WORKING WITH THE DATA

1. Define Your Cohort

The first step to working with the data is to create a cohort that defines all inclusion and exclusion criteria. It is really useful to create a flow diagram (see example) to know where/when exclusions are made.

2. Define Your Variables

Second, all variables to be used in the analysis must be defined.  A really useful tool to for defining variables is the MCHP Concept Dictionary. Often cohort formation and defining variables are done together as defined variables are used as inclusion/exclusion criteria.  It is useful to create a table (that can be put into an appendix) that has the definition used for all your variables, and the sources you used to define these variables (see example). 

3. Write the SAS Code

This is often the most intimidating part of the process (especially if you have not done any programming in the past). Luckily, we have some great resources and a lot of example code for you to use.

  1. SAS Training is offered several times a year at MCHP - the training workshop handbook can be found here.

  2. SAS Code in the internal concept dictionary: Many of the concepts available on the concept dictionary have SAS code attached to them. To access this code, you have to log onto the internal version of the concept dictionary. To do this, follow these steps:

    1. Log into the VDI on one of the computers in the Training Room. The username is 'train' and the password is 'training'​​

2. Click on the 'Concept Dictionary and Glossary' tab under 'Research Resources'

    1. Click on 'Internal Version'

    2. There is a 'SAS Code' link at the bottom of most concepts

  1. If you have access to the RAS, there is also a link to the internal concept dictionary on the desktop.

  2. In RAS, there is a 'Support' Folder, within this folder, there are two folder of example code:

    1. Code used in the PATHS project

    2. Code contributed by Students