I recently received this interesting question regarding Multilevel Models after one of my last blog posts: Question: Can you tell me when a multilevel-model is not appropriate? I have data that by design is clustered but the random intercept in the null model is not significant. I have seen advice
In the previous blog on CLASS variables, we developed the idea of creating design variables and examined GLM and reference coding. Another coding scheme, known as ‘Effect Coding’ or ‘Deviation from the Mean Coding’ is also commonly used. Consider our previous scenario of modeling the average amount spent on a
I haven't seen an official definition of 'infographic' that I really like ... but in my mind it's something 1/2 way between data visualization & artwork. It borrows elements from graphs & dashboards, and combines that with an artistic poster. If you perform a Google image search on infographic you'll see
While many analysts understand how to interpret the parameter estimates from linear regression with continuous input variables, they may feel less comfortable with parameter estimates from models with categorical inputs. SAS modeling procedures (such as PROC REG and others) need numerical information to represent all inputs—including classification variables. Suppose we
I recently attended the O'Reilly Tools of Change for Publishing Conference in New York, and I have to confess, I had a Twitter eureka moment. I was sitting in a conference room, waiting for the next session to start and reading tweets from a publishing industry contact I've longtime admired
This week's SAS tip is from Lauren Haworth, Cynthia L. Zender, and Michele Burlew's book Output Delivery System: The Basics and Beyond. This monumental guide is packed with a wide-array of techniques and examples. As SAS programmer Christine Iodice said, "This book is one-stop shopping for all your ODS needs!"
Multilevel models (also called hierarchical linear models) are used to analyze clustered or grouped data, as well as longitudinal or repeated measures data. Consider the simple scenario shown below, where Y is continuous and is shown as a function of a continuous predictor variable, X (which has been standardized). If
If you have data where the observations are not independent due to nesting or clustering, you may need a multilevel model. Another scenario that would require a multilevel model is if you have data where observations have been gathered multiple times on the same subject (a.k.a., longitudinal data or repeated
“It’s not about how much money you make but to work on stuff that matters," according to Tim O’Reilly, founder of O'Reilly Media. At the recent Tools of Change conference, Tim O’Reilly mentioned during his keynote that writers tend to be motivated to work on “stuff that matters.” Through SAS
A big part of "winning" these days (be it sports or a business) is performing analytics better than your competition. This is demonstrated in awe-inspiring fashion in the book (and movie) "Moneyball." And on that topic, I'd like to show you a few ways SAS can be used to analyze sports data
