Keepin’ the Customers satisfied

As a young lad in Geelong, I sold newspapers, for the Experience. Around that time I bought a 45 single, Bridge Over Troubled Water by Simon & Garfunkel. Bit orchestrated but amazing song, although  I didn’t much care for the flip side ‘Keep the customer satisfied’. I didn’t think it all that good a song, and it didn’t seem to have much to do with customers, or not the sort of customers I was likely to come across across back then.

I like that song more now, but still don’t think it has much to do with customers, apart from the snappy title. Similarly, the great Australian song Esmerelda by the Indelible Murtceps (an anagram of their alter ego Spectrum), sang about ‘always one more customer to go’, but once again, not the sort of customers a newspaper ‘distributor’ or Statistical consultant is (hopefully) likely to meet.

Bridge Over Troubled Water sounds better though, with its message of hope and reassurance (although a some of the words are admittedly a bit dodgy). Reassurance is vital in Statistical Consulting, where clients are often scared of statistics. The other important thing, in any form of consulting, is that clients must feel they come out with something positive that they didn’t come in with (information, a new ‘clever consultant riff they can try in SPSS or whatever), as well as feeling a bit more relaxed.

Whether it be cafe’s or consulting , offering a glass of water or cup of tea or coffee, keeping up an interesting patter about the daily specials, or how it is that median quartile regression may help to make length of stay data clearer, you always got to

Keep the Customer Satisfied!

Power Analysis Surge

December-February (and generally March) is Summer in the Southern Hemisphere. It can get pretty hot even in the more temperate southern states (= northern states in the Northern Hemisphere). Today, the last day of 2015, has an expected top of 40 Celsius, which is 104 Fahrenheit and a nose-peeling 313.5 Kelvin!

Summer is also the start of the Australian Academic Year, and Grant Season, where everyone is looking for Statisticians (who are in swimming pools or pool halls) to run power analyses for them. With no statisticians to be found, it’s off to the library and borrow whatever randoms they can find, or use one of the free web packages, which often works out like a home haircut.

There’s great specific software such as PASS http://www.ncss.com and NQuery Adviser & Nterim  http://www.statsols.com/products/nquery-advisor-nterim/

They’re not cheap (but neither are grants!), and aid professional statisticians as well.  There’s R of course, and the excellent menu-driven power and sample size routines in SAS and Stata.

But first  define the differences you’re expecting, based on the actual Literature as well as Clinical Judgement, and always see a Statistician!

Wisdom of the Cloud

Many summers ago when I started out in the Craft, I could log onto the trusty DEC-20 literally anywhere in the world, and use SPSS or BMDP to analyse data. Nowadays, I have to have IBM SPSS or Stata installed on the right laptop or computer, and bring it with me, wherever I may roam, and wonder dreamily  if I could just access my licensed stats packages from anywhere, like a library, a beach, a forest, a coffee shop.

One option would to subscribe to a stats package in the Cloud! Iin terms of main line stats packages, https://www.apponfly.com/en/ has R (free plus 8 euro’s ($A12.08) per month for platform, NCSS 10 at 18/27.19 per month + platform, IBM SPSS 23 Base 99/149.54 ditto and Standard (adds logistic regression, hierarchical linear modelling, survival analysis etc) for 199/300.59 per month + platform.

Another option, particularly if you’re more into six sigma / quality control type analyses, is Engineroom from http://www.moresteam.com at $US275 ($A378.55) per year.

Obviously,  compare the prices against actually buying the software , but to be able to log in from anywhere, on different computers, and analyse data,  sigh, it’s almost like the summer of ’85!

When I grow up, I’m gonna be a Statistician!

How many of us said that, I wonder? Rather than children dressing up as sheriffs or doctors or possibly even scientists (?), how many dressed up like Statisticians? Did anyone even know much about Statisticians then? Mathematicians yes, they were sort of nerdy (although that word wasn’t around when I was a kid) but could do important things, like calculate odds of winning at Las Vegas or horse racing, and the chance of thermonuclear war.

 

But when I was young, inspired by Get Smart and The Man from UNCLE and James Bond I  mainly wanted to be a secret agent! I played with the idea of  becoming a private detective, sorry investigator, for a while until I found out that in real life, as opposed to TvLand and BookWorld, they mainly seemed to be involved in divorces. So, when I was in my very early teens, I toyed with the idea of joining the FBI. As an Australian citizen, this would have been rather difficult, I would have had to become a US citizen,  as well as either a lawyer or an accountant first. So I put that idea in the ‘too hard basket’. (Imagine, a lawyer or an accountant!).

Well, I suppose it shows evidence of an inquiring mind. Further steps, trots, canters and gallops along the road to Statistics is a story for another time. But there were a couple of ‘residuals’ from that childhood long ago. Asking questions, even if  no one else was. The desire to do the right thing, and wear the right colour hat (even if in truth the Jack Palance baddie wearing black was far cooler/groovier/jazzier in the Shane movie, although not the book, than the light coloured cloth-wearing goodie, Alan Ladd).

And a 1963 book which I got for Christmas a year or two later, called The How and Why Wonder Book of Robots and Electronic Brains. I still have that book and I cited it in my PhD Thesis, although back then I was more interested in the robots, especially the black and red tin ones that could be wound up with a key!

 

But it was a 1979 Texas Instruments TI-55 (simple) programmable LED calculator I got for my 21st, that came with quite a thick manual, showing how one could do fun things like predicting future sales from advertising expenditure, that gave much more excitement, practicality and crunch to the Psych 101 Stats that I was undertaking.

http://www.datamath.org/Sci/MAJESTIC/TI-55.htm

And then, in the early summer of 1981 when I first used SPSS (submitted to be ran at 2300 hours) on a DEC System 20-60 I was truly hooked.

True, James Bond had his Beretta and Walther PPK and Aston Martin and Bentley and Sea Island shirts and Shaken Not Stirred, but at least in the early days, he never used a programmable calculator, let alone a Computer!

 

John and Betty’s Journey into Statistics Packages*

In past days of our lives, those who wanted to learn a stats package, would attend courses, and bail up/bake cakes for statisticians, but would mainly raise the drawbridge, lock the computer lab door and settle down with the VT100 terminal or Apple II or IBM PC and a copy of the brown or update blue SPSS Manual, or whatever.

Nowadays, folks tend to look things up on the web, something of a mixed blessing, and so maybe software consultants will now say LIUOTFW (‘Look It Up On The Flipping Web’) rather than the late, great RYFM (‘Read Your Flipping Manual’).

And yes, there are some great websites, and great online documentation supplied by the software venders, but there are also some great books, available in electronic and print form. A list of three of the many wonderful texts available for each package (IBM SPSS, SAS, Stata, R and Minitab) can be downloaded from the Downloadables section on this site.

IBM SPSS (in particular), R (ever growing), and to a slightly lesser extent SAS, seem to have the best range of primers and introductory texts.
IMHO though, Stata could do with a new colourful, fun primer (not necessarily a Dummies Guide, although there’s Roberto Pedace’s Econometrics for Dummies (Wiley, New York, 2013) which features Stata), perhaps one by Andy Field, who has already done superb books on SPSS, R and SAS.

While up on the soapbox, I reckon Minitab could do with a new primer for Psychologists / Social Scientists, much like that early ripsnorter by Ray Watson, Pip Pattison and Sue Finch, Beginning Statistics for Psychology (Prentice Hall, Sydney, 1993).

Anyway, in memories of days gone by, brew a pot of coffee or tea, unplug email, turn off the phone and the mobile/cell, and settle in for an initial night’s journey, on a set or two of real and interesting data, with a good stats package book, or two!

*(The title of this post riffs off the improbably boring and stereotyped 1950’s early readers still used in Victorian primary (grade) schools in the 1960’s
http://nla.gov.au/nla.aus-vn4738114 (think Dick and Jane, or Alice and Jerry), as well as the far more entertaining and recent John and Betty’s Journey into Complex Numbers by Matt Bower http://www.slideshare.net/aus_autarch/john-and-betty )

Happy Numbers

Why there are so very few statisticians as heroes (or even dashing villains) in novels is a pop culture mystery even bigger than the true identity of reggae magicians Johnny and the Attractions, or the actual final resting place of Butch and Sundance.

I have heard of, but don’t have, the 2008 novel Dancing with Dr Kildare, which features British medical statistician Nina, as well as the Finnish composer Sibelius, and the Tango, by Jane Yardley PhD, in real life a co-ordinator of medical trials for a small pharma.

http://onlinelibrary.wiley.com/doi/10.1111/j.1740-9713.2009.00341.x/abstract.

http://www.transworld-publishers.co.uk/catalog/book.htm?command=Search&db=twmain.txt&eqisbndata=0552773107

I’m now performing statistical consulting at two major hospitals so I’m about to re-read that wonderful book by major scriptwriter / drama writer Jim Keeble ‘The Happy Numbers of Julius Miles’, originally published in 2012 by independent outfit Alma

http://www.almabooks.com/the-happy-numbers-of-julius-miles-p-387-book.html

but there seems to be an April 2014 printing for the US.

It’s a great book about a big fellow, Julius Miles, a professional statistician with Barts Health NHS Trust, Royal London Hospital, Whitechapel, East London, England. Julius loves stats – nose-counting ones such as the fact that it takes him 2 minutes to polish his shoes (with 30 seconds airing between polish, application and buff), as well as meaty methods such as multilevel Poisson regression for length of hospital stay.

Julius is about 1.93 metres (6 foot 4 inches) and wears size 13 (UK) shoes, a solid fellow (although   not reminiscent of the solid Ignatius Reilly in John Kennedy Toole’s classic posthumous 1980  novel ‘A Confederacy of Dunces’).

There’s something about the name Julius, Julius Sumner Miller the US physicist and educator whose ‘Why is it so?’ ran on Australian TV for over 20 years from the 1960’s, and the frothy US drink Orange Julius, named after Julius Freed, around since 1926, taking off in ’29 (the official drink of the 1964 New York World’s Fair).

I can thoroughly recommend this colourful & warm book about Julius Miles, medical statistician.

Give p’s a Chance? Hoochie Coochie Hypothesis Tests

A common request for jobbing analysts is to ‘run these results through the computer and see if they’re significant’. Now, unfortunately, many folk, including scarily, even lecturers in our craft, have a misconception as to what ‘significance’ actually means.

Shout in a desperate monotone “it’s the probability of getting a result as large as, or larger than, what we would obtain if the ‘null hypothesis’ of no difference or association was actually true” and people look flummoxed, yes flummoxed, as if you were speaking to them in the language of the ancient Huns, (another) language no-one has been able to figure out.

True, testing ‘something’ against the concept of ‘nothing’ is a bit kooky. If we really did have a situation where two groups ended up with identical averages we’d think it was a trifle dodgy to say the least.

And as for the notion of effect sizes! Picture, on an enchanted desert isle, two group means of 131.5 and 130, with a pooled standard deviation (sd) of 15. A difference of 1.5 divided by 15, is a Cohen’s (the late great Jacob Cohen; Cohen’s kappa, populariser of power analysis, maven of multiple regression) effect size of 0.10, where given Jack’s arbitrary but conventional guidelines for mean differences, 0.20 is a small effect size, 0.50 medium, 0.80 large.

Using an online calculator e.g.

http://www.graphpad.com/quickcalcs/ttest1/

we find, that if there were 1000 in each group, the t test value would be 2.24 and our p value 0.026.

Voila, Eureka, Significance, as cook smiles and puts an extra dollop of custard on our pudding!

But if we ‘only’ had 100 in each group, our t value would be 0.71, our p value would be 0.48, and there’d be a sigh, a frown, a closing of doors and a grim faced cook doling out the thrice-boiled cabbage….

But they’re the same means, the same sd, and the same effect size!

Coming Up:  Guest Post on a possible, probable, Salvation.

Further/Future reading

G Cumming (2014) How significant is P? Australasian Science, March 2014. p. 37.

http://www.australasianscience.com.au/article/issue-march-2014/how-significant-p.html

also check out Prof G’s website

http://www.latrobe.edu.au/psy/research/cognitive-and-developmental-psychology/esci

with free Excel ESCI program and details of his illuminating 2012 book ‘The New Statistics’.

Now, back to honest resting from honest labour!