CADDIS Volume 4: Data Analysis
Selecting an Analysis Approach
- How Can I Use My Data?
- Establishing Differences from Expectations
- Estimating Stressor-Response Relationships
How Can I Use My Data?
- Author: L.L. Yuan
How Can I Use My Data?
The Stressor Identification approach does not require a minimum data set, and existing data often are sufficient to determine the cause of impairment. When available data are too sparse to support a causal analysis, the Stressor Identification process can be used at a screening level to identify data that would be most useful to collect. When larger monitoring datasets are available, the analyst can use them to do a quantitative causal analysis.
Analysis of state and regional monitoring data sets can inform two questions that are particularly relevant to causal assessment:
Are environmental conditions or biological characteristics at a test site different from expectations?
Establishing differences from expectations.
In most causal assessments, we start with information that the observed biota at a test site differ from our reference expectations for that site, i.e., the site is impaired. Data analysis can often inform causal assessment by determining whether certain stressors also differ from our reference expectations for the site. Answering this basic question can support evaluation of spatial/temporal co-occurrence. That is, to establish co-occurrence we would like to assess whether a stressor is present when the biological effect is observed, by considering whether the stressor level at the impaired site is similar or different from stressor levels expected at one or more reference sites.
This same question often underlies evaluation of the verified prediction type of evidence. In this case, we hypothesize that a particular stressor causes the observed impairment, and based on this hypothesis, we make a prediction regarding the biological characteristics at the impaired site. For example, if increased bedded sediment is a possible stressor, we might predict that the number of clinger taxa at the impaired site is low (see page on traits). Operationally, to determine whether the number of clinger taxa is "low", we must assess whether clinger taxa richness at the impaired site differs from reference expectations.
What is the relationship between an environmental stressor and a biological response in a particular region?
Estimating stressor-response relationships.
Stressor-response relationships can provide an estimate of the magnitude of the effect we would expect for a given stressor level. Different statistical approaches can be used to estimate stressor-response relationships with varying degrees of confidence. Estimating and applying stressor-response relationships to different types of evidence (stressor-response from the case and stressor-response from other field studies) may require consideration of different issues, such as the number and types of covariants present across a larger study area.