Home   About Kurtosis   Courses   Course Calendar   Booking Information   Ideas   Technique   Clients   Contact
     

Microsoft Word. Microsoft Excel.

Numbers with narratives; numbers without narratives

(2 comments)

This week has been unusual for me. Iíve been spending a fair bit of time working in an NHS headquarters as a data analyst. Iíve been given a Visitorís Pass, an NHS email address and lots of forms to fill in and sign. Iím a quasi-employee again. Itís almost as if Iím back as part of a team of analysts.

Almost.

The trouble is, by spending so much time in the company of analysts, I think I mightíve discovered this week just how unusual I am.

Look, when you walk past the desks and workspaces of your co-workers, itís difficult not to glance at their workstations, right? So you kind of notice what theyíre working on. And what everyone seems to be working on is Excel or Oracle or SPSS or CliqView or some other numbers-based piece of software.

Whereas me, Iím mainly using Microsoft Word.

Why?

Iím supposed to be a number cruncher. Not a wordsmith.

Whatís going on?

Well, itís to do with what Iíve learned this week to call ďhigher objectivesĒ.

One of the other things I've been doing this week, as well as being an NHS analyst again, is I've been reading a book by Matt Watkinson called The Ten Principles Behind Great Customer Experiences. Iíd stumbled across a short piece by him in Wired magazine at the weekend, and thought it was so good I bought the book.

The second principle behind Great Customer Experiences is:

Great customer experiences satisfy our higher objectives.

And to illustrate what this means there is a great quote from Theodore Levitt: "People don't want to buy a quarter-inch drill; they want a quarter-inch hole."

So that got me thinking about what the ďhigher objectivesĒ of NHS managers might be.

Specifically, is "data" one of them?

I don't think so.

Managers might say they want data. But actually they donít want data; they want what the data is telling them.

So I find myself using Microsoft Word a lot because I believe thatóas well as the dataóIíve also got to give managers a narrative that joins up the different bits of data that I'm spitting out. Narratives overcome the Battle of Camperdown problem. Narrativesóin my humble opinionóprompt views, discussion, debate. And they help people understand.

And what's the best software medium for weaving together data and text? What's the best medium for telling a story? Well, it's a Word document, not an Excel workbook. Standalone Excel workbooks are for people who know exactly what data they want.

But if you are trying to make sense of a problem that nobody's yet worked out the answer to, or a problem where people disagree about the solution, you don't want data; you want what the data is telling you.

Which means crafting a narrative around the tables and charts that you've pasted into Word.

A few years ago, Stephen Few, a man I often turn to when I'm in search of wisdom on subjects like this, wrote, in an article called Dashboard Confusion Revisited:

The greatest clarification that is needed today is a distinction between dashboards, which are used for monitoring whatís going on, and displays that combine several charts on a screen for the purpose of analysis.

And I reckon that when you place several charts on a screen for the purpose of analysis, you also need to think seriously about whether you also need a written narrative to help the reader make sense of the relationships between all the charts in the display.

[14 February 2014]

Note: I spoke to the manager for whom I was doing the analysis work this week. We agreed that this argument only works for complex problems (what I have called elsewhere the problems that need "there-and-then" data as opposed to "here-and-now" data). For the simple problems, you can just use Excel Ė and indeed thatís exactly what I do when faced with simple, dashboard-type transactions.

 

Make a comment on this blogpost:

Name:

Job Title:

Organisation:

Email address:

 

     
© Kurtosis 2014. All Rights Reserved.    

Comments on this article

14 February 2014:

So true, the analyst role is the cruncher, the assurer, the interpreter, the presenter and the narrator. This is the full analysis cycle but in reality there are data producers, those who churn; and the analysts.

As finance becomes tighter where are the boundaries between the information function and the managerial functions within healthcare? Where does one stop and the other begin? A4C details job descriptions and competencies detail a depth of requirement in knowledge for NHS managers relating to the use of information and the associated skills.

In effect we are all analysts. So: a few questions....

What level of "spoon feeding" do information functions now need to maintain to provide a service?

Is it possible to strike a balance that generates efficiency gains for all parties?

Does "self service" really work?

Chris Gresty

Assistant Director, Health Informatics, North West Ambulance Service NHS Trust

16 February 2014:

Shewhart's First Rule is "Data without context is meaningless" and it is the narrative that provides the context. So often I see data presented without context - and the risk this incurs is that the reader adds their own context, the data is then misinterpreted, an invalid conclusion arrived at; an unwise decision made; and ineffective action sanctioned; and the unintended outcome achieved. Russell Ackoff describes it as the 'content of the human mind hierarchy': Data-Information-Knowledge-Understanding and Wisdom (DIKUW). It is context that converts data to information. Without context-by-narrative the data is just 1s and 0s.

Simon Dodds

System Architect, SAASoft Ltd