Sanjay Matange is R&D Director in the Data Visualization Division responsible for the development and support of the ODS Graphics system, including the Graph Template Language (GTL), Statistical Graphics (SG) procedures, ODS Graphics Designer and related software. Sanjay has co-authored a book on SG Procedures with SAS/PRESS.
A week ago I had the pleasure of attending MWSUG 2012 in Minneapolis. There were many presentations on Data Visualization in various tracks including the following: Using SAS ODS Graphics - Chuck Kincaid Get the best out of SAS ODS Graphics... - LeRoy Bessler Program Assisted Patient Narratives - Faye
A week ago, SAS "wusses" got together in Long Beach, California for another successful WUSS conference. Personally, I really enjoy all SAS user conferences as it gives me an opportunity to see the creative ways in which SAS users are exploiting our software. One such case was the paper "Power
A few weeks ago I wrote an article on using the non breaking space character to prevent stripping of leading and trailing blanks in a string. Since then, I have discovered a few more instances where the nbsp can be a useful tool for creating graphs. One such instance came up last week
Pie charts have been the subject of some criticism when they are used to compare measures across multiple categories. It is generally accepted that comparison of magnitudes represented as angular measures from varying baselines is not effective. However here are some use cases where a pie chart does quite well. When it comes
The Scatter Plot Matrix is a great tool that provides a quick visual of potential associations between variables. This may provide the analyst some hints on how to proceed with the analysis. Matrix of lab values for liver function tests are commonly used in clinical research. The SGSCATTER procedure provides an easy way
Forest plots come in many flavors. An example of a Forest plot using GTL is available on the SAS support web site. Simple forest plots can also be created using SGPLOT procedure by using the SCATTER statement with MARKERCHAR to display data aligned with the plot by study names. One issue
In the Clinical Research domain, there is often the need to display lab values by treatment or test and it is often useful to view this data along with reference lines showing the normal ranges. The obvious way is to use reference lines to denote the normal ranges. SGPLOT Code: proc sgplot data=band; scatter x=x y=y
When comparing results by category and group, putting the items to be compared close together usually leads to a graph that is easier to decode. Take the case of the data (simulated) shown below. Here we have population by age group and sex. To compare the population by sex, it is
A key element of graphs used for analysis of safety data for clinical research is the inclusion of statistical data (or tables) about the study that are aligned with the x axis of the graph. A common example of this comes from the paper "Graphical Approaches to the Analysis of
Spark lines, made popular by Edward Tufte, provide a way to visualize trends in a concise space, often inline with the rest of the narrative or data. Previously, I posted an article on Spark Plots in which I created different plot types, some of which included multiple graphs and data in each row. For such
Statistical graphs often include display of derived statistics along with the raw data. Often these statistics are presented in a tabular format inside the graph. With SGPLOT procedure, a table of statistics can be added to the graph as an inset table, as shown below. Using a Stat Table: SGPLOT code:
ODS Graphics components like GTL and SG procedures are designed to work with Styles to create graphs that are effective in the delivery of information and aesthetically pleasing out of the box. You no longer have to tweak the colors to ensure a nice graph. The graph derives all the
CTSPedia.org is a website of Knowledge Base for Clinical and Translational Research. On this site you can find sample graphs for statistical analysis of safety data for Clinical Research. Graphs included in this resource have been submitted by contributors, and include a graph for Liver Function for different tests by treatment.
A graph in a recent article in Fortune magazine caught my eye. The graph shows the cost of hosting the Summer Olympics over the past eight events. Here is what I termed the "Medal" graph. Now, practitioners of the art of Effective Graphics would likely find some shortcomings in the graph. Clearly
In the previous post on Broken Y-Axis, I reviewed different ways to display data as a Bar Chart, where the response values for some categories are many orders of magnitude larger than the other values. These tall bars force the display of other values to be squeezed down thus making it harder to compare
Often we want to display data as a bar chart where a few observations have large values compared to the rest. Comparison between the smaller values becomes hard as the small bars are squeezed by the tall bars. Here is an example data, and a bar chart showing the data. The large values
Recently, I came across an interesting graph showing Euro contries bank exposuro to GIIPS countries, as percent of GNP. Here is the graph: I thought I would see how far I can get in making a similar graph using SAS. I made up some data with response values for a Product x
This post is a result of an intriguing question posed by a user on the SAS communities' page. How to create a bar chart where the category is a combination of two variables var_1 and var_2 (each with values of '+' and '-'), and get a X axis that shows each
At SAS Global Forum, and again at PharmaSUG, we had the pleasure of attending many papers and presentations on various topics that included graphs in the power point decks or in the papers. More often than not, the graphs exist along other text, and occupy a smaller part of the screen
PharmaSUG 2012 conference drew to a close today, concluding two and a half days packed with papers, presentations, posters, hands-on demos and super demos by SAS staff. While the weather outside was a bit chilly from time to time, the conference what hopping with many user papers on how to
Recently I posted an article on this blog on how to create bar charts with log response axes in response to a question by a user. This generated some feedback suggesting that bar charts should not be used with log response axes or with a baseline of anything other than
Getting the axis values just right generally requires some work, and the values you want can change from case to case. One such example was discussed by Dan Heath in his post on custom axis values. Here Dan shows the usage of non uniform axis values using the VALUES option on
Creating bar charts with log response axis has come up a few times in the past few days. Before we look into how we could do this, it would be worth pointing out the considerable opinion in the blogosphere against use of log response axes for bar charts. See BizIntelGuru and
On Friday before the conference, I presented a 1 day "developer led" seminar on SG Procedures and GTL, along with a discussion of new features for SAS 9.3. The experience was very gratifying as all users were now using SAS 9.2, and some were using SAS 9.3. We had a lively
Recently, an interesting question was posed on the previous article on this blog by a reader. Can we use the new DiscreteAttrMap feature to create just a legend with specific entries, with no graph. The question was intriguing enough that I did not wait to ask - "Why?". I just got
SAS Global Forum 2012 at Orlando, Florida is just round the corner and we are excited to see so many presentations offered by users on SG procedures and GTL. We'll add a few more on new SAS 9.3 features of SG procedures and GTL. These include cluster groups for discrete and interval axes, cluster
Recently, a user asked about creating a Bar Chart of Value by Date, where the dates are displayed on a scaled interval axis. Consider this simulated data set of value by date and treatment shown below. This data set only has one value for each date and treatment combination. We can use the VBAR statement
ODS Graphics have matured. With SAS 9.2, GTL and SG procedures were a new direction for creating analytical graphs in SAS. The motivation and design of the GTL framework and the SG procedures was driven primarily by the needs of the procedure writers within SAS to enable the automatic creation
A recent article in the SAS and R blog was about current winter temperatures in Albany, NY. The temperature data for the recent winter (Dec 2011 - Mar 2012) was plotted on a polar graph. Robert Allison posted an article on displaying the same data as a Polar Graph using SAS/GRAPH . Here is his
A user recently posted a question in the SAS communities forum about how to best display two measures by one classifier using a Bar-Line graph, where the scale of the two measures is vastly different. This got me thinking about various different ways to represent such data. Here are some of my thoughts,
