Task 3: How to Consult the Documentation

Before analyzing the data, you will need to know how the variable is coded, data editing, processing, and collection information, and the frequency (or sample size) of the variable. This information is found in the documentation for the components:

Step 1: Codebook

First, you will check the demographics codebook to learn how the gender variables, RIAGENDR, is coded in NHANES data.

If you are still at the variable list, click your browser's Back button twice to return to the NHANES 2001-2002 page. If you returned to the NHANES homepage, click the Questionnaires, Datasets, and Related Documentation link, then click the NHANES 2001-2002 link.

Click the Demographics link, then Docs link. The documentation is a web page and you can search for the variable, in this case RIAGENDR, and learn how the variable's values are coded, the target population, SAS label, and if the variable was part of a skip pattern in the codebook portion of the documentation.

You should check the codebooks of the other components on your own.

 

Step 2: Data file documentation

Now, you will check the laboratory data file documentation to determine which data file to use in your analysis — Total Cholesterol and HDL Cholesterol or Standard Biochemistry Profile. Their counterparts in the Second Day lab files are based on a convenience sample and can be used to determine intra-individual biological variation for methodologic studies. However, they are not appropriate for this type of analysis.

If you are still at the demographic documentation, click your browser's Back button twice to return to the NHANES 2001-2002 page. If you returned to the NHANES homepage, click the Questionnaires, Datasets, and Related Documentation link, then click the NHANES 2001-2002 link.

Click the Laboratory link, then the Total Cholesterol and HDL Cholesterol Docs link. The documentation is a web page and you can use the find feature to search for the variables "LBXTC," one of the cholesterol variables you identified in the previous task. As you read the notes for each variable, you will find that LBXTC from the Total Cholesterol and HDL Cholesterol data file is the appropriate variable to use in your analysis.

If you wish, go back to the 2001-2002 Laboratory files and open the data file documentation for Standard Biocemistry Profile, search for the other cholesterol variable, "LBXSCH," and confirm that it is not the variable to use by reading the analytic notes.

The self-guided activity is to locate and read the data file documentation for the other variables in your analysis — blood pressure examination, blood pressure questionnaire, and demographics.

 

Step 3: Frequency tables

Finally, you will check the frequency table for the gender variable, RIAGENDR, from the demographics component, to determine the sample size.

If you are still at the data file documentation, click your browser's Back button twice to return to the NHANES 2001-2002 page. If you returned to the NHANES homepage, click the Questionnaires, Datasets, and Related Documentation link in the navigation menu, then click the NHANES 2001-2002 link.

Click the Demographics link, then the Docs link. The frequency table is combined with the codebook in the data documentation. On the right side you will see a list of hyperlinked variable names under Codebook. Find the gender variable, RIAGENDR, in this list and click it. The web page advances to the RIAGENDR codebook and frequency counts for the possible responses to the question. Use the final row in the Cumulative Frequency column as your sample size.

 

close window icon Close Window to return to module page.