Wednesday, November 30, 2016

Thinking about Retirement #32

According to, as of today I have 3,598 days until I retire at the age of 67 - that's just under 10 years from now. This is a long way off and I hope that there is still life in the old dog just yet. My dream is to retire earlier than this, but only if I can afford to. I have a modest pension plan from my previous job in SmartForce, and I have what is regarded as a golden ticket as I am now in a Defined Benefit pension scheme.

In addition to my two pensions (which is two more than a lot of people) I will also get the state pension - which combined may be enough to live on. As long as I can afford petrol for my bike, and wine for my glass - I think I will be OK. (Yes - I know this is not what state pensions are intended for).

But the "experts" tell us that there is a pension time-bomb for my generation (and future generations). Defined Benefit schemes are being closed, or wound down, or closed to new entrants. Other "experts" say that there is no problem. Smarter people than me (like actuaries) make the calculations and forecasts - we must trust and react to their predictions. If an aging population just digs its head into the sand on pensions, younger people who will fund many of our pensions, will not tolerate this. Threats to pension schemes make me think that it may not be the best investment - would I be better putting my monthly pension contributions under the mattress? Only time (and retirement) will tell!

How many of us see retirement.
Image source: US News.

Tuesday, November 29, 2016

The Thanksgiving Effect on my YouTube Channel #analytics #33

Every year at Thanksgiving time there is a drop in daily views of videos on my YouTube Channel. Normally Thursday is one of the best days in a typical week, but in the chart below for November you can see that views on both Thanksgiving and Black Friday have dropped dramatically while Americans were having turkey and going shopping. Looking at the analytics country by country, only in America is there a fall off like this. For Thanksgiving Day itself (24th Nov) the daily view figure was 561 compared to almost 3,000 the previous Thursday - even in the lead-up days there was a steeper than usual decline. In other countries, there is no such dramatic decline. As America accounts for about 25% of all views, this has a big effect on the overall view totals. 

The next big holiday is the Christmas and New Year period, and the annual decline leading into this period will begin in the next week or so. While it is no surprise that views for educational videos drop during holiday periods, I am alway amused (and gratified) that there are still lots of people who watch my videos during holiday time.

Monday, November 28, 2016

Working from Home #34

Today I am working from home for the full day - I rarely do this (I need to be home today as I am expecting/hoping for a call on my home phone). Also - I have no classes today. I dislike working from home - I much prefer going to the office. I like the hustle and bustle of College life, and talking with both students and staff. Many of my colleagues work from home, but it is not for me - I don't really have the discipline for it, there are too many distractions at home. For me, the office is where I should work and I want to keep my work separate from home life.

Image source: Wear Your Voice.
Working from home causes mixed feelings in the workplace. For some, it makes no difference - I once heard a manager say "I don't care where you do your work as long as it is done". Clearly, as a Lecturer, I need to be present in the College for classes, but what about days like today when I don't have any classes? If I don't have any classes, do I need to be in the College? My personal preference is to be in College even if I don't have classes, but not everyone agrees with this. Many people will be more productive working from home. Many live long distances from their office. Some like to be able to prepare classes or correct assessments at home rather than come to the office. Others may be writing research reports or are working on an academic publication.

With technology, we can work from anywhere. I have excellent broadband speed at home (245 Mbps), so I can access all College systems. Most of what I am doing today is grading assignments submitted on Moodle. I would be doing the exact same work if I was located in the office. The major difference, of course, is that I am not available to colleagues and students who may knock on my office door. 

Yahoo CEO Marissa Mayer once tried to ban working from home - she is quoted as saying that it "is critical that we are all present in our offices" and that "best decisions and insights come from hallway and cafeteria discussions, meeting new people, and impromptu team meetings". While I would not prevent others from working at home, I tend to agree with this sentiment. 

Sunday, November 27, 2016

Fixing an iPhone 6s #35

So I had a go at fixing something - replacing an iPhone broken screen. I had some help from YouTube - the Official iPhone 6s Display Assembly Replacement Guide from This is an excellent step-by-step procedure for doing the job. I spent almost an entire Sunday afternoon at this. I found the small screws difficult to deal with but didn't give up. Hopefully, I followed the instructions to the letter. 

It didn't work.

Saturday, November 26, 2016

The Dying Art of Fixing Things #36

Like a lot of people of my generation - I like fixing things. Growing up on a farm I learned to be quite handy with DIY, I can turn my hand to most things around the house. However, this is a dying art - lots of people can't do basic stuff like wire a plug or fix a light switch. We have also become a throw-away society where we will replace something rather than try to fix it. In America, they have a great idea where people can bring things to a Repair Café where volunteer fixer experts can help to fix them - this encourages the art of fixing things, and of course saves money and the environment.

In honour of this, I am going to try and fix something I've never done before - an iPhone 6S. The screen is smashed and needs to be replaced. I bought a replacement screen on eBay and will go through this tomorrow afternoon.

Friday, November 25, 2016

Real Black Friday Bargain @Udemy #blackfriday #37

How did we get into adopting an American phenomenon known as "Black Friday"? How did it spread from America to the rest of the world? And there is still "Cyber Monday" to come? Since the US holiday of Thanksgiving always falls on a Thursday, many also take the Friday afterwards as a holiday to go shopping. Even in Ireland, where we don't do Thanksgiving, we are now inundated with TV, radio, and Internet ads for Black Friday bargains - it's hard to ignore them. Sure - I'd love to spend the day in gadget shops looking for bargains, but like most Irish people, I have to work today. 

However, I did come across a real bargain! Udemy, an on-line education provider, has slashed (oops - I sound like a Black Friday ad!) the prices of courses. One, in particular, caught my eye: Tableau 9 For Data Science: REAL-Life Data Science Exercises. It is just €15 (reduced from €200), so I enrolled in this and a follow-on class to brush up on my Tableau skills. I am already making use of Tableau features to create much better data visualisations than I have ever done before. A true educational bargain!

Here's the Promo video:

Thursday, November 24, 2016

Computer Science in Secondary Schools #STEM #38

The STEM Education in the Irish School System report was published today by the Minister for Education Richard Bruton. The 72-page report into the teaching of STEM subjects (science, technology, engineering, and maths) was commissioned three years ago. How did it take this long? The review group was chaired by Pro Brian MacCraith of DCU.

Image source: Dept of Education.
There are 50 recommendations in the report (of which the Minister says he will act on 21) - the headline one is that "computer science (including coding) as a Leaving Certificate curriculum subject" - this is regarded as "critical to address the ICT skills deficit in Ireland". The report also suggests the establishment of a National STEM Education Research Centre, modelled on similar in the UK.

I definitely have no argument with this and the many other recommendations in the report. It recognises that teachers will be central to any initiative, though one of the proposed initiatives requires primary teachers to "pass all STEM-methodology-related subjects (without compensation) in final examinations, where this is not currently the case" - good luck with getting that past the INTO!

Adding more computer science to the Leaving Cert has consequences for Third-level education where this is typically done today. Very few students going to College know how to code, Colleges would have to adjust so that we are building upon what is learned in secondary school. I would imagine that a Leaving Cert Honours subject in Programming would match a lot of what students learn in first and second year of College - for an honours subject the level would have to be set quite high. If school leavers know how to write code - do they need to go on to College to learn more? Would/could they be hired straight away by companies who have a skills shortage in this area?

Wednesday, November 23, 2016

Top 8 Data Trends for 2016 via @tableau #datatrends16 #analytics #HDSDA #39

Tableau Software has recently published The Top 8 Data Trends for 2016, which makes for interesting reading for all those interested in data science/analytics. These are interesting times for Big Data, and companies like Tableau are leading the way in how we analyse, present, and interpret data. I only use Tableau briefly in my Statistics classes, but next semester I am hoping to be involved in a Data Visualization module and Tableau will be one of the tools I will use.

From the list of the Top 8 Trends according to Tableau, the most interesting to me are:

4: Big data grows up
As Hadoop becomes a core element of IT architecture in organisations, more money is being spent on data security - this will lead to greater adoption in the enterprise.

5: Big Data gets fast
While Hadoop adoption increases, there is a growing demand for fast data analysis as is expected from the traditional data warehouse. New technologies such as Cloudera Impala, AtScale, Actian Vector and Jethro Data, will become more prominent .

6: The number of options for preparing end users to discover all forms of data grows
Tools such as Tableau, a self-styled "Self-service data preparation tool" are exploding in popularity. Put simply, they reduce the time to explore and analyse data. Business users (who are not Data Scientists) now want to also want to be able to reduce the time and complexity of preparing data for analysis, especially when dealing with a variety of data types and formats.

8. The buzzwords converge!
Tableau sees the Internet of Things, Cloud, and Big Data coming together. Tableau predicts that data from devices in the Internet of Things will become one of the “killer apps” for the cloud and a driver of petabyte (that's 1,000,000,000,000,000 bytes!) scale data explosion.

There's no doubt that the Big Data landscape is changing and developing fast. Lots of companies are seeing opportunities to analyse their data to look for something that they did not see before. This drives the technology needed, so institutions such as the National College of Ireland must be prepared to educate the next generation of Data Scientists and Analysts.

The full list of Tableau Top 8 Trends can be downloaded from:

Tuesday, November 22, 2016

How To... Perform a One-Way ANOVA Test in SPSS #HDSDA #analytics #40

So far this year I have released 12 new "How To..." videos - this brings the total to 147 in my channel. Only two of these new videos have exceeded 1,000 views so far this year: "How To... Perform the Mann-Whitney U Test" (6,812 views) and "How To... Perform a Kruskal-Wallis H Test" (3,511) - both were released earlier in the year. This year was also the first time I started a series of videos for SPSS. I had low expectations of high views for these videos, as similar by other content creators on YouTube do not have many views - my expectations have proven to be true. At least they should be of use to my own students.

My latest video shows how to conduct a One-Way Analysis of variance (ANOVA) test in SPSS. This is a statistics technique to test for significant difference between three or more samples. The data in each sample must be normally distributed. The result of the test simply tells you if you have found a significant difference between at least two samples, but not which two. A follow-up test called a Tukey Test is required to tell which two samples are significantly different. Performing an ANOVA testing SPSS is slightly awkward in that you have to first group the data, assign a value to each group, and give each group a label. Here's my short video showing how to do all this:

Monday, November 21, 2016

YouTube Channel - Back in Five Figures #Comeback #41

Last week my YouTube Channel passed 10,000 daily views for the first time since 28th April 2015. As always, I am both delighted and humbled that so many people watch my videos and find them useful. From early 2015 to the end of April 2015, the channel regularly surpassed 10,000 daily views. At the end of April I did some stupid stuff to increase views, including poorly designed thumbnails for each video. This resulted in a collapse in views clearly visible in the chart below - it has taken a year and a half to recover after I reversed these changes earlier this year:

The number of views followed a similar seasonal pattern to previous years, last Tuesday views reached 10,431, and actually surpassed 10,000 views three days in a row. This is still a long way off my record day on 25th March 2014 when the channel had 11,944 views. 

I have added 12 new videos this year, but videos take a long time to get a lot of views - these 12 have just about 12,000 views between them this year. Revenue from ads has also fallen, though it too is showing some signs of recovery in recent months. I'll be interested to see if trends continue. There should be a fall-off this week due to the Thanksgiving Holiday in America, and also it is getting close to the seasonal fall-off for Christmas and the New Year. 

Sunday, November 20, 2016

SmartForce to NCI #42

Around about this time 14 years ago I became a part-time Lecturer at the National College of Ireland. In October 2002 I was made redundant by SmartForce, "the e-Learning Company" - it was an amicable split, I had applied for and got voluntary redundancy. A former colleague told me about NCI and that they were starting up a new Diploma in e-Learning, and I got to deliver my first module. I spent about 14 years with SmartForce, and am now 14 years in NCI. So pretty much my working life so far has been divided into just two companies. Now I am longer in NCI than I was in SmartForce.

Of course, SmartForce no longer exists as it was taken over by SkillSoft in 2002 - the first thing they did after the takeover was fire loads of people. At one time there were over 500 people working for this company in Dublin, now I believe there are less than 50. It was certainly the correct and fortuitous decision for me to leave - I have enjoyed (almost!) every minute in NCI since November 2002, with a bit of luck I will finish out my working days there.

Many people change jobs frequently - good for them. When I was leaving school (1978) a job-for-life like the Civil Service or teaching was the thing many school leavers were aiming for. I recall my Grandmother Kathleen O'Loughlin urging me to apply for the bank - I'd be set up for life. She didn't really believe in all that College stuff (Seminaries excepted!). I would absolutely hate to think that I had just spent nearly 40 years working for a bank. 

Saturday, November 19, 2016

The End of the Battle of the Somme #43

Yesterday (November 18th) marked the 100th anniversary of the last day of the Battle of the Somme. By the end of the battle, the British Army (including thousands of Irishmen) had suffered 420,000 casualties including nearly 60,000 on the first day alone. The French lost 200,000 men and the Germans nearly 500,000. Nothing was achieved by either side, as bad weather ended the battle. 

Image source: The History Learning Site.
In our family, we still think that my great-grandfather James Byrne fought at this battle and may have been gassed. We know he survived the war as he died in 1925. Despite long searches, we have not found any record of his service in World War I. His name is very common, and there is the difficulty with the spelling of his name - it is both "Byrne" and "Burns" in the various genealogy sites where I have found records of him. For example, here is a copy of the registration of his marriage to Margaret Colborn (also spelt "Cobourne" and "Coburn") that I found today:

He lived in Lorrha in North Tipperary, his house was located where a grotto to Our Lady is now to be found - here it is on Google Street View:

Most of the men who enlisted from his area did so in Birr, Co Offaly. In the main, they enlisted into the Irish Guards (who were at the Somme) or the Leinster Regiment. I got this clue from Gerard O'Meara who is launching a new book "Lorrha People in the Great War" this month. He is an expert on the World War I soldiers from that area, but he too has no record of James Burns/Byrne. He was not a young man when he joined up - he was 41 years old, and had a family with seven children (two more were born after the war). He worked as a farm labourer all his life. No doubt he was poorly paid and may have found the "King's Schilling" attractive, plus of course the home payments to his wife Maggie.

Image Source: Ask Ideas.
On this 100th anniversary of the end of the Battle of the Somme, I wonder what he was thinking in 1916. Was he terrified? Relieved? Shell-shocked? Was he thinking of the aftermath of the Easter Rising a few months earlier? How many of his friends were lost in the battle? Was he longing to be home in Lorrha? Could he bear the separation from Maggie and his children any longer (including my grandfather Paddy)? Did he kill any Germans?

In memory of James Byrne - I am posting a poppy on my blog for the first time.

Friday, November 18, 2016

Missing NCI Graduation Ceremony #NCIGrad2016 #gutted #44

Graduation 1988.
For the first time in my 14 years at National College of Ireland, I will not be attending our annual Graduation Ceremony being held at the National Convention Centre today. I have an almost full schedule of classes during both the morning and afternoon conferrals, so this year I am working with next year's graduates!

I have written in the past about how Graduation is my favourite day of the academic year. I love seeing graduates dressed in their finery, meeting with their families, and sharing in the celebration. I like to think that of the several hundred students graduating who were in my various classes, that I had a small part to play in their success. I get a special satisfaction seeing students who struggled throughout their time in College finally make it to Graduation - no matter how long or difficult it was, they finally made it. I have fond memories of my own graduation - though now it is nearly 30 years ago.

My message to all graduates, but especially to those who were in my classes, is to first - enjoy the day and celebrate a well-deserved achievement. Secondly - though for many this day may be the only time that they graduate with gowns and lots of formality, for most there will be many other forms of graduation to come. Postgraduate study is a path that many will travel, but others will complete more courses, and mark other special achievements. Graduation is not just receiving a parchment with your name and degree on it, it is also the point for many who will graduate from formal education for the first time and move into the workplace. Lifelong learning means just that - and it continues from today.

Many congratulations to all NCI graduates today!

Thursday, November 17, 2016

Europe for Big Data via @forbes #BigData #HDSDA #45

Law firm William Fry, in partnership with Forbes Insights, surveyed 200 senior business executives from leading companies around the world to explore the business challenges and opportunities that big data offers. They produced some interesting findings - especially as 96% viewed Ireland as a "favourable investment location". 82% rated Ireland’s data privacy regulatory regime as good to excellent - we are therefore "well placed as a jurisdiction of choice for the location of data operations for the European region" according to Forbes. You can get a full copy of the report here.

The survey also shows that "tax rate" is the "number one most important location issue for data-driven investment" (Duh!) followed by "ease of doing business", "legal framework", and "talent". The report is an excellent read for Data Analytics students and is also a comfort to see that Ireland is up there with the best with regard to Big Data. 

Some figures from the report...
Image source: Forbes.

Wednesday, November 16, 2016

Your Data Belongs to You! Part II @fitbit #analytics #46

Following on from yesterday, I am taking another look at personal data gathered by my ChangeHR Fitbit. Yesterday I compared my "Active" to "Sedentary" activity to discover that I am not very active. The Fitbit breaks activity down into three types: minutes "Lightly", "Fairly", and "Very" active. When I compare these three meaures for last month together I get the following chart:

As I am not very active it is no surprise to see that most of the activity that I do have is "light activity" (blue line). As I found out yesterday, this should be a motivation for me to get more active - especially to try to be more fairly/very active.

I also wear the Fitbit when I am asleep - it can tell when I am asleep (presumably by matching the time with long period of inactivity associate with sleep). 8 hours (480 minutes) is the default recommendation for a good night's sleep, and most of the time I achieved this. Some days at the end of October are missing data. 

Not seen here, but the Fitbit Dashboard will also tell if my sleep is disturbed or if I am awake. Sometimes I feel in the morning that I have gotten very little sleep or that I have been "awake all night" - the Fitbit can track this, and often it tells me that I was only disturbed or awake for very short periods contrary to what I thought.

Some day in the future, probably after about a year of wearing the Fitbit, I'll check the data again and see what a year's pattern looks like. I really like Fitbit making this data available under the "Your Data Belongs to You!" principle. How often can we get all data about ourselves? Eh Google? Facebook?

Tuesday, November 15, 2016

Your Data Belongs to You! Part I @fitbit #analytics #HDSDA #47

Since I got my new ChangeHR Fitbit last month I have been wearing it most days and tracking things like the number of steps I take, distance walked, calories burned, minutes active, sleep patterns, and much more. I started to wonder where these data are saved on my computer? They are not. The data are saved in the "cloud", but Fitbit have a "Your Data Belongs to You" policy and you can download CSV or Excel files with your data for analysis. Fitbit provides fairly decent dashboards to read the data, but it can be fun to do your own analysis. I looked to compare one set of data with another for the past month.

The first chart below compares the number of steps taken each day with an estimate of the number of calories burned. Not surprisingly, the trends follow each other closely:

I really don't know how accurate the Fitbit is as detecting steps - when I brush my teeth it racks up 300-400 "steps". When I ride the 7 miles to work on my motorbike, the vibrations cause it to record over 2,000 "steps". 10,000 steps a day is the recommended minimum for a healthy life style - I see I have fallen below this regularly in the past month.

The second chart below is a little more interesting - it shows the number of minutes active (brown line) compared to sedentary (blue). Again I question the accuracy because each day should total 1,440 minutes (24 hours * 60 minutes). While many days do add up to 1,440 minutes below, others don't. Nevertheless, the surprising thing for me here is to see how much time I spend inactive! The blue is very dominant. 

While I am not surprised that I am more sedentary than active, this shows the stark contrast between the two measures. Have to do something about this! Perhaps a device like a Fitbit is in itself a motivational force to make us (me) exercise more?

The above is only a small sample of the data gathered for just one month. Imagine years of data, or even a lifetime of data recorded. The benefits for each wearer's health should be great, but the contribution to overall health science in the future will be massive.

Monday, November 14, 2016

How To... Select the Correct t-test to Compare Two Sample Means #48

One of the confusing things that takes a lot of time to explain in my Statistics classes occurs when we are discussing Student's t-test. This is a test to compare two sample means for a significant difference, and it was invented by Irishman William Sealy Gosset in 1908. There are three possible formulas for calculating the t statistic which is then  used to determine if a difference between the two samples has been found. My new video maps out the path to determine which test to use - it is important to choose the correct path, otherwise conclusions based on the data analysis could be false.

Sunday, November 13, 2016

10th Anniversary of this Blog #49

Exactly ten years ago I started this blog. My first post on 13 November 2006, was very short:

My First Post

It's taken a long time but I've finally created my own blog. Here I hope to publish some thoughts, articles, reports, short stories, other publications - all of my own.

This is my first post!

1,820 posts later I am still at it - that's close to one post every two days. At the time I had no idea that blogging would become a huge part of what I do. Over the past ten years I have posted at irregular intervals. During 2011 (when I had been nominated for the Irish Blog Awards) I posted quite a lot - about 20-25 posts per month. Since 01 January 2016 I have been posting every day in an effort to see if I could keep this up for a year. - just 48 more posts to go! The chart below tracks monthly posts since I started:

This year has so far been the most prolific, though I freely admit that on many days I was very stuck for something to write. Since 2011 the trend was declining on a monthly basis - I guess the effort to post daily was a reaction to this. 

Not that many people read this blog - I remind myself that the word "blog" is short for "Web Log" - a sort of on-line diary. So it is a very personal document, that just happens to be public. It's for me, and I like the freedom to be able to express myself like this. Quite a few people that see my posts do so in Facebook and LinkedIn - each post is automatically published to both these platforms (and to Twitter as well). This year I have my Mum and Dad reading this blog for the first time (Hiiiiiii!) - apart from occasional bad language and some shite (sorry!) posts, they like reading it.

In 2006 I set up the blog when I was teaching on the MSc in e-Learning that at the time was running in the College. It was a relatively new thing at the time as Web 2.0 started to allow more and more people publish content on-line. My blog is hosted on Google's Blogger platform. A lot of bloggers use WordPress, but I am happy with Blogger. At the time we were discussing in class how blogging could be used in education. Will Richardson's "Blogs, Wikis, Podcasts, and Other Powerful Web Tools for Classrooms" was the textbook we referred to most. None of us could have predicted that people would be making money out of blogging (I don't), and there would be professional bloggers breaking the Internet. 

Anyone can blog, and there is no failure. I like to think that I will keep on blogging for a long time to come. Actor Kirk Douglas will be 100 years old on 9th December next - he must be one of the world's oldest bloggers, He used to blog on MySpace, but now he blogs using the Huffington Post. Just last September he blogged about the the US Presidential Election. When he was born, Woodrow Wilson was President! If I survive as long as Douglas, I have 43 more years of blogging to go. Hopefully I will be able to say the same as him - he wrote in his last post:

My 100th birthday is exactly one month and one day after the next presidential election. I’d like to celebrate it by blowing out the candles on my cake, then whistling “Happy Days Are Here Again.”

Saturday, November 12, 2016

Nothing beats a day with your Mum and Dad #Family #50

There is nothing that will separate us from where we grew up. I grew up in the townland of Ballingate, which is about three miles (5km), from Carnew in Co Wicklow. Today on an autumnal day in the sunshine, Ballingate looked great - I snapped the photo below of a maple tree in the garden.

I visited with my Mum Phil, my Dad Joe, and my brother Joe, for lunch and to help with cutting logs in the bog. Cutting logs ("sticks" as we call them) is a hard working activity that requires a nice pint afterwards to slate the thirst of those involved in the heavy work. We adjourned to Conways in Kildavin for a very pleasant jar. Happy Days!

Friday, November 11, 2016

Losing it in class #51

In one of my classes today I had to ask myself "Why the fuck do I do this?"
I didn't apologise for saying this
The click click click of mouse activity was constant,
There was constant chatting
There was lots of attention being paid to computer screens 
(I'm certain that this was not related to class activity)
Very few were paying any attention
I refuse to police this

It was 16:00 on a Friday afternoon
The College was almost empty
I regularly asked students not to distract others beside/behind them
I regularly asked for the chit-chat to stop
I was ignored
All the time
Why the fuck do I do this?

I want to make a difference
I can for some
But not for all
I was not happy
I showed this
I was not polite
I lost it

I'm sorry.

Thursday, November 10, 2016

A Vision of Students Today #52

In 2007, Michael Wesch of Kansas State University published a video that struck me at the time as a landmark video about the use of technology in the classroom. Prior to this, I used to stop the few students who had them from using their laptops in the classroom - on the basis that they were disrupting the class. Facebook was strictly forbidden. But - after seeing Wesch's video, I did a complete about-turn and allowed students to use whatever they wanted to in class. My reasoning for this is that they could be using technology (including Facebook) for learning. I do advise that no one should disturb or disrupt others in the class - occasionally some students try to test me on this and how far I will let them go. However, I don't police the classroom any longer and only admonish students if they appear to be disturbing others.

I feel that Wesch's video, though nine years old now (and watched over 5.2 million times), still holds some truths for us all today. Here it is...

There is a great quote from Josiah F. Bumstead in the video:

The inventor of the system deserves to be ranked among the best contributors to learning and science, if not the greatest benefactors of mankind 

This quote is from 1841, and the "system" referred to was a blackboard. The book by Bumstead that this is taken from is called "The Black Board in the Primary School: A Manual for Teachers". Elsewhere in the book, Bumstead recalls asking a Clergyman on a school committee if the school had a blackboard. "No" replied the clergyman, "it is of no use to get them. If we had blackboards, we have no teachers to use them to advantage". Bumstead was of course astonished at this (so he wrote the book) - 175 years later we should be equally astonished if our teachers could not use technology to advantage.

Wednesday, November 09, 2016

How did the Pollsters get the USA Presidential Election so wrong? #OpinionPoll #Fail #HDSDA #Analytics #53

Well - there goes one of my topics in my Statistics classes. When introducing the topic of random sampling, I often refer to opinion polls to illustrate how an inference about a population can be made with a sample. I pointed to the incredibly accurate opinion polls for the 2012 US Presidential Election  - but yesterday's election (plus the recent Brexit vote) are causing Opinion Poll companies to reconsider their strategies and techniques.

Image source: Well This is What I Think.

Nathan Bomey, writing in USA Today, asks "How did pollsters get Trump, Clinton election so wrong?". Various arguments are put forward: "some voters were apparently sheepish about admitting to a human pollster that they were backing Trump", that "many pollsters may have incorrectly ruled out the prospect that people who didn't vote in 2012 would nonetheless cast ballots in 2016", and that pollsters "overestimated Clinton's support among minorities and underestimated Trump's support among white voters". 

All pools seemed to predict a comfortable win for Clinton, with one exception: the LA Times/USC poll. The LA Times polls saw "what other surveys missed: A wave of Trump support". Their polls show increasing support for Trump after the 3rd Debate with Clinton.

Here's what the LA Times says about why their methodology is different from other polls:

The poll asks a different question than other surveys. Most polls ask people which candidate they support and, if they are undecided, whether there is a candidate they lean to. The Daybreak poll asks people to estimate, on a scale of 0 to 100, how likely they are to vote for each of the two major candidates or for some other candidate. Those estimates are then put together to produce a daily forecast.

Basically, in the LA Times poll, someone who is 100% sure of their vote counts more heavily than someone who is only 60% sure? And someone who says she is 100% certain to vote weighs more heavily than someone only 70% certain. This is an interesting way to gauge public opinion, and their's is an on-line poll rather than a face-to-face or telephone poll. They also got the 2012 election almost spot on - they predicted a margin of 3.32% for Obama over Romney - the actual was 3.85%. This is bound to influence the way polls are taken in the future, and in turn cause questions to be asked about inferential statistics.

Tuesday, November 08, 2016

How To... Calculate Student's t Statistic (Paired Samples) by Hand #Analytics #Statistics #54

One of the easiest statistical tests to carry out using pen and paper is the Paired (Dependent) t Test. This is a test to find a significant difference between two sets of sample data (normal distribution assumed) - the easiest way I know to remember what is needed for this type of test is that you have one group measured twice. Most commonly this would be a "before" and "after" situation. This is different from an unpaired (independent) t Test where you would have two groups measured once each.

You don't neet to calculate the mean, variance, or standard deviation, which can be time-consuming (especially in an exam situation). The solution is based on the differences between each pair - so if you can add, subtract, divide, and get the square root of a number, you can do a paired t Test. Here's my short video showing how to do this test with a small sample by hand:

Monday, November 07, 2016

No Learning and Teaching in some schools today #55

Learning can take place without teachers - it happens every day. However, as a civilized society we choose to structure education with schools and classrooms populated by learners and teachers - so far it is the best way we have come up with to teach and learn subjects like maths, science, history, and many more. While technology enables on-line learning and teaching to take place, our education system is a long way away from providing this at second level education in this country.

Image Source: RTÉ Morning Ireland Twitter Page.
I take no side in the current teachers' dispute. I don't work at second level and it's not for me to argue the merits of claims and counter claims. I am not a member of a union, and have no plans to join one. News reports are increasingly and repetitively dominated by pay claims and restoration by public sector unions - will it ever end?

The sad thing about all of this is that learning and teaching is not taking place today in many schools. For some learners it will be a welcome day off, for others it will be of little consequence as it is just one day (so far), and for many there will be the added worry of missing out on learning in a Leaving Certificate year. For some teachers it will be a welcome day off, for others it will be of little consequence as it is just one day (so far), and for many there will be the added worry of missing out on teaching in a Leaving Certificate year. There are no winners, just losers.

In the end, this dispute will be resolved - they always are. Money will fix this - it will have to be either borrowed, raised from taxes, or some other public service will have to be cut. Whatever needs to be done to restore learning and teaching in our schools should be done, at whatever the cost is to be. Then there will be no losers, just winners.

Sunday, November 06, 2016

Who will America pick for President? #56

This week, Americans face a choice of Donald Trump or Hillary Clinton to be their next President. I genuinely hope that whoever gets the job that they will be up to it. If I was an American with a vote, I definitely would find it very difficult to vote for either character - I don't like either of them.

Image source: The Fiscal Times.

I think Hillary will win, and win big. The swing states that decide US elections seem to be in her favour. According to the 270 to Win website, Clinton has 252 Electoral votes against Trump's 163, with 123 electoral votes still to fight for. She needs just 18 more votes out of the 123 - she can't lose. What type of President will she make? I don't know. I often feel that candidates promise a lot, but when they get into office they discover that they can't make much difference as the "system" changes little.

I'm sure Americans would prefer true leaders to come forward at election time. In the election Primaries, Bernie Sanders stood out. Also, current Vice-President Joe Biden would have made an excellent candidate. How Americans would wish for a Washington, Lincoln, Roosevelt, or a Kennedy to step forward. I have no doubt that the next President will be in office for just one term. It's up to the Republican party (who since Ronald Reagan have just come up with the Bushs) to find a candidate capable of taking on Hillary in the 2020 Election.

God (really needs to) Bless America!

Saturday, November 05, 2016

The Mellon Educate Building Blitz #57

This year's Mellon Educate Project is to rebuild and renovate two primary schools (Ummangaliso and Kuyasa) in Khayelitsha, Capetown. One of the 270 or so volunteer builders is (for the sixth time) my lovely wife Roma. The Building Blitz starts today, many volunteers are already there. Together with several other volunteers, she set out from Dublin (via London) for the long trip to Capetown. They can look forward to a very long flight (11+ hours) which will, in turn, be followed by six days of very early mornings and long days of hard work. 

Education is the most important and valuable gift anyone can give a child. In the video below (from Mellon Educate) you will see children from Capetown who want to grow up to be doctors, teachers, and lawyers - just like children all over the world. But their school infrastructure lacks the resources they need, especially in science and technology, to give the kids the opportunities they want. The work that Roma and her volunteer colleagues are doing is vital to contribute to better education facilities for the children in the Ummangaliso and Kuyasa schools.

Over the past year, Roma has been raising funds from friends, neighbours, and many others who have contributed generously to the Mellon Educate cause. Indeed at times our house has looked like a clothes recycling plant as she collected old clothes to raise hundreds of euro. All in a good cause.

Go Roma!

Friday, November 04, 2016

How People Learn at Work via @degreed #LifelongLearning #58

I've been reading about interesting research carried out by a company called Degreed who I had not heard about before, but have been around since 2012. Degreed claims to be the first company to provide a lifelong learning platform to track all learning, not just degrees and other formal qualifications. Sounds like an interesting idea, long overdue - as part of their launch statement states "Imagine if your diploma actually measured EVERYTHING you have learned!". The research I mentioned above backs up what their mission is. This research was a survey of 519 respondents in the USA, mostly Millenials (57%) and Generation X (42%) - just 1% were Baby Boomers (that's me!).

The "HOW THE WORKFORCE LEARNS IN 2016" report shares some interesting results from Degreed's research. Almost 85% of respondents said they "learn things for work by searching online at least once a week. Nearly 70% learn from peers or by reading articles and blogs every week, and 53% learn from videos in any given week". Another really interesting find was that workers "spend about 1% of the average work week (37 minutes) on their employers’ training", but that they "invest 3.3 hours a week on their own". We don't confine our learning to "offices, shops, factories and warehouses—or to “normal” working hours". A whopping 85% of people in the survey said they learn at work - no surprise, but get this: "67% do so on personal time and 18% are learning during travel or commutes". Here's a graphic from Degreed showing how we learn (according to their survey):


The breakdown above is interesting, though not surprising - web search and video are now central parts to our everyday learning. As you can see, most types of learning are "self-directed" and are more frequent. It makes sense for both learners and employers to try and capture all of this. Degreed recognises that there is no single path to expertise - we become experts in our chosen fields through many different paths. Degreed integrates many different sources under a single platform - it would be interesting to see this . Here's a flavour of what they do:

Thursday, November 03, 2016

Computers for students in Kenya #SiliconSavannah #59

According to news reports on radio this week, Kenya is rapidly becoming one of Africa's leading technology-driven economies. The impressive Konza Technology City project is planned to be a world-class technology hub and a major economic driver for Kenya. Clearly, the Kenyan Government, in difficult circumstances, sees that the future for its citizens will be enhanced by technology. Good for them and I hope that it works out well for not just the people of Kenya, but everyone in Africa.

Of course, one of the things that Kenya will need to do to support projects like Konza, is to prepare its children for a modern economy. Will Goodbody of RTÉ tells us this week that there is an initiative in Kenya where "every child starting school gets a tablet is giving children there an opportunity to catch up with their digital peers internationally" - a simple (though not novel) idea. This is a project that has resulted from an election promise being delivered (now there's a novel idea!). In total 1.2 million tablets will be delivered to 24,000 primary schools in Kenya at a cost of about €150 million. In a country with ten times (44 million) the population of Ireland (where our education budget in 2016 is just over €9,000,000,000) - this seems to be fantastic value for money. 

I wonder of the Kenyan Education Minister would like a job here?

Image source: RTÉ.

Wednesday, November 02, 2016

Ooooooh! - New (work) PC! #60

It's goodies time for me in NCI today as I have received a new desktop PC with a touchscreen monitor. I had decided that I no longer needed or wanted a laptop. The old one was very big and heavy, and I almost never took it out of the office. I never brought it home, nor did I ever bring it to classes or to meetings. In short - I did not need an expensive new laptop. My only luxury with the new (tower) desktop PC is that I requested (and got) a touchscreen. The 22" monitor is brilliantly clear and with Windows 10 I hope to make more use of touch. 

Most software that I need and use is already installed on the PC, the biggest piece of work is syncing accounts. Google Drive has to sync 81,901 files which will take a long time. I am holding on to the old laptop for a while to ensure that I match up all files and applications. 

I can remember the days when it took a long time to set up accounts and install software - it could easily use up a day or longer of work time. Within minutes of getting this new PC, I was able to get working again. 

Thank you, NCI!

Tuesday, November 01, 2016

Where did the title of this blog "Careful With That Axe, Eugene" come from? #61

Almost four years ago to the day, I renamed this blog from "Eugene' Blog" to "Careful With That Axe, Eugene". I was recently asked where did I get the title from, and coincidentally I heard it mentioned by Dave Fanning on RTÉ Radio 1 yesterday when he was discussing the 1970 movie Zabriskie Point - a film I'd never heard of but which has actor Harrison Ford in an uncredited role. Music for the movie was by Jerry Garcia (Grateful Dead) and Pink Floyd. "Careful With That Axe, Eugene" was re-recorded by Pink Floyd for the movie and given the title "Come In Number 51, Your Time Is Up". According to IMDB, the movie was not a success - it cost an estimated $7,000,000 to make, but grossed just less than €1,000,000. The main star of the movie, Mark Frechette, was just 20 when he made the movie, but he later died in jail in 1975 (he was convicted of robbing a bank). He died in a weight-lifting accident when a 150-pound weight fell on his neck and suffocated him (foul play was not suspected). The female lead in the movie, Daria Halprin, was married to Denis Hopper from 1972 to 1976.

Image Source: Wikipedia.

The original track is mostly instrumental with lots of whispers and some screams by Roger Waters. In the version below on YouTube by a tribute band, the line "Careful With That Axe, Eugene" is mentioned at about 5 mins 50 seconds into the video. A link to the original by Pink Floyd is here - this is protected by copyright and will be blocked if I embed it here.

Originally, the song was a B-side to the single "Point me at the Sky", it also features on two Floyd albums: "Relics" and "Ummagumma".