|
Analysis |
Broadening
Participation in Biological Monitoring: |
|
Planning analyses—Data are of little value if they are not used. Because participatory monitoring takes a lot of time and effort, all participants have a stake in the analysis and reporting of the data they have collected. Not doing so is a sure way to discourage participation in subsequent projects. Careful budgeting and foresight is required to ensure that sufficient resources exist in advance to complete analyses and report the results. Cost estimates can be obtained through consultation with a statistician. Having the analysis planned in advance will also ensure the sampling and project design will produce usable results. Hoping that support will come along later, if the data are simply collected now, is a risky strategy. Data analyses should be planned at the stage of sampling design so the intended use of the data is consistent with the statistical design of the sampling procedures, the measurements taken, and the format of the databases. Advance planning of the analysis not only speeds the process when the data are ready, it also averts the natural, but bias-inducing, inclination to simply explore the data for any meaningful results that might be found. For instance, searching for any significant correlations that might be found among multiple factors in a large dataset might seem to yield meaningful insights when, if enough combinations of factors are compared, some percentage of these factors are likely to seem statistically correlated simply by chance. Analyses can range in complexity from simply charting the data to the application of sophisticated statistical techniques. Any combination of valid methods that meet project goals and user needs is appropriate. Regardless of the selected analyses, it is essential that the individuals designing and conducting the analyses have an adequate understanding of the statistical methods involved. Even experienced scientists routinely consult with statisticians, and a participatory monitoring project should be no different in this regard. Reviewing and interpreting results—Once data have been analyzed, results are usually subject to interpretation. Such interpretation typically is more informed and objective if approached from the multiple perspectives that participants impart. Collaborative interpretation of results also contributes to the on-going process of enhancing mutual trust. Involving participants in the interpretation of results need not be a difficult process, if for example, the findings are presented graphically as well as numerically and discussed in a group setting. Reaching collaborative interpretations, especially if consensus can be attained, does much to prevent disagreements or controversy that might arise from the way managers or policy makers use the results. If criteria for reaching stipulated conclusions are agreed to in advance, then the data analysis can speak for itself, reducing the likelihood of dissention about the meaning of the results.
|
|