Reid 2001

Reid, L. M. 2001. The epidemiology of monitoring. Journal of The American Water Resources Association 37(4):815-819.

An informal sample of 30 flawed monitoring projects was examined to identify the most common problems and to determine how they could have been prevented. Problems fall into two general categories: 70 percent of the sampled projects had design problems, and 50 percent of the sampled projects had procedural problems. Monitoring projects implemented by land-management agencies tended to have a higher proportion of procedural problems than did university-based programs (generally graduate student research), while the frequency of design problems was similar between agencies and universities. The most common problems were poorly trained or unmotivated field crews (37 percent of projects, a procedural problem), a sampling plan that was not capable of measuring what was needed to meet project objectives (30 percent, design), delays in analyzing data (27 percent, procedure), inadequate monitoring durations (27 percent, design), and absence of the collateral information needed to interpret results (20 percent, procedure). Most of the problems could have been avoided by submission of the study design to thorough technical and statistical review, active participation of the principal investigators in field data collection, and analysis of at least some of the data as soon as information was collected so that problems could be recognized early enough to be corrected.