• View
• Edit
• History
• Print

An introductory video tutorial is available at: https://youtu.be/kRBCEQO9IHw

Below is step-by-step walk though the video.

1. Loading the NY Jazz dataset

• If the nyjazz dataset is not listed in the Data Viewer, select Import RDS Data from the File menu. Click here for more information about the Data Viewer.
• Use the dialog boxes to select the file nyjazz.rdsat. HInt: In Windows look in C:\Program Files\RDS Analyst\library\RDS\extdata, though the exact path depends on where you installed RDS Analyst. For Mac, search for the location of the extdata folder.
• The data is already in RDS format and upon loading is listed in the Data Viewer with an rds preface to indicate this (rather than a df preface for a regular spreadsheet).
• If you do not see the nyjazz data listed in the Data Viewer, try opening and closing the Data Viewer window.

2. Viewing and Editing Data in the Data Viewer

• In the Data Viewer select the (rds) nyjazz data set from the Data Set menu (center of the window pane). If you load more than one data set you can select the one to view here.
• Now you can look at the data in spreadsheet format.
• On the spreadsheet are three tabs:
  • Data View which is the current spreadsheet view
  • Variable View which summarizes the variables and their properties.
  • RDS where you can add information (meta data) about the data set (such as estimates of the population size). It is OK to go with the default values for now.
• Click on the Variable View tab. Click the value for Gender.MF. Under the Type column select Factor. Repeat for Race.WBO., Airplay.yn., and Union.yn.. This makes sure that the program recognizes these as categorical variables.
• Click on the RDS tab. Click the value for Mid in the Population Size Estimate section. Enter 20000 as an estimate of the population size for the NY jazz population. It is used in the computations, but they are somewhat insensitive to it as long as it is not close to the sample size.
  • The data in the video is in Recruiter ID format. See the Convert to RDS help page for information about RDS data formats.

3. Looking at the data (the Plots menu)

• From the Plots menu select Plot Recruitment Tree. Click Run. The plot will appear in a new graphics window (called "JavaGD (#)").
• By default, nodes are labeled by subject ID, but there are options in the dialog box you can play with to adjust the nodes.
• Select the "File" menu when the JavaGD window is in focus to save the plot as a PDF, JPG, etc.
• Select Recruitment Diagnostics from Plots menu.
  • Unselect the Recruitment tree box (you just saw that) and select all the others.
  • Select PDF Report and click Run.

4. Analyzing the data (the Population menu)

For exploratory data analysis:

• Select Population Crosstabs from the Population menu.
• Select Airplay.yn. as the Row and Race.WBO. as the Column using the arrow buttons.
• Click Run
• Re-open the Population Crosstabs dialog.
• Select Union.yn to Stratify By
• Leave the other settings as they are and click Run.
• Look in the Console window for the "Contingency Table" output. On top of the console window there are two tabs: Console view and Element View.
• In the Console you should see two contingency tables, the first for respondents in the union (Union = 1) and the second for those not in the union (Union=2).
  • Console view gives a continuous view of the commands sent to the RDS Analyst engine.
  • Element View shows the results of each command separately. Click on the Element View tab to isolate the latest Contingency Tables.

For inference:

• Select Frequency Estimates from the Population menu.
• Select Airplay.yn. as the Variable using the arrow buttons and specify 20000 as the population size estimate. Leave the other settings as their defaults.
• Click Run.
• Look in the Console window for the results. This function uses a computationally-intensive algorithm to compute the confidence interval and will take 20 seconds or more to return output to the Console.
• The output is a summary table with the point estimates of the percentage of respondents who have and have not received radio airplay and 95% confidence intervals on the estimates. It also reports the design effect (the factor by which an RDS sample must be bigger than a simple random sample to achieve the same power), standard error and the sample size adjusted for any missing values.

5. Saving data

• To save the data in RDS format:

• Note that data saved in RDS format will not be compatible with Excel, for example. For that capability, you can save as a Flat File with a .csv suffix.

6. Saving your work

• To save the contents of the console (results interspersed with the commands that produced them), click on the Console and choose Save from the File menu.
• You may not be required to add an extension to the file name, but it is a good idea to use '".txt", e.g., "nyjazzanalysis.txt" for maximum compatibility--even a simple text editor such as WordPad could open it.