Credibility of Evaluation: The Kenyan case