National Household Survey on Drug Abuse Finding Specific Variables in the NHSDA
April 18, 2003

Finding Specific Variables in the NHSDA

Search Tips

  • Searches are not case–sensitive.
  • Quotation marks are interpreted literally — don't use them.
  • Only whole words are matched — no wildcards or partial words.
  • Try different forms of the search term or use synonyms (e.g., employ, employment, job, work).
  • Be patient. This program searches thousands of variables and, therefore, it is sometimes slow.

The Substance Abuse and Mental Health Data Archive (SAMHDA) contains a search feature for conducting variable–level searches among studies included in the archive's online data analysis system (DAS).1 This new tool searches the question text and variable and value labels for tens of thousands of variables in order to find variables of interest across many studies. This is helpful for locating variables:

The search tool works in stages, beginning with the studies and then proceeding to the variables. It first identifies every study containing the word(s) of the search term in the question text, value labels, or variable labels. Next, matching variables are listed for a single study or for all studies. For search terms with multiple words, the tool will first generate a list of variables that contain any of the words in the search term (an "OR" search). This list can then be narrowed to only those variables containing all words in the search term (an "AND" search).

The following examples of the variable–level search utility use the 2001 National Household Survey on Drug Abuse.


How to Access SAMHDA Variable–level Search
To access the variable–level search utility:

Figure 1. Input for Variable–level Search.

Figure 1. Input for Variable-level Search.

Figure 2. Excerpt from Study–level Search Results.

Figure 2. Excerpt from Study-level Search Results.

Figure 3. Partial Variable–level Search Results.

Figure 3. Partial Variable-level Search Results.


Examples

Example 1: Single word search

This example looks for studies that include one or more variables with the word "methamphetamine", using only SAMHDA studies on the DAS (Figure 1).

Figure 2 shows a partial set of study–level search results, which include hypertext links to:

Click on "Find matching variables for all studies" to produce a single page with the matching variables for all studies. The results may display slowly, depending on the number of studies and variables that match the search term.

Select "List matching variables" for the 2001 NHSDA. Figure 3 shows a partial list of variables that contain "methamphetamine" in the question text or labels.

From these results, click on the hypertext links to examine the codebook entry for any of the variables listed (Figure 4). The online codebook provides:

To use selected variables in online analysis, click the browser's "back" button until the study–level results page (Figure 2) is displayed, then select "Online analysis" from the study options.

Example 2: Multiple–word search ("OR" search)

For some concepts (e.g., binge drinking during pregnancy), the data in a single study may not appear in same variable. To search for studies containing this combination, enter the words of the search term separated by a space (e.g., binge pregnant). The return is a list of studies that include both "binge" AND "pregnant" in the question text or labels of any variables. The 1995–2001 NHSDAs meet these search criteria. From the choice of actions available, select "List matching variables" for the 2001 NHSDA.

With a multiple–word search, the initial results will include variables that contain any of the words in the search term (in this example, "binge" OR "pregnant") (Figure 5). Narrow the search results by applying an "AND" search as shown in Example 3.

Follow the hypertext links to review the codebook entries for variables of interest. To conduct an analysis based on these results, access the DAS, select the year of the study and statistical program, and enter the variable names (e.g., a crosstabulation of PREGNANT by BINGEDRK).

Example 3: Multiple–word search term ("AND" search)

This example searches for "marijuana recency". The initial study–level search finds 19 studies containing both of these words. Select "List matching variables" for the 2001 NHSDA. The variable–level "OR" search matches numerous variables for "marijuana" OR "recency". Next, select "Narrow results with an AND search." This yields 3 variables (Figure 6). The broader results are still accessible either by clicking the browser's "back" button or by selecting "Show all results with an OR search".

Figure 4. Example of Online Codebook Entry for Matched Variable.

Figure 4. Example of Online Codebook Entry for Matched Variable.

Figure 5. Multiple–word Search Results Using "OR" Search.

Figure 5. Multiple-word Search Results Using OR Search.

Figure 6. Multiple–word Search Results Using "AND" Search.

Figure 6. Multiple-word Search Results Using AND Search.


Additional Information
For additional information about the variable–level search utility, the DAS, or the databases publicly available from the archive, email: samhda@icpsr.umich.edu or phone toll–free (888) 741–7242.


Notes
1The archive is supported by the Office of Applied Studies at SAMHSA through a subcontract with the National Opinion Research Center at the University of Chicago (NORC) and is based at the Inter–university Consortium for Political and Social Research (ICPSR) at the University of Michigan. SAMHDA and Computing and Network Services (CNS) staff at ICPSR developed the variable–level search utility. The DAS is based on the Survey Documentation and Analysis (SDA) software developed at the Computer–assisted Survey Methods Program (CSM) at the University of California at Berkeley.

The National Household Survey on Drug Abuse (NHSDA) is an annual survey sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA). The 2001 data are based on information obtained from 69,000 persons aged 12 or older, including approximately 23,000 youths aged 12 to 17. The survey collects data by administering questionnaires to a representative sample of the population through face–to–face interviews at their place of residence.

The NHSDA Report is prepared by the Office of Applied Studies (OAS), SAMHSA, and by RTI in Research Triangle Park, North Carolina.

The NHSDA Report is published periodically by the Office of Applied Studies, Substance Abuse and Mental Health Services Administration (SAMHSA). All material appearing in this report is in the public domain and may be reproduced or copied without permission from SAMHSA. Additional copies of this report or other reports from the Office of Applied Studies are available online: http://www.oas.samhsa.gov. Citation of the source is appreciated.

This page was last updated on December 30, 2008.