often you will find what appears to be a meaningful or important piece of data. If you are to publish it, I suggest that you triangulate it. That means you corroborate it from a different source. Using non-triangulated data can be problematic and may make your findings easier to challenge.
6. Consider letting the sponsor and key stakeholders see copies of your report prior to publication – assuming your agreement with your sponsor allows this. This is particularly useful if you have some controversial data. You may get asked to remove it – which you can then do if you think it appropriate – however, you will already have made your point to those people who matter.

By following this guidance you should avoid the common pitfalls, keep on track, protect yourself and be able to produce a very effective evaluation report.

Good luck!

Paul