Skip to main content

more options


The Flower and the Bee - Symbiosis and Co-Evolution

 

The ideas of symbiosis and co-evolution are critically important in evolutionary biology. One of the most familiar examples of this phenomenon is the relationship of the flower and the bee. Each provides something to the other. The flower provides nectar that is produced into honey, and the bee acts as the vehicle for plant sexual reproduction by moving pollen from one flower to another. Both benefit from the exchange. Neither participates in this exchange consciously. Flowers didn't strategize one day that they needed bees as a vehicle for reproduction. And bees didn't decide that flowers would be good vessels for honey production. They co-evolved over millennia in a manner that makes them co-dependent.

There are several ways that symbiosis and co-evolution are important for evaluation. First, if all programs evolve through different stages over time, then we must recognize that the evaluation approaches we use at each stage need to differ throughout the life of the program. That is, the way we would evaluate a program during its initiation stage would not likely be appropriate for evaluating it during its growth stage, and so on. In effect, the evaluation of a program has its own lifecycle and one of the major tasks of systems evaluation is to encourage the symbiotic or co-evolutionary relationship between program and evaluation lifecycles. In the initiation phase an evaluation needs to be dynamic and flexible, providing rapid feedback about implementation and process. In many program evaluations this is accomplished with simple monitoring or post-only feedback forms, unstructured observation, qualitative methods, informal debriefing and feedback, and through communications systems. In the development phase of an evaluation, the focus tends to shift to the observation and assessment of change and we focus on things like designing observational procedures and measures of key outcomes, assessing the consistency and construct validity of measures, looking at pre-post differences and examining the relationships among different observations, qualitative or quantitative. The mature phase of an evaluation tends to emphasize the idea of control. At this point the program is routinized and stable enough to compare performance of participants with some standard expectation of performance or with outcomes of people who participate in alternative programs or none at all. This is the realm of experimental and quasi-experimental designs and of more structured and comparative qualitative approaches. The translation or dissemination phase in evaluation is typically concerned with generalizability or external validity. It examines the consistency of outcomes across different settings, populations or program variations. This is the realm of secondary and meta-analysis and of program review approaches that seek general inferences about the transferability of the program. Encouraging a symbiotic relationship between the evaluation approach and the program lifecycle is a critically important systems evaluation process.

Second, the ideas of symbiosis and co-evolution also have important practical implications for the level of support people have for evaluation. In many evaluation contexts, one hears a series of laments about how unmotivated people are to evaluate or their resistance to doing evaluation. For instance, the evaluator asks "Why don't these program people just cooperate when I ask them for data?" Program implementers ask "Why don't these evaluations address something that would be useful for us?" Program participants want to know "Why do they keep bugging us for data? We don't get anything from this." In the ideal, we would want the situation to be a co-evolutionary one where program participants are providing information naturally as part of their participation, where program administrators are getting what they want from the provided data, and where evaluation happens almost transparently as an integrated aspect of program implementation. That is, the ideal is the flower and the bee. This is a difficult ideal to achieve in practice. It requires that the evaluation systems be engineered in such a way that each stakeholder group's incentive to participate in the evaluation is well understood.