Skip navigation
National Library of Medicine, HTTP://www.nlm.nih.gov WebMIRS Title Lister Hill National Center for Biomedical Communications, 
		HTTP://www.lhncbc.nlm.nih.gov/
 
Home
CEB Home
Installation
Dataset
Site Map
User Guide
Help

WebMIRS Tutorial

Next Previous Table of Contents

1. Tutorial

This Tutorial is part of the WebMIRS User Guide documentation. It is intended to get you up and running WebMIRS by only reading a minimum amount of documentation.

In the following, general explanatory text is shown in dark type, like this.

Explanatory text for the specific example we will present, and actions for you to take in WebMIRS are shown in green type, like this. The specific actions for you to take are preceded with the words ACTIONS FOR YOU =====> in green and in upper case.

1.1 WebMIRS URL and login

(If you are viewing this help file after starting WebMIRS, some of the information is redundant for you, since you've already found out how to start the system. If you are viewing this help file from another source, such as a printed version, you may find this section useful, however, and for that reason we have included it.)

Using Netscape release 3.x or higher, or Internet Explorer 4.x or higher, go to the URL for the Communications Engineering Branch (CEB) of the National Library of Medicine:

  • http://archive.nlm.nih.gov

From the CEB home page, take these hyperlinks, in the order shown:

  • Projects
  • World Wide Web Medical Information Retrieval System (WebMIRS)

Follow any instructions on the screen for registering as a WebMIRS user. After you have registered, the WebMIRS team will send you e-mail with a user name and password to allow you access to WebMIRS. There is no charge to use WebMIRS.

SUN JAVA PLUG-IN: If you have not previously installed the Sun Java Plug-in program, you will be prompted to do so. Follow the instructions on the screen. You need only install this plug-in one time, not every time you run WebMIRS.

SAVING RESULTS TO YOUR HARD DISK: If you wish to be able to save WebMIRS results to your hard disk, follow the instructions on the screen to grant WebMIRS write access to your computer system. (WebMIRS is a Java applet; these by default cannot write to your system; specific permission must be granted by the user.) You need only carry out these instructions one time, not every time you run WebMIRS.

Once the above steps are completed, if the WebMIRS Login screen is not up, select the run WebMIRS hyperlink to start WebMIRS and bring up the Login screen. Use the pull-down menu to select the database you want to query (NHANES II or NHANES III) and enter your user name and password.

ACTIONS FOR YOU =====>

  • Select the nhanes2 database.
  • Enter webmirs for user name and webmirs for password.
  • Click your mouse button on the Login option.

1.2 Using the query screen to run a query

The first WebMIRS screen to appear is the Query screen. The left-most part of the screen provides a tree representation of the contents of the database you have selected to query. By using your mouse to click on nodes in this tree, you may navigate the database to select items in the database to query and or to have returned to you. In fact, you have the option to navigate the database using three different trees:

  • Taxonomy: this gives you a view of the database with a structure closely corresponding to the structure of the NCHS documentation for the data
  • Category: not implemented yet
  • Index: this gives you a view of the database by SAS variable name (currently implemented for NHANES III only)

We will create a query to extract data variables NHANES II data base for self-reported back pain on most days for two weeks and self-reported episodes of back pain for 6 months or longer. We will ask WebMIRS to give us race, sex, height, and weight for these people, as well as the appropriate statistical weighting factor, and the PSU (Primary Sampling Unit) and strata codes.

ACTIONS FOR YOU =====>

  • Be sure that the Taxonomy tab is selected.
  • Click on the Demographics node. You should see a list of sub-nodes open.
  • In this sub-node list, click on the Sample Person node. You should see a list of items open which pertain to the Sample Person. You will notice that the icons for these items are different from the icons for the higher-level nodes, indicating that you have reached the level of the individual data points in the database.
  • Click on the item: Age - years (at interview).
  • The upper right-hand part of the Query screen should now read Create a new query component. Underneath these words you should see the name of the item you have selected: Age - years (at interview).

This upper right-hand part of the screen will let you:

  • Specify the value or value range for the variable you are querying for.
  • Mark the variable for return on your query.
ACTIONS FOR YOU =====>
  • To the right of the word Value: you will see a pull-down control, marked with a downward-pointing arrowhead. Click on this, and select the >= sign. If you do this correctly, you will see >= sign appear to the left of the arrowhead.
  • To the right of the pull-down control, you will see a text entry field. Using your keyboard, type the number 0.
  • Hit the Enter key on your keyboard or click the Add button.
  • You should see the phrase Age - years (at interview) >= 0 appear in the Created Query Components part of the Query screen, located in the middle of the right-hand side of the Query screen, just under the Create a Query Component section.

At this point you have created a search clause (not quite a query yet) which will be used to delimit your query to search for people without any restriction of age (0 years or older).

Before we finish with this age variable, let's request that WebMIRS return its values to us as part of the query results.

ACTIONS FOR YOU =====>

  • Check the box Check to make this variable a return field in the Create a new query component section. (In the tree at the left, if you now click on any variable except Age, you should see that Age is marked with a checkbox, indicating that it has been selected as a variable to be returned on the query.)

Now we'll select two additional demographic items to be returned as part of our query results. (We won't use them as part of the query; we just want to get their values back in our query results.)

ACTIONS FOR YOU =====>

  • From the tree on the left part of the screen, select Sex. In the Create a new query component section on the right part of the screen, check the box Check to make this variable a return field.
  • Similarly, select Race from the tree on the left, and check the box on the right to make this variable a return field.

In the tree on the left, note that Age, Race, and Sex are all marked with special checkbox icons in the tree, indicating that they will have their values returned on the query. (If you have just marked a variable for returned, you will have to click on any other variable in the tree to see the checkbox appear for the item most recently selected for return.)<

Now we will select additional items as part of our query.

ACTIONS FOR YOU =====>

  • In the tree, click on the Sample Person Data node. The node should collapse.
  • Click on the Survey Data node. Under that node, click on Anthropometric, then Height, and finally select the item In inches.
  • On the right part of the screen, check the box Check to make this item a return field. You may ignore any settings of the Value: pull-down or the text entry box, since we are only marking this item for return, not using it in our query.
  • Now, in the tree, click on the Weight node and select the item In pounds. On the right part of the screen, check the box to make this item a return field.
  • In the tree click on the Anthropometric node to collapse it.
  • In the tree click on the Health History Supplement node to expand it. Then click on Back Pain and finally select the item, Have you ever had pain in your back on most days for two weeks? Note that if you let the mouse pointer rest on any item, the complete item description appears on the screen. This helps in the case of items with long descriptions, which otherwise might be obscured, if the screen is sized small.
  • On the right part of the screen, check the box Check to make this item a return field.
  • Return to the tree and select the item, What is the longest episode of back pain you have ever had?
  • On the right part of the screen, check the box Check to make this item a return field.

We return to the tree one final time to select the appropriate statistical weight to be returned and the PSU and strata codes.

ACTIONS FOR YOU =====>

  • In the tree click on the Survey Data node to collapse it.
  • Click on the Weights node to expand it, and select the item, Medical history interview final weight.
  • On the right part of the screen, check the Check to make this item a return field.
  • Return to the tree and select Strata code.
  • On the right part of the screen, check the Check to make this item a return field.
  • Return to the tree, and select Pseudo PSU code.

Now, we'll create and submit the query.

ACTIONS FOR YOU =====>

  • On the right part of the screen, under Created Query Components, you will see a button with two checkmarks. Click on the button. This will select all query components (search clauses) that you have created (just one, in this example).
  • Click on the AND button at the bottom of the Created Query Components section. This will join your search clauses by putting a logical AND between them. (Since we have a single query component, it just copies this component to the Query Ready for Submission area). You will see the completed query appear in the Query Ready for Submission area at the bottom of the right part of the screen.
  • Click on the Submit button at the bottom of Query Ready for Submission to submit your query. The button will change color, indicating that WebMIRS is processing your query. When the button returns to its original color, your query results are ready.
  • At this point you should see a the Confirm Record Download message box appear, telling you that you have 14,479 records that satisfy your query. (You may have to use your mouse to resize this box so that you can read the number of records; just click on the right border of the box, hold your left mouse button down and drag your mouse to the right a little, then release the left mouse button.) Click OK on this message box to have the query results downloaded for display on your computer. When the query results have been downloaded to you, they will automatically be displayed.

1.3 Using the Results screen

To view the results of the query, be sure that you have downloaded the results as described in 1.2 above, and that you have the Tabular View tab selected. (It's in the left-hand corner of the screen.)

ACTIONS FOR YOU =====>

  • You should be now be able to scroll through the data you requested for the 14,479 people in the NHANES II database who satisfied your query (which essentially was to find certain data fields for everyone who took the health questionnaire). You should see these fields in the table:
  • Row #
  • Person ID
  • What is the longest episode of back pain...
  • Have you ever had pain in you back on most days for at least two weeks?
  • Pseudo PSU code
  • Strata code
  • Medical history interview final weight
  • Height in inches
  • Weight in pounds
  • Race
  • Sex
  • Age - years

You may scroll vertically or horizontally in the data. You may resize any data column by clicking on one edge of the column, holding down your left mouse button, dragging your mouse until the column is the size you want, and releasing.

If you see blanks or periods (.) for some data fields, these items were classified with the descriptor "Blank" in the NHANES datasets. (You will have to study the NHANES documentation to interpret the meaning of "Blank" for specific data items.) You can control the representation of "Blank" values in the WebMIRS results by the WebMIRS menu choices: Options | Save Results Options... | Code to Use for Missing Data.

By using the checkbox Show Codes in Results, you may toggle back and forth between the representation of the data as numeric fields (such as 1,2 for Race) or descriptive text ("White", "Black" for Race).

Note: If you wish to view the x-ray images associated with your query results, you must return to Search Mode, (use the Search Mode tab at the top of the screen), be sure that the checkbox Enable Downloading of Images is checked, under the Options menu choice, and resubmit your query. Then, when the results are ready, you can use the Image View/Tabular View tabs at the bottom of the screen to move between the two views of the data.]

If you want to save the query results (text only; no image saves at this time) to your local computer disk, click on the Save Results button. (You must follow the instruction in Section 1.1, SAVING RESULTS TO YOUR HARD DISK, to allow WebMIRS access to your hard disk, before you do this.) Be aware that the character used to delimit fields in the saved results is controlled by the Options | Save Results Options... | Field Delimiter setting on the main menu. (You can have space-, tab-, comma-, or semicolon-delimited output.) When you click on Save Results, you will get a standard file dialog that will let you specify a file name to save in. Be aware that, if you are on a Windows NT system, you will only be allowed to save in the profiles directory for your user id. A typical directory name for a user with id "rsmith" is "c:\winnt\profiles\rsmith". If you specify the file name "run1-results.txt", then this file will contain the query results data. The data is unlabeled, to simplify importing into statistical packages like SAS. A second "map" file will be automatically created. If your output query results files is named "run1-results.txt", the map file will be named "map_run1-results.txt". This file contains (1) an INPUT card that may be copied into a SAS job to read your query results into SAS, and (2) a mapping of these SAS names into short descriptors from the NHANES documentation to assist you in keeping track of the variables. Be aware that the SAS names WebMIRS outputs for NHANES II were created by the WebMIRS team and do not correspond to any names in SAS datasets released in the original NHANES II data release; for NHANES III, the SAS names output by WebMIRS are, however, the actual SAS data names as given in the NHANES III data release.

Next Previous Table of Contents
    Return to top of page

CEB Home | CEB Projects | Related Work | Publications | Repositories | NHANES | Site Index

U.S. National Library of Medicine, 8600 Rockville Pike, Bethesda, MD 20894
National Institutes of Health | U.S. Dept. of Health and Human Services
Copyright information | Privacy policy | NLM Accessibility
USA.gov | Need a plug-in?

URL: http://archive.nlm.nih.gov/proj/webmirs/userGuide/index-1.php
Updated April 05, 2010

Send questions or comments about this site to