Random Analytics

Charts, Infographics & Analytics. No Spinning the Data. No Juking the Stats

Month: July, 2014

Random Analytics: Ebola in Liberia (to 30 Jul 2014)

I have now been doing Random Analytics since October 2012 and analytics on Ebola itself was my sixth post (having been interested in the subject since I read The Hot Zone in the mid 1990’s. I’ve been doing complete posts on the subject recently, by impacted 2014 country, by African exposure, by top 10 outbreaks and by classification and year but this outbreak is moving faster than I can keep up with. I’ll now just concentrate on infographics that don’t get done in the mainstream narrative (or my flublogist colleagues don’t do). If you want to follow the detail more closely then follow Crawford Kilian, Ian Mackay or Maia Majumber (who does the dangdest infographics in the flublogist space).

Ebola in Liberia (to 30 July 2014)

01 - Ebola_Liberia_140730

***** Please note that this EVD infographic was updated with public source information to 2100hrs 30 July 2014 (EST) *****

As I go to print the overall Ebola situation can be found in the latest World Health Organisation advice (circa 27 July 2014). This infographic comes from the latest information from the Liberian Ministry of Health and Social Welfare via reliefweb. Situation Report No. 72 on the EBOLA Virus disease epidemic in Liberia as of 29th May to 28 July 2014. Detail:

Highlights

Lofa County

Bakedou

Citizens and CHSWT have reached a compromise to allow health workers to investigate and follow up cases. Unofficial: About 15 persons are currently ill.

Voinjama

One of the two patients in the holding room left against medical advice and relocated to Zorzor Curran Hospital. However, specimen was taken.

Bong County

There no lab technician to collect specimen

Bomi County

There is no trained personnel to manage cases – Also, there is no lab technician to collect specimen

Nimba County

Family of confirmed case continuously refuses to allow patient to be carried to treatment facility

Advertisements

Random Analytics: MH17

1 - MH17_Infographic_140722

The above infographic displays the known nationalities of the 298 passengers and crew aboard Malaysian Airlines flight MH17 which departed from Amsterdam and was en-route to Kuala Lumpur. Details of the nationalities are primarily sourced by the Malaysian Airlines statement (released 18 Jul 2014) and further updates via ABC (19 Jul 2014) and the BBC (22 Jul 2014).

Depending on where you are in the World you have either awoken or are going to sleep after hearing about the terrible tragedy that has occurred near Shakhtars’k, Ukraine. Via the Australian Broadcasting Corporation. Live: Australians among dead after Malaysia Airlines jet shot down by missile over Ukraine. Excerpt:

At least 27 Australians have died on board a Malaysia Airlines passenger jet that was shot down by a ground-to-air missile over Ukraine, killing all 298 passengers and crew.

Flight MH17 was en route from Amsterdam to Kuala Lumpur when it came down in rebel-held territory near Ukraine’s border with Russia. US vice-president Joe Biden says it was “blown out of the sky” and that it was “not an accident”.

Prime Minister Tony Abbott has told Parliament it appears that Russian-backed rebels shot the plane down.

Ukrainian wire taps appear to have captured pro-Russian separatists claiming responsibility for downing the jet, but there has been no official confirmation.

 

To the Lost.

Random Analytics: 44th Parliament 94(a) and 94(b) ejections (to 8 Jul 2014)

1-BronwynBishop_viaABC

Here is a look at the Standing Order 94(a) and Standing Order 94(b) ejections for the 44th Parliament of the Commonwealth of Australia House of Representatives to the end of the 14th week (8 July 2014).

2 - 94aEjectionsbyPerson_140714

The first chart details the list of those ejected from the house under Standing Orders 94(a). At the end of the 14th sitting week there were 144 ejections, of which 141 (98.0%) were from the Australian Labor Party, two (1.4%) were from the Liberal National Party and just the one (0.6%) for the National Party.

The Top-3 from the ALP:

3-NickChampion_viaAPH

Nicholas David “Nick” Champion MP (20-ejections) is the ALP Member representing the electorate of Wakefield (SA). He won the seat at the 2007 Australian federal election.

4-MarkDreyfus_viaAPH

Mark Alfred Dreyfus QC MP (12-ejections) is the ALP Member representing the electorate of Isaacs (WA). He won the seat at the 2007 Australian federal election. He is the former Attorney-General, Minister for the Public Service and Integrity, Minister for Emergency Management, and Special Minister of State in the Second Rudd Ministry.

5-PatConroy_viaAPH

Pat Conroy MP (11-ejections) is the Member representing the electorate of Charlton (NSW). He won the seat at the 2013 Australian federal election.

From the Coalition:

6-EwenJones_viaAPH

Ewen Jones MP (2-ejections) is the Member representing the electorate of Herbert (QLD). He won the seat at the 2010 Australian federal election.

There have only been two Standing Order 94(b) ejections thus far. They are Mark Dreyfus (ALP) and Wayne Swan (ALP).

7 - 94aEjectionsbySittingDay_140714

The final chart is a look at the Standing Order 94(a) ejections by sitting day. As you can see they tend to pick up in the second last week of each large sitting period or at the end of the calendar year. In this update I’ve been able to include some colour as three ejections have come from the Coalition.

Image Sources: Bronwyn Bishop’s image care of the ABC while photos of individual members are taken from their official Australian Parliament House photo.

Random Analytics: MERS by Occupation (to >375)

Since my last Middle Eastern Respiratory Syndrome update (21 May 2014) there have been a number of key developments and even some improvements in data quality coming out of the Kingdom of Saudi Arabia.

The big announcement since my previous post was the addition of 113 legacy cases by the KSA Ministry of Health. Of the 113-cases, 42 were Health Care Workers and are included in the following charts.

Almost as importantly, the Saudi’s are now sharing their data with the World Health Organisation (WHO). The Disease Outbreak Notifications, or DON’s, are very comprehensive, light-years away from the Health Ministries updates of the past. With some irony I was pleased to see that Iran and the Kingdom featured together in the most recent DON. Perhaps we are seeing a form of MERS diplomacy occurring.

Here are the latest MERS by Occupation charts.

***** Please note that all infographics for this MERS-CoV article are using publically sourced information to 1200hrs 4 July 2014 (EST) *****

 

01 - MERSbyJobTitle_140704

This first chart looks at those infected with MERS by Job Title or Function.

Key Notes:

  • Retired: The largest group. There are 150-retirees (39.5% of known job titles) represented in this chart but only 2 have been confirmed (1.3%). The bulk of the retirees represented in the chart are included if they did not have a job title or function attributed to them AND if their ages are greater than the official retirement age for their home country.
  • Health Care Workers (HCW): The second largest group. Includes all types of unidentified workers in the Health sector (i.e. Nurses and Doctors).
  • Nurse: I have been able to identify 23-Nurses and in at least two cases, their speciality (ER & ICU).
  • Farmer: Includes both Owners (9/75%) and Employees (3/25%). I suspect the higher weighting toward owners is due to the fact that they are all nationals (from KSA, Qatar and the UAE). The three farm employees that have been identified are all resident workers. Just a thought here. Rich owners get to see the doctor while residents might have a range of barriers which reduce their ability to receive primary care services or choose to work through what they might believe is a bad flu.
  • Pilgrim: Of the 11-Pilgrims I have been able to identify I believe at least 8 were Umrah linked while three were potentially due to the Haj.
  • Doctor: Six identified, including one surgeon and one ICU specialist.This infographic looks at those infected with MERS-CoV by Job Family. In short I think this is a key infographic for MERS as it gives you some confidence in the key narratives (i.e. that Health Care Workers are over represented in the data as an example).

Next chart, Job Families:

02 - MERSbyJobFamily_140704

Key Notes:

  • With the inclusion of an additional 42 Health Care Workers the Non-Participatory (156/18.6%) group (Paediatrics, Students, Retiree’s and the Unemployed) move from the largest to the second largest Job Family.
  • Health Practitioners/Technical Operations (159/19.0%) or HCW as there more commonly known become the largest Job Family represented. This number includes the Nursing Assistant that was identified in Iran (but more on that later).
  • Paediatrics (18/2.1%) numbers have declined since the last update when they represented just 2.8% of the data then. Still seems low and Maia Majumder picked up on this in a recent post.
  • Pilgrim/Tourist (14/1.6%) has seen a slight increase due to some Umrah inclusions recently.
  • Healthcare Support (6/0.7%) numbers remain static so not sure if the Saudi announcement of legacy cases conflates HCW and HCSpt numbers.
  • Construction (2/0.2%) is a new inclusion from the previous update. Given the amount of building going on the in the Middle East, especially in Qatar this number seems on the very low side. I’d expect to see this number increase with more robust reporting.

Last chart.

03 - MERSbyMainJobTitle_140704

The last chart looks at those overall main job families that are most impacted by MERS, specifically Farmers, Travellers, Paediatrics, Retirees, HCW & HCSpt (combined), Other and Unknown.

Key Notes:

  • Farmer (1.7%): With only 14 confirmed cases apart from 2013 you can barely see them across an entire year, quarter or month. Numbers seem low.
  • Traveller (1.7%): Like farming, numbers seem low.
  • Paediatrics (2.1%): As suggested previously, no new paediatric cases since my last update so the numbers have declined somewhat.
  • HCW & HCSpt (19.7%): Health Care Workers and I have also included Health Care Support Workers in this grouping as well. Numbers up on previous update due to the additional 42-cases.
  • Other (2.3%): All other occupations that have been publically released. I’ve actually reduced the number in this group by one from the last update due to improved reporting from Saudi Arabia.
  • Unknown (54.7%): Unknown occupations. Up slightly but with improved reporting I’m hoping that this will reduce (over time).

Final Thoughts (on the difference between a Health Care Worker and Health Care Support

Last month I tweeted that the Iranian Nursing Assistant (FT #827) should be counted as a Health Care Support worker rather than a HCW. I then got a number of return tweets from the likes of Helen Branswell, Ian M Mackay and others who disagreed with that line of thought.

When Helen and Ian ‘guide and advise’ it’s probably worth not disregarding that advice. Upon some personal review I decided that perhaps I had taken a too hard Workforce Planning line to my job functions without fully considering the clinical implications.

I have subsequently reviewed my thinking and have re-organised my data along the following lines.

Health Practitioners/Technical Operations (nee HCW) are any job title or function that is included in the Bureau of Labor Statistics SOC Occupations 29-0000 Healthcare Practitioners and Technical Occupations PLUS any clinical function that is included within the 31-0000 Healthcare Support Occupations, such as Nursing Assistants.

I am continuing to track Health Care Support personnel (there are four job titles already identified in the MERS data including Health Clinic Admin Officer, Health Domain Worker, Hospital Employee, Hospital Receptionist) as I believe the differentiation from HCW is important but I am including their data in job family charts and infographics.

In the end, I made a bad call and I thank those of you who took the time to correct my thinking.

Flublogia is certainly a community and one I truly appreciate being involved in.

Random Analytics: Australian Mining Employment Update (to end June 2014)

Some years ago while working within the Workforce Planning fraternity I quickly understood that June was a tough month for employment, no matter what sector you either guided or worked in. It might have something to do with the seasons given that companies traditionally cut staff in the coldest months. Alternately it might be impacted on the business cycle. After more than a decade in Workforce Planning I’ve found that, whatever your view, business reflects more pragmatic ends.

Last month I suggested that June is often a horror month for mining employment.

My prediction was borne out (no matter the reason). What I didn’t guess was that the job gains versus the job losses would be the worst in more than two-years, potentially making June 2014 the worst month in mining employment since the commencement of the Global Recession (effectively Nov/Dec 2008).

Here are the charts for Australian Mining Employment through to the end of June 2014.

1 - MiningJobsByState_Infographic_Jun2014_140701

 

The opening infographic looks at job gains and losses by State or Territory for the month of June 2014.

Data-points: All key mining states (with the exception of the Northern Territory) have lost significant numbers of operational mining (human) capital. The only positive is that Tasmania picked up an additional 50-positions.

2 - MiningGainsLosses_Chart_140701

The first of my mining employment charts looks at the previous 24-months from a mining employment gain and loss perspective. The positive employment numbers are split into those that reflect infrastructure (tan) and operational (blue) gains. Job losses are represented in red.

The biggest data-point for June 2014 is that the losses in employment are the greatest in two-years, effectively making June 2014, officially, the worst month for two years but, in effect, the worst month for mining employment since the Global Recession which kicked off in Nov/Dec 2008.

3 - MiningSectorGainsLosses_Chart_140701

 

The next employment chart looks in more details at the main resource types (Iron Ore, Coal, Gold/Copper, Zinc/Lead/Nickel, CSG/LNG and Uranium) by either a job gain or a job loss.

Of the 3,939 job losses for the month of June the key points are:

  • 100% were operational job losses (effectively, zero in construction);
  • 79% were included in my tally of key resources;
  • 37.5% of the total operational losses in June were in the Coal sector, the 24th month of consecutive losses in that sector;
  • 0% of the losses were in mining construction.

It should also be noted that the figures above do not include projected job cuts from BHP Billiton who are currently reviewing their iron ore business in West Australia.

4 - MiningSectorSentiment_Chart_140701

The last chart tracks employment gains and losses sentiment.

Given that I haven’t commented on this chart for some time the key points are that August –November 2012 and Q4 FY 2012/2013 plus July 2013 were very negative times for mining employment (effectively corresponding with declining commodity prices). We are currently undergoing a similar period with soft mineral pricing from late 2013 through to even more depressed prices in the last six weeks of the financial year.

One last point on the uptick in June 2014. There were lots of positive announcements for mining in June from government (i.e. positive EIS announcements) but little in the way of immediate positive employment benefits.

Juking the Stats (June 2014)

Whitehaven Coal is cutting jobs at its coal handling and preparation plant at Gunnedah in response to “tough” coal market conditions. At the same time they are not releasing much in the way of details or numbers of the people they are putting out of work and potentially on the bench for an extended time. I look forward to a future update.

Even Money Prediction

After two years of consecutive coal mining employment losses due to depressed thermal and metallurgical coal prices the continued low iron ore pricing will see a similar (and consecutive) drip feed of employment decline in that resource group.