Key Concepts About Creating New Variables

Creating new variables by recoding or deriving is an important step in preparing an analytic dataset containing environmental chemical data. You may want to recode an existing variable to consolidate responses. You may also wish to derive a new variable from multiple existing variables to fit your analytic needs.

Generally, two variables are provided for each analyte in environmental chemical data: one to record values of each analyte, and another to indicate at or above or below Limit of Detection (LOD) for that analyte. Sometimes only one variable (values of measurement) is provided for each analyte in environmental chemical data. So when you combine data from different cycles, you need to create a new variable to indicate at or above or below the LOD using multiple existing variables and different cut points.

For environmental chemicals measured in urine, levels are presented per volume of urine. You could create a new variable to adjust urine dilution (i.e., creatinine-adjusted). The urinary creatinine variable is always included in environmental chemical data if the chemicals are measured in the urine.

Recording or creating variables is useful in multiple scenarios, such as:

• Recoding a continuous variable (e.g., age) into a categorical variable (e.g. age group);
• Regrouping a categorical variable (e.g., race/ethnicity);
• Deriving a new variable (e.g., variable to indicate above or below the LOD); and
• Create a creatinine-adjusted variable of interest.

 

Each of these examples is described in the "How to Create New Variables" section of this task.