Statistical analysis

Abstract

Effective monitoring of populations is critical for assessing the efficacy of conservation and management activities, determining the conservation status of a species, and examining trends over time. However, far too often, monitoring efforts fall short of adequately describing the population. Poorly defined problems and objectives, a lack of standard operating procedures, a poor selection of attributes to monitor (We define attributes as characteristics such as presence-absence, relative abundance, size structure, body condition, and age structure data), inadequate survey designs (e.g., low survey effort or statistical power), a lack of consistency due to budget shortages or staffing turnover, and a lack of rigorous analyses of the data focused on improving monitoring methods are the causes of ineffective monitoring (Lindenmayer and Likens 2010; Reynolds 2012). Many of these issues can be addressed prior to implementing a monitoring program by defining clear objectives, developing a formal protocol with specific instructions on how to conduct the survey, analyzing historical data to estimate the required sample size, exploring collaborations between biologists and statisticians, and instituting a formal goal to regularly analyze the data collected to improve the monitoring process (Lindenmayer and Likens 2010). The goal of this chapter is to provide an overview of defining the sampling frame, site selection from the sampling frame, sample size estimation, data summarization and analysis techniques, parameter estimation, and hypothesis testing, as well as recognizing when a monitoring program is unlikely to produce the desired results.