[WEBINAR] Broadening Access to Substance Abuse and Mental Health Data with the R-DAS

Learn about the National Survey on Drug Use and Health (NSDUH) data available through the R-DAS. The Broadening Access to Substance Abuse and Mental Health Data with the R-DAS webinar provides:

• A general understanding of the data and resources available through SAMHDA;
• Differences between the public-use and restricted-use NSDUH data files;
• Instructions on how to locate and access restricted-use NSDUH data in the R-DAS; and
• A brief demonstration on how to create a cross-tabulation in the R-DAS.

What are the differences between NSDUH public-use and restricted-use data?

NSDUH public-use and restricted-use data differ in terms of access, availability, and variable groups. Users can reference the Analysis Options for NSDUH Public-use and Restricted-use Data page for help with determining which available option best meets their research needs: public-use (downloadable data), SDA (?) (online analysis of public-use data), R-DAS (?) (online analysis with disclosure restrictions), or the Data Portal (?) (virtual desktop access to restricted-use microdata).

Which variables are available in the restricted-use NSDUH data files?

The following variable crosswalk displays all variables in the restricted-use NSDUH data files and their availability for specific groups of years.

• Restricted-use data variable crosswalk

How do I perform a homogeneity test of proportions or percentages in the R-DAS?

The R-DAS does not have the Comparison of Means analysis available. The Frequencies/Crosstabulation program has an option of Summary Statistics to perform a test of independence (or no association) between two categorical variables using the Rao-Scott F statistics. These statistics take the complex design effect into account. The test of independence of a two-way contingency table is equivalent to the test of homogeneity of row (or column) percents (StataCorp 2011, page: 141-142). The null hypothesis for the later test is that row (or column) percents are equal for every category of the column (or row) variable.

For example, if your variable of interest is levels of alcohol consumption in the Column field, you can use the row percentage option and the resulting table output to approximately determine homogeneity (or the lack of homogeneity) of row (ethnic) groups among the levels of alcohol consumption. In other words, this test determines whether the distributions of each of the ethnic groups (among the alcohol levels) are equal.

The Rao-Scott F statistics are calculated from the contingency table for Row by Column variables. The test is significant at x% level of significance if the p-value of the Rao-Scott F statistic is less than x%. And, overall, the test concludes that there is association (or dependence) between Row and Column variables. The first screenshot shows RACE4 x ALCREC (recoded) table output for Total percent with the Summary Statistics box checked. From this table display of cell percent (i.e., total percent), confidence intervals and weighted cell frequencies, it is difficult to compare the prevalence of alcohol in different racial groups. In order to interpret the table output with regard to the test of homogeneity, we have to look at the table display in a better way. The second screenshot shows the contingency table output for Row percent. A larger percent (i.e., 63.7%) of whites have consumed alcohol within the last 30 days than blacks, other, or Hispanics. Since we only changed the way the percentages are displayed, the Rao-Scott F statistic is identical for both screenshots.

Reference:

StataCorp. 2011. Stata Survey Data Reference Manual, Release 12. Statistical Software. College Station, TX: StataCorp LP.

How can I transfer output from SDA and R-DAS to a document, spreadsheet, or presentation?

You can copy and paste output, including tables and charts, from SDA and R-DAS into a document, spreadsheet, or presentation. To copy and paste, use your mouse to highlight the output and then click "copy". When pasting the output into a document, use the "paste special" option to retain the same display as in SDA and R-DAS. To transfer data only, paste the output using the standard paste option.

If you have a PDF file creator or print driver, you can also print the output to a PDF file.

My results in R-DAS were blocked by the disclosure protection settings. How do I avoid having my output blocked?

Because of confidentiality concerns, we are unable to provide specific details about what is causing the disclosure protection settings to block output for a specific analytic run. However, we are able to provide solutions for several common reasons that analytic results are blocked.

When output is blocked, you may get one of these messages:

• "The Row Total is equal to the value of one of the cells."
• "To preserve confidentiality, tables cannot be displayed when the number of observations in any cell in the table is too low."

Definitions of the various blocked result messages are available in another FAQ.

Below are several examples of analytic requests where the results were blocked, and possible solutions for how to change your request to receive some analytic results.

Example 1: A user runs a crosstabulation where State is the column variable.

Possible solutions:

If interested in a single state, you might try placing the State variable in the Filter field to specify the one state for analysis. For example, entering STATE(1) in the filter field will give you results for just Alabama. Focusing your analysis on only one state might help you avoid a circumstance where a different state is causing your results to be blocked.

Another option would be to use a geographic variable like Census Region or Division in an attempt to avoid low record counts that can result in causing your results to be blocked.

Example 2: A user runs a crosstabulation where AGE is the column variable.

Possible Solutions:

The AGE variable spans an age range from 12 to 103 years old. You could try using one of the categorized age variables within the data file.

Alternatively, you could utilize the temporary recode feature in R-DAS that allows you to recode a variable into fewer categories.

Help documentation on doing temporary recodes can be found at: http://www.icpsr.umich.edu/icpsrweb/content/SAMHDA/help/helpan.htm#recode

Example 3: A user runs a three-way crosstabulation using the Row, Column, and Control fields. However, the results are blocked, and the user has no idea which variable or combination of variables contains the low record count.

Possible solutions:

Run frequencies for the variables in your analysis one at a time. One variable may stand out as having a value with a particularly low weighted frequency. It is possible that a variable has a value with such a small record count that the univariate frequency is blocked. If one variable does stand out as being the primary cause of the problem, then you could check to see if a similar variable exists with fewer categories, or you could do a temporary recode to create larger record counts.

If no single variable stands out as causing the problem, then try running crosstabs on two of your variables. If any cross combination of values from the two variables has a particularly low weighted frequency, then this can be an indicator that the combination is the cause of the problem. If one combination does stand out, you could find similar variables to the ones you chose, but have fewer categories. Again, you could do a temporary recode on one or more of your variables to create larger record counts for the categories/values of the two variables that are the possible cause of the problem.

My results in R-DAS were blocked by the disclosure protection settings. What do the various messages mean?

Below are descriptions of the most common messages that display when analytic results are blocked.

1. The Row Total is equal to the value of one of the cells.

   This message refers to a built in disclosure limitation protection for specific crosstab output. In the following 5 X 3 crosstabulation example, the sum of the 4th row is equal to a single cell in that row. The whole table is suppressed when this happens.

   6	15	8
   9	17	8
   3	20	5
   0	5	0
   30	4	7

   
   
2. To preserve confidentiality, tables cannot be displayed when the number of observations in any cell is too low.

   This error message states that at least one cell in the frequency of the table or crosstabulation does not meet the threshold established by CBHSQ/SAMHSA for protecting the confidentiality of respondents.

3. To preserve confidentiality, analyses are not permitted to use the following variable(s): 'variable name'

   This message appears when one of the complex design variables (weight, strata, or cluster) is entered into one of the analysis fields (i.e. ROW). While the complex sampling design variables are used by the R-DAS system to calculate accurate statistics, the design variables are not available because of the potential disclosure risk involved.

How do I produce correct estimates for NSDUH: 4- and 2-Year R-DAS data files?

To generate correct estimates for the NSDUH: 4- and 2-Year R-DAS data files, users must use the year pair indicator variable (YRPRIND) in one of the following ways:

1. As a filter to subset the data file for a specific group of years such as:
   - For the 2-Year R-DAS data file use "YRPRIND(1)" for 2002-2003, "YRPRIND(2)" for 2004-2005, "YRPRIND(3)" for 2006-2007, "YRPRIND(4)" for 2008-2009, "YRPRIND(8)" for 2010-2011 or "YRPRIND(10)" for 2012-2013 in the filter field.
   - For the 4-Year R-DAS data file use "YRPRIND(5)" for 2002-2005 or "YRPRIND(6)" for 2006-2009 in the filter field.
2. As a control variable. PLEASE NOTE: This option will produce separate results for each combined group of years (e.g., 2002-2005). The last table with total estimates is invalid and should be ignored.
3. As a row variable in the row field. PLEASE NOTE: Under this option the column total estimates are invalid and should be ignored.
4. As a column in the column field. PLEASE NOTE: Under this option the row total estimates are invalid and should be ignored.

The NSDUH R-DAS data files do not allow for the creation of single-year estimates because of the potential for disclosure of confidential information. The NSDUH R-DAS data files contain revised weights designed to produce estimates that are representative of the average population across combined two-year (2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011, and 2012-2013), four-year (2002-2005 and 2006-2009), eight-year (2002-2009), or ten-year (2002-2011) periods. There are no weights on the 4-year R-DAS that are representative of all eight years; thus all estimates of totals across the period 2002-2009 will be twice as large as they should be. Similarly, there are no weights on the 2-year R-DAS that are representative of all twelve years. To get the correct estimates of totals across the eight year period, the 8-year R-DAS file should be used. Similarly, the 10-year R-DAS file should be used to get the correct estimates of totals across the ten year period.

What is the Restricted-use Data Analysis System (R-DAS)?

The R-DAS is an online analysis system that allows researchers to produce frequencies and cross-tabulations using restricted-use data files. The R-DAS provides output in the form of tables and frequencies for viewing and exporting. Advanced statistical methods are not available in the R-DAS at this time.

The R-DAS does not permit listing of individual cases and does not provide unweighted frequencies in the R-DAS codebook, nor are users able to generate unweighted frequencies (no unweighted sample sizes are provided). These limitations have been put in place to reduce the potential for disclosing confidential information.

The R-DAS provides standard errors that take into account the complex survey design. All weighted totals and point estimates are rounded to the nearest thousand and all percents and associated statistics are rounded to one decimal point. If any cell in a table contains too few unweighted cases, then the entire table is suppressed.

The R-DAS does not currently allow for the creation of composite variables (i.e., the creation of new variables using other variables), but that capability is under development. The R-DAS does allow for temporary recoding of existing continuous and categorical variables. See the SDA 3.5 help documentation for assistance with how to Temporarily Recode a Variable.

Watch the "Broadening Access to Substance Abuse and Mental Health Data with the R-DAS" webinar to learn about the National Survey on Drug Use and Health (NSDUH) data available through the R-DAS. For more information on analyzing data with the R-DAS, consult the FAQ section on Help with the Restricted-use Data Analysis System (R-DAS).