How to Conduct Water Quality Statistical Analysis
Statistical analysis of water quality provides timely information for regulatory agencies and resource managers that helps in assessing safety for recreational activities such as swimming. It also helps gauge effects on wildlife and in planning water treatment measures. Agencies such as the U.S. Geological Survey, state regulatory commissions, and environmental scientists analyze water quality using regression analysis, which asks what measures of water quality can be predicted, given the values of certain independent variables that impact quality.
Other People Are Reading
Things You'll Need
- Computer
- Statistical software or spreadsheet program
- Data from water analyses
Instructions
-
Using Regression to Analyze Water Quality
-
1
Open your data file with the statistical software or spreadsheet program you are using. Popular statistical software packages include SPSS, Stata and SAS, all three of which are popular among scientists, university researchers, and government agencies. The more widely accessible spreadsheet program Excel can, however, conduct regression analysis with the use of the program's Data Analysis tool. The data analysis toolpak is included in all versions of Excel, but will not appear in the Tools menu until you unpack it from the Add-ins and load it. Fortunately, the option is easy to load. To load this tool, select "Add-ins" from the Excel tools menu and click "Analysis ToolPak." After doing this, click "OK." The Data Analysis option should appear in your Tools menu, ready for use.
-
2
Decide what measure or measures of water quality will serve as your dependent variable(s). Measures favored by the U.S. Geological Survey include levels of nutrients, such as phosphorous and nitrogen, and levels of bacteria. High concentrations of nutrients in water could have adverse effects on humans and on the development of fish and other aquatic life. Meanwhile, presence of fecal coliform bacteria, for example, could indicate other disease-causing organisms. If you use multiple dependent variables, each one will have a separate regression equation.
-
-
3
Select the independent variables for your water-quality analysis. You should include those variables for which there is a sound physical basis for inclusion. Water quality analyses are specific to the sites from which samples are collected. However, some variables are common to most water quality analyses. Common independent variables include water temperature and turbidity. The latter variable measures the amount of matter transported by a stream, and is important because runoff can transport sediment and other material into bodies of water.
-
4
Specify the selected variables from your water quality data in your software or spreadsheet program. If you're using Excel, open the Data Analysis tool and select "Regression" from the menu of procedures. You will then specify a range of values for Y (your dependent variable, or water quality measure) and for X (your independent, or explanatory, variables).
-
5
Interpret your regression output, paying attention to the independent variables with statistical significance, as they are the ones with the greatest impact on water quality. In addition, note the value of R-square, which provides an overall measure of how much variability in water quality (the dependent variable) is explained by your equation.
-
1
Tips & Warnings
Be aware that the reliability of your analysis will be affected by the water samples that provide your data. The better the data collection, the better your analysis will be.
