Which SAS course should I choose?

Newsletter Icon help blueAs an instructor for SAS, I receive a wide variety of queries before, during and after delivering my courses. Most frequently, I am asked questions such as: Should I learn SAS programming or a point and click tool instead? I know lots of code, should I go straight to the advanced programming course?  I need administrator training, should I do the Quick Start or the Fast Track?  I’ve taken these particular courses, what would you recommend I take next? Really, all of these questions can be categorized into one: which SAS course should I choose?

Perhaps you have started a new role which requires you to use SAS and you need training; maybe you have been allocated some training budget to develop your existing SAS skills; possibly the company you work for are transitioning their data platform to SAS; or perhaps you are wanting to reskill into SAS? Formal training is often the best way to quickly gain or grow your SAS knowledge. This is not an exhaustive list of scenarios but all of them lead to that question I mentioned above: which SAS course should I choose?

In my last post How Can I Learn SAS?, I discussed the various training formats SAS offers. Usually individuals do not decide the method of training format first – their initial search is to find a course to attend. If you have looked at the complete course list, it can be an overwhelming starting point. Below I have listed some of the most popular SAS courses and some pointers to help guide you. This is by no means an exhaustive list. For every SAS course offered on the SAS Education website, there is a detailed breakdown, including:


  • Overview which provides a high level synopsis of the course, a summary of what you will learn and a guide to who should attend. Some courses also include a link to an online assessment, Ready for this course?, to determine if you are a good candidate for a particular class.
  • Prerequisites which details what advance knowledge you should have prior to attending the course.
  • Course Outline which lists the content chapters and sections of the course.

Read More »

Post a Comment

How much does each country spend on healthcare?

With the US Affordable Care Act (ACA)  and other health topics in the news lately, I wondered how much people spend on healthcare in various countries. Of course spending varies from person to person, so I decided to take a look at the average per capita spending in each country (it's not perfect, but it's still interesting!).

First, I decided to look around to see what's already out there. I found the following graph in a Kaiser Family Foundation (KFF) publication. It was generally what I was looking for, but only showed a handful of countries, and wasn't the easiest to read.


I did a little digging, and found that similar data were available for several other countries too. So I downloaded the csv from the Organisation for Economic Co-operation and Development webpage (oecd.org), and read it into SAS.

Since I was including more countries in my graph, I knew that having a different color for each country was not going to work (it was already difficult to discern the 6 colors used in the original graph). Instead of using colors and a color legend, I added html hover-text with the country names to each line (you'll need to click the snapshot below, to see the interactive version), and labeled certain countries to provide some quick/easy reference points.


I liked my graph better than the original, but it was still difficult to compare the values in the 1970s and 80s. Since the graph showed current-dollar values, the more recent years had much higher values (because of inflation), which squished the values on the left side of the graph. I tried using the consumer price index (cpi) and converting all the values to constant (~1983) dollars, but I wasn't really confident that applying the US cpi to other countries was a valid thing to do. Also, I wasn't sure whether to use the cpi for all items, medical care, or something else.

Let's take a little intermission, while we think of a different way to visualize this data. Here's a picture that my friend Margie made for me (special, just for this blog). What better way to take a "closer look" at data, than to use a magnifying glass!?!  Thanks Margie!


And now, with my mind cleared and thinking outside the box, I came up with a different way to graph the data. What I really wanted to do was compare the other countries with the US, therefore why not do something more direct, such as calculate the other countries' spending as a percent of the US spending! This way, I have a number that I can compare from year to year much more easily.


This graph shows that since about 1990, the other countries have spent less than 75% of what the US spent on per capita healthcare (most of them way less). If the US spends more, do we get more? If other countries spend less, do they get worse healthcare? I guess those are the big questions ... and hopefully I will find some data to answer those questions in an upcoming blog!

So, what are your theories and thoughts on healthcare, and why the US spends so much more per capita than the other countries? I wonder if there's a lot of variance on the per capita spending, depending on income level, and such? What's the "right amount" to spend, for an acceptable level of health care? Feel free to share your thoughts and suggestions in the comments section!


Post a Comment

Which drinks have the most, and least, caffeine?

A friend and I were recently debating how much caffeine is in various drinks. I felt like my hands were tied behind my back, since I had no graphs to show them. Of course, the first thing I did when I got to work was create some caffeine graphs! They turned out pretty interesting, so I thought I'd share them with you...

I found a really informative website called caffeineinformer.com that provides detailed data on many different kinds of drinks. I downloaded the data for soda, tea, and coffee by copy-n-pasting them into Excel spreadsheets, and then using Proc Import to read the data into SAS datasets.

Each category still had a few too many drinks to quickly get a mental grasp on (typically 100+), so I took a two-stage approach -- first I plotted just the handful of drinks in each category that I recognized and/or might drink, and then I plotted the entire list. I'm including images of the small subset graphs below, and you can click them to see the graphs of all the drinks in each category:

Graph of how much caffeine is in various sodas.

Graph of how much caffeine is in various teas.

Graph of how much caffeine is in various coffees.

I'm not usually a big fan of embellishing graphs with things like artwork or dancing bologna. But in this case I decided to add a simple image in each graph, to let help the user quickly distinguish between them (a glass of soda, a glass of ice tea, and a cup of coffee).

So, how did your favorite drinks fare in this comparison? Did they have more, or less, caffeine than you thought? (And what do you think about the images added to the graphs - do they help, or distract?)


Post a Comment

9 skills to become a more modern marketer

Today, the need to take an analytical approach to business challenges no longer only applies to traditional, data-driven functions of a business, like IT, but is becoming more and more relevant to every part of an organization, including the marketing function.

Employers tasked with staffing modern marketing teams are facing a well-documented digital marketing skills gap.  But, if you’re someone who’s seeking a new marketing career opportunity or wanting to expand your current skill set to keep up with the industry, meeting the challenge to be more analytical can provide you with tons of opportunity -- if you can set yourself apart from the competitive pack with the right set of skills.  So, exactly which skills do you need?

In her book, The Analytical Marketer, SAS global marketing executive Adele Sweetwood shares related insights she’s gained from her years leading the transformation of a large marketing organization into a data-driven, analytically-focused team, and she highlights nine skills as essential for success. She states:

  1. Sales skills. It’s no longer good enough for marketing to simply focus on filling the top of the funnel and passing leads off to sales. The analytical marketer needs to be active throughout the buying process and work hand in hand with salespeople to provide the information that will help them close deals.
  2. Social media skills. Social media dramatically change the buyer-seller-influencer dynamic. But only those actively participating in social media tangibly appreciate the differences between old-style, one-way media conversations and group interactivity.
  3. Journalism and storytelling skills. With buyers getting the majority of their information from the web, and with potential sales an increasing priority, there’s no end to the need for juicy, targeted content. Storytelling also comes into play in campaign design.
  4. Process design skills. Automation is just beginning to penetrate the market. As anyone who has been part of a reengineering effort can attest, it’s not the automation that increases productivity. It’s the process changes that automation enables and enforces. Deploying marketing automation will require skills such as process modeling, project management, the ability to train and manage change, and ease with technology.
  5. Data and analytics skills. Technology captures and makes available enormous amounts of data about buyer and seller behavior. A marketer must be a data guru with a passion for analytics and curiosity.
  6. Domain expertise. Customers don’t care about our products. They care about themselves and their problems. Building a bridge between our products and the customer requires knowledge of both realms.
  7. Collaboration and exceptional communication. These skills are not mutually exclusive. On just about every job posting these days, you will see that “communication” skills are a must. Communication has a different meaning for marketers in our world. Traditional communication skills need to be supplemented with an intense focus on collaboration through effective communication. There are no one-man or -woman bands, only full orchestras with very clear objectives and constant interaction.
  8. Creativity and innovation. We need people to reach for the next idea. The term “creativity” is no longer applicable to just the agencies or the designers. Today’s channels and digital work approaches enable and encourage creativity at all stages of marketing and the marketing process. Creativity is at the heart of innovation, which is not only required but rewarded.
  9. A leader is someone who is willing to take risks, drive change, and build trust. We need these skills at every level of the organization, not just the vice president level. Today’s marketers, regardless of their role, have a unique opportunity to demonstrate leadership in their field and across their business for maximum impact.

Read More »

Post a Comment

Where did Hurricane Matthew drop the most water?

This past weekend, Hurricane Matthew came through the Carolinas. Some areas had record flooding, while other areas didn't. I was anxious to get back to work today, so I could use SAS software and create a custom map showing who got how much rain.

But before we get to the official rainfall data, let's have a bit of fun with some unofficial data... Below is a picture of my team's dragon boat after Hurricane Matthew. The cover blew off, and there was about a foot of water in it. Since it's about 40 feet long, and 3 feet wide, that's about 120 cubic feet of water -- which is about 898 gallons. At 8.3 pounds per gallon, that's (roughly) 7,500 pounds of water!

dragon boat full of water

Now, on to the more serious analytics! ... I downloaded the rainfall data from the National Weather Service. It was in shapefile format, so I used Proc Mapimport to get it into a SAS dataset. I used a data step and a where clause to subset the data, and limit it to just the range of latitudes/longitudes I was interested in. I then used annotate to place a colored dot at each data point on the map. The dots are so close together that they blend together and resemble a contour map, which is the effect I was wanting. I think the resulting map looks pretty good! Click the image below to see the full-size map, with html hover-text showing county names.

Hurricane Matthew rainfall map

Notice that there is a definite band of the magenta color, showing the areas along the edge of the hurricane that got the most rain. It is interesting that some areas to the east (actually closer to the eye of the hurricane) got less precipitation. Was your area affected by this hurricane? If so, how did it compare to previous hurricanes?

Update: Some people have questioned how I picked the color scale. I'm not a weather expert, so I found a NOAA precipitation map, and used the same color scale they had used. If you'd like to re-use my example with different colors, they are easily changed via macro variables in the SAS code.

Post a Comment

Digging deeper into weather data, with custom SAS Analytics!

When you were growing up, did you ever imagine that the most popular show on TV would be 24-hour coverage of a storm?!? Yes, for better or for worse, the weather has become our new reality TV.

And as you watch the weather coverage, you probably see all the weather reporters showing the same old weather maps. With all the data available these days, it seems like they'd be able to provide something a little more in-depth or customized, eh? And speaking of in-depth (pun intended!), here's a photo from my friend Paul, who was doing his own analytics on a recent snowstorm ...

5 inches of snow

Well, that's what this blog post is about - taking the reams of weather data, and creating custom visualizations using SAS software. Rather than writing a lot of text about my weather-related examples, I'll just show them to you. I've set up an index page that will let you see them all (and download the SAS code used to create them). Click the screen-capture below to go to the index page:






And what better way to end a blog post than with a nice rainbow. This picture is from my old college friend Mubarra.

a nice rainbow

Post a Comment

Introducing SAS Studio to SAS programmers

sas-studio-for-sas-programmersOver the past 40 years, SAS programmers have used several different methods and interfaces for writing and submitting SAS code. Have you heard about SAS Studio? It is the future of SAS programming! SAS Studio is a web-based application that allows you to access your data, write your code and look at your results from any machine with a browser! Want to use your Mac or tablet? SAS Studio makes that possible. If you downloaded the free SAS University Edition, you many not have realized it, but you are working with a version of SAS Studio. And if you have started hearing buzz about SAS Viya, you might be interested to hear that SAS Studio is a critical part of programming in this new and exciting cloud-enabled platform.

To introduce SAS programmers to SAS Studio, we are offering a ½ day Live Web course titled SAS Studio for SAS Programmers. Check out this video to learn a bit about SAS Studio and some of the great features you will learn about in class!

Ready to register? Visit our course outline page.

Post a Comment

Will Hurricane Matthew follow in its predecessors' footsteps?

Like many words, if you just say "Cat 5" different people might think you're talking about different things. Computer geeks might assume you're talking about category 5 network cabling. But around this time of the year, most people in the southeast US will probably assume you're talking about a category 5 hurricane ... and that's the topic of my blog post ...

Before we get started, here are a couple of hurricane-related photos to get you into the right mindset. These photos are both provided by my blog editor Maggie, from the last NC hurricane she covered as a news reporter. The first one is a picture of a brave (or is that crazy?) NBC reporter, and the second one shows how some of our coastal locals feel about hurricanes.



This past weekend, while I was at Lake James enjoying great weather and racing dragon boats, Hurricane Matthew grew to a category 5 storm, which means its winds were 157 mph or higher. It is still located south of Cuba, but is forecast to probably start heading north, and could affect many states in the southeastern US (including North Carolina). Read More »

Post a Comment

Looking after your digital self

digitalselfJared Cohen, President of Jigsaw, formerly Google Ideas, and Chief Advisor to Alphabet Executive Chairman (former Google CEO) Eric Schmidt, presented a very interesting keynote at Analytics Experience this month. In it he mentioned the idea of a ‘digital self’. He explained most people have not one but numerous digital selves out in the Ethernet and they should start taking their digital selves just as seriously as their physical one. Your digital selves expose themselves to many more people than your physical one could ever do and their presence is longer lasting!

So what is a digital self? Well, it is your presence on the web, on Facebook, Twitter, LinkedIn, to name but a few. Quite often these take on many different personalities – one for work, one for your friends, one for your family, perhaps even one for your alter ego! The point is that they are all out there, and you need to take care of them, to make your digital presence a powerful, professional, and authoritative one. Read More »

Post a Comment

CIMB Data Science Challenge

datachallenge4Similar to hunting for Mewtwo (Pokémon), identifying and recruiting talented data scientists is just as elusive. CIMB decided that the best way would be to bring budding data scientist together for a challenge.

This will provide candidates the opportunity to win cash prizes along with potential employment with CIMB’s Decision Management division. There was a prize fund of RM30,000 up for grabs with the top 6 teams taking home a cash prize. The winning team would walk away with RM12,000.

The Challenge

It was felt that the challenge should be based on an actual problems CIMB wants to address. Therefore, the problem statement issued to the participants is as follows:

“CIMB wants to improve cross-sell take-up of its main consumer banking products by identifying the most relevant product to cross-sell to each of its existing customers. This is to enable CIMB to run targeted marketing campaigns, offering relevant products to each customer, deepen banking relationship and share of wallet.” Read More »

Post a Comment