SSA Canberra is greatly saddened to advise that Professor Susan Wilson passed away on Monday 16 March 2020. Sue was a strong supporter of the SSA Canberra for her entire professional career, and attended meetings until very recently. In 1997 she presented the Knibbs lecture entitled “Who counts and why? A perspective of the history of women and men in statistics”. She was later a discussant for Oliver Mayo’s Knibbs lecture in 2002, when the topic was “To what extent has Fisher’s research programme been fulfilled In Australia?”

Sue also presented named lectures for the SSA on multiple occasions. She gave the first Moran lecture at the Australian Statistical Conference in Canberra in 2016. Her topic was “The ongoing influence of genetics/omics on advances in statistical science”. Outside of the SSA, Sue was an elected member of the International Statistical Institute (ISI), elected Fellow of the American Statistical Association (ASA) and elected Fellow of the Institute of Mathematical Statistics (IMS, 1995). She was the President of the International Biometric Society (IBS, 1998-1999). 

Sue provided excellent support to female statisticians in a generally male-dominated profession and also to early career statisticians, encouraging them to attend national conferences as well as local meetings such as Canberra Branch meetings. We will all miss her wisdom and humour.

She is survived by her son Jonathan.

Regards,

Alice Richardson, Conrad Burden, Alan Welsh, Francis Hui

On 5 March 2020 Science & Technology Australia (STA) announced 17 STEM Ambassadors who will work together with their local MP to help bridge the gap between science and government in Australia.

These STEM Ambassadors represent 17 different electorates across Australia and have been matched with MPs from across the country and the political spectrum who have expressed a desire to build stronger scientific networks.

STA President Associate Professor Jeremy Brownlie said that the STEM Ambassador Program encourages the involvement of science in Australian politics and aims to put science and evidence-based policy on the national agenda.

“The STEM Ambassador Program is vital to forging relationships between science and parliament,” he said.

“STA advocates for evidence-based, science-informed policy, and the STEM Ambassador Program helps to connect our national decision-makers with scientific experts who can give them direct access to research, data and evidence.

Each Ambassador will meet regularly with their local MP to build associations between parliament and the broader STEM sector. It enables federal politicians to gain a deeper understanding of the potential impact of science, technology, engineering and mathematics (STEM) in Australia.

“It’s brilliant to see so many of federal parliamentarians involved in the program and keen to know more about the science and technology happening in their own back yards,” Associate Professor Brownlie said.

“Australia’s STEM professionals play an incredibly important role in shaping Australia’s health and wellbeing, economic prosperity and environmental sustainability. This program empowers our highly skilled STEM workforce to make positive change and use their work to help build better policy and shape Australia’s future.”

The STEM Ambassador program builds on the successful 2019 pilot program and brings the total number of STA STEM Ambassadors to 24 across Australia.

The latest cohort of STEM Ambassadors come from a wide range of science, technology, and engineering mathematics professions, representing a range of sectors. 

Congratulations to SSA member Dr Alex Russell on his nominations as STA STEM Ambassador. Alex will be working with MP Zali Steggal in the electorate of Warringah.

• Media contact: Leesa Maroske 0439 784 216

In case you missed it, a report from the 2019 Research Quality Workshop of the National Health and Medical Research Council (NHMRC) was released in February 2020. This committee has arisen because of concerns with the quality of health and medical research, with estimates that 85% of health and medical research is currently wasted. The importance of statistics got several mentions in the NHMRC report, and was called a “critical competence” for researchers. The report also mentioned the need for institutional support, including “centrally funded statistical support for research groups”. There was also support for “statisticians as advisors to/members of ethics committees” and we lobbied for this recommendation by writing to the NHMRC with the letter led by members Prof. Sue Wilson (ANU) and Prof. Michael Martin (ANU).

This highlights the value of contributing to consultation processes which have an important statistical component or interest. If there’s an issue of national importance that you’d like our support on, email us to let us know.

Adrian Barnett

President, SSA

A/Prof Ayse Bilgin (President-Elect of IASE) and I were Guest Editors for a Special Issue of the Statistics Education Research Journal published 29 Feb 2020, entitled ‘Building Future Generations of Statisticians’.

Following on from collaborative SSA national initiatives such as STEMS2016 ‘Putting Statistics into STEM in the Age of Data’, as I noted in the SERJ Editorial pp 8-10, for many, the theme of this Special Issue has been a perennial focus, discussed at national and international fora but perhaps lacking a consolidated emphasis in the literature.

The aim for this Special Issue was to provide such a platform for sharing the many and varied international ‘outreach’ initiatives, experiences, resources and supporting mechanisms for increasing the number engaging with the field of statistics.

Before you view the articles perhaps you may like to consider why you were drawn to statistics and whether the work you are performing is as you’d expected it to be when you began your studies.

Professor Peter Howley

Section Chair – Statistical Education

@peterhowley0

Thanks to support from the ABS, a fresh and engaging SSA NSPC website has been created – please have a look and let me know your thoughts.

The NSPC is in its 6^th year as a national activity following its 2014 pilot and continues to engage new participants annually from primary and secondary schools.

Please inform friends, family and local schools of this enjoyable activity which engages teams of students with statistics via investigations and aligns with and supports national curriculum outcomes.

Professor Peter Howley

@peterhowley0

Section Chair – Statistical Education

The SA Branch welcomed Professor Omer Ozturk, currently on sabbatical from Ohio State University, to our February meeting to speak on his work on probability-proportional-to-size ranked-set sampling from stratified populations. This is the topic of his current research, following on from many years of work on sampling methodology. Omer’s talk took us through a number of examples to demonstrate the value of finite sampling, particularly in agriculture and environmental science.

Motivating examples included (1) the Quarterly Agricultural Survey performed by the Ohio Agricultural Statistics Department to estimate corn yield. Due to the presence of both regular farms and ‘mega farms’ auxiliary data on the farms is required to be used to ensure an appropriate sampling scheme; (2) the US Census Bureau’s Monthly Retail Trade Survey. Size data based on the previous year’s annual revenues is assumed to be approximately proportional to the current year’s revenue and can be used to optimise sampling; and (3) the Turkey Statistical Institute estimates the total apple production in Turkey. Auxiliary data identified two sub-populations, and this was able to be used to develop a sampling scheme that increased the available information.

Omer’s work combines theory from the finite population setting, proportional-to-size (PPS) and ranked-set (RS) methodologies, and extends this to more than one population (eg, geographic regions) for a stratified PPS-RS approach. Proportional-to-size sampling requires an auxiliary variable assumed to be proportional to the outcome, and ranked-set sampling requires determining, with reasonable accuracy, the relative position of the sampled units. Omer presented results for estimating the mean, variance, and confidence intervals. Finally, he discussed methods for sample size allocation across strata including equal allocation, proportional allocation, Neyman allocation, and allocation under a fixed cost plan. Results demonstrated that stratified PPS-RS was more efficient than stratified PPS and stratified simple random sampling for all allocation procedures.

In conclusion, Omar demonstrated that ranking information induces more structure and improves the information content of the sample. This can be extended to more complex population structures such as clustered populations, with further work to be published soon. It was an entertaining and engaging presentation.

Key reference: Ozturk O & Bayramoglu Kavlak K. (2018). Model based inference using ranked set samples. Survey Methodology, 44(1): 1-16.

Author: Kylie Lange

Family caring responsibilities (such as caring for children or elders) can be a barrier to conference participation. The SSA will make a number of bursaries available to people wishing to attend ANZSC2020  to help defray the additional costs of familial caring responsibilities.

SSA members who will incur additional costs associated with familial caring responsibilities due to their attendance at ANZSC2020 are invited to apply for a bursary. These bursaries are available to people of all genders. These funds could be spent on additional childcare or eldercare expenses that would arise as a result of attendance at the conference, or on flights for an accompanying person to look after children while the awardee is at the conference, for example. Bursary awardees do not need to present work at the conference to be eligible for a bursary. The SSA has a maximum of $2000 available for these bursaries, with the maximum amount available to any one applicant being $500.

The application form is available here. Applicants need to confirm they are a member of SSA, and intend to attend ANZSC2020. A short statement is required on how attendance at ANZSC2020 would be beneficial for their career, and details of the requested funds. Successful applicants would be reimbursed upon presentation of ANSZC2020 registration confirmation and receipts/invoices. The deadline for applications is March 25 2020, with decisions to be communicated to applicants by April 6 2020 (the Early Bird Registration deadline of ANZSC2020).

There will be a pre-conference workshop at ANZSC in July on logistic regression in Autostat. We have a trip report from the BayesComp workshop in January, which was attended by members of the Bayes SSA section. The ISBA Program Council have extended the early registration and poster submission deadlines for the ISBA World Meeting and BAYSM (June/July, 2020). A reminder about ABC in Grenoble (March, 2020) and announcement of an SMC workshop in Madrid (May, 2020).

There will be a pre-conference workshop “From machine learning to statistical modelling (Frequentist & Bayesian), via logistic regression in Autostat” held on Sunday, July 5, prior to the Australian Statistical Society and New Zealand Statistical Association Conference (ANZSC) on the Gold Coast. This year Griffith University has planned 70 workshops and other training events, delivered by 20 presenters, on statistics or mixed (quant/qual) methods. Several workshops introduce a Bayesian approach for particular models, with computing in Autostat and/or R. There are also introductory workshops on

“Understanding Probability”, and “Classical vs Bayesian Statistical Modelling.” All workshops are advertised a few months ahead here.  Please have patience with a new process for attendees external to Griffith University to

enrol in workshops here.

The biennial BayesComp workshop, sponsored by the Bayesian computation section of ISBA, took place January 7-10 in Gainesville, Florida. The conference promotes original research into computational methods across all strands of Bayesian research. This year’s plenary talks included cutting-edge topics such as Continuous-time MCMC, variational inference (VI), as well as Bayesian analysis of large-scale time series. The advancement of methodology for high-dimensional and complex models was a persistent theme of the conference. An underlying theme of research present in many talks was the move away from more traditional, exact, Bayesian methods, to alternative methods that are approximate or robust to model misspecification. 

The 5th Bayesian Young Statisticians Meeting (BAYSM) will take place in Kunming, China from June 26-27, 2020. Deadline for submission of abstracts is March 29. Due to concerns about the current coronavirus epidemic, the ISBA Program Council have extended the early-bird registration deadline for the ISBA World Meeting (Kunming, China, June 29 – July 3) until May 15. They will keep submission of contributed posters open until May 15 as well. More details about the conference programme are available here.

A workshop on Approximate Bayesian Computation, ABC in Grenoble, will be held in France on March 19 and 20. More details are available here.

A workshop on Sequential Monte Carlo methods, SMC 2020, will be held in Madrid, Spain, May 27 – 29. More details are available here.

Executive Committee, Bayes Section of SSA (Chris Drovandi, Clara Grazian, David Frazier, Sama Low-Choy & Matt Moores)

Twitter: @BayesSSA

https://statsoc.org.au/Bayesian-Statistics

Editors: Helen MacGillivray, Robert Gould, Jim Ridgway

The special issue of Teaching Statistics will be published in 2020 as an e-book, titled Teaching Data Science and Statistics: senior school or introductory tertiary.

A variety of chapters is envisaged, with a small number of invited chapters, and most oriented to classroom-ready ideas, case studies and/or materials embodying good teaching practice. Each of the latter type of chapter should focus on one or more aspects of statistical data investigations and data science. Authors will find valuable guidance in Unit 1 of the framework of the International Data Science for School Project (IDSSP). Case study approaches should embody classroom-ready implementation, with a rich data context with a number of variables.

Chapters should be between 2,000 and 4,000 words, and no more than 5,000 words, including references, figures and any appendices. Resources such as data, code, notes for teachers/instructors, and videos, may be placed in an online repository. Authors may wish to provide short, dynamic videos.

Timeline

• Submissions to be made online to Teaching Statistics via ScholarOne.
• Deadline for submissions, 25^th April, 2020.
• Referee reports to authors by 30^th June, 2020.
• Deadline for revised submissions, 31^st August, 2020.
• Proofs to authors by 15^th October, 2020.
• Deadline for corrections to proofs, 22^nd October, 2020.
• Issue compilation, 29^th October, 2020.
• Issue approved, 7^th November, 2020.
• To printer and published online, 14^th November, 2020.
• Published in print, 21^st November, 2020.

Teaching Statistics, https://onlinelibrary.wiley.com/journal/14679639, is published by Wiley on behalf of the Teaching Statistics Trust. It is intended for all those who teach statistics to students aged up to 19 years. The emphasis is on good practice in teaching statistics and statistical thinking in any context. Teaching Statistics seeks to inform, enlighten, stimulate, guide, correct, inspire, entertain and encourage. Teaching Statistics is a refereed journal, with double-blind reviewing.

For more information contact the Teaching Statistics editor h.macgillivray@qut.edu.au . Also see Teaching Statistics, 42 (1).

Helen MacGillivray

The Statistical Society of Australian and CSIRO/Data61 are delighted to announce the inaugural winners of the Betty Allan travel award as:

•  Karen Lamb who will visit the Winton Centre for Risk and Evidence Communication in Cambridge
• Sharmala Thuraisingam who will visit the University of Calgary and attend the North American Primary Care Research Group Conference

The award of $3000 each is to support early career female statisticians in Australia to travel anywhere in the world in order to further their career.

There were 14 applications for what should have been just one award, but the judging panel recommended two awards because they were so impressed with the quality of the applications.

The president of the Statistical Society of Australia, Adrian Barnett, said, “The quality of the applications was inspiring. It proves that we have incredible early career female statisticians in Australia. But it did make the final decision very difficult.”

Frances Elizabeth (“Betty”) Allan (1905 to 1952) was CSIRO’s first statistician and provided statistical support across the organisation. She also lectured in mathematics and statistics at Canberra University College and Australian Forestry School. When she married in 1940 she was forced to retire, which was the law for female public servants.

The award will be open again in 2020 with applications called for in November. For details on last year’s application click here.