Dee Keilholz

��˾��ֱ�� statistics students analyze oceans of data to understand how water temperature affects sharks

Christopher.Sorensen — Mon, 30 Aug 2021 15:28:25 +0000

��˾��ֱ�� statistics students analyze oceans of data to understand how water temperature affects sharks

Christopher.Sorensen Mon, 08/30/2021 - 11:28

Cutline

Undergraduate student Vinky Wang and recent graduate Jessica Long analyzed reams of data to understand the impact of water temperature changes on leopard sharks (Photo by Cavan Images/Getty Images)

Dee Keilholz

Topic

Breaking Research

��˾��ֱ��

Climate Change

Faculty of Arts & Science

Statistical Sciences

Sustainability

Trinity College

Undergraduate Students

If you’ve ever watched Shark Week, you might think the life of a shark researcher is full of adventure involving near-death encounters with sharp teeth and snapping jaws. The reality looks much different, often requiring patience rather than courage in the face of danger, as Vinky Wang and Jessica Long know.

Rather than swimming with sharks, Wang, an undergraduate student in the University of Toronto's Faculty of Arts & Science, and Long, a recent alumna, spent hours hunched over their computers cleaning, analyzing and visualizing data from observations of sharks in their natural habitat. The data set was supplied by the Shark Lab, a research unit studying the ecology of marine animals at the California State University, Long Beach.

Under the supervision of Vianey Leos Barajas, an assistant professor in the department of statistical sciences and the School of the Environment in the Faculty of Arts & Science, Wang teamed up with Shark Lab researcher Jack May to determine how changes in water temperature impact the life and behaviour of leopard sharks along Southern California’s coast.

Shark Lab researcher Jack May used drones and underwater temperature loggers to collect the information the team of ��˾��ֱ�� researchers needed for their data analysis.

“The data is basically just a bunch of longitude and latitude coordinates and the number of sharks observed,” explains Wang, a statistics, geography and math student. “After you fit your statistical model and print out your output, you basically just have a lot of numbers. We then created a visualization that mapped all the coordinates, so you can see a map of the study region. You can see how the sharks move within each grid cell and where they prefer to aggregate and cluster.”

Preliminary findings show that leopard sharks are rather picky when it comes to water temperature. After analyzing the data collected by May over two summers, it seems leopard sharks seek out areas in a specific range of temperature. They might prefer to hang out in one spot but don’t hesitate to move on to a different area if temperatures change and end up falling below or above the sharks’ preferred range.

With scientists ringing the alarm bells about climate change – most recently through an Intergovernmental Panel on Climate Change report – the importance of leveraging data to understand the impact of environmental changes on marine animals is a vital tool in protecting the oceans and its inhabitants, May says.

“The better we can understand how temperature affects sharks distribution and behaviour, the better we can manage them and make sure they have a healthy population for some time to come,” he says. “Statistics and coding play a huge role in that. With better technology comes better data and, as researchers, we need to know how to make sense of that.”

Long, who recently graduated with a bachelor's degree in statistics, math and economics as a member of Trinity College, spent her summer working alongside Shark Lab researcher Emily Meese on a similar project, analyzing the movements and behaviours of California horn sharks.

The most challenging part of her work was learning how to use “hidden Markov models” (HMMs), a complex statistical approach frequently used to model biological sequences and reveal patterns in animal behaviour.

This visualization, created by Wang, takes data from drone footage and other sources to match areas where sharks like to cluster to certain seafloor temperatures.

“I had no prior experience with HMMs, so I had to read through textbooks and papers to get an understanding of what to do. It was a huge learning curve. Using statistics to study and classify animal behaviour in general is actually quite complex,” says Long. “I learned a lot through this project, and I really enjoyed teaming up with people in the field and learning about their work.”

The research projects are part of Leos Barajas's Bayesian Ecological & Environmental Statistics (B.E.E.S.) research group. The group’s goal is to develop statistical methods to help answer pressing ecological and environmental questions.

Leos Barajas emphasizes that working in interdisciplinary teams is a key component of all B.E.E.S projects and an important factor in student learning, as well as statistical research in general.

“What I’ve seen this summer is that pairing students with researchers from other disciplines has worked very well,” says Leos Barajas. “Statisticians alone can't make sense of the data. We need to work with experts in other fields to really understand what it is we’re doing.”

The team is currently working on expanding their model to include data from other locations in order to see if there are other factors in addition to temperature that might affect the sharks’ behaviour.

“I find it rewarding to be able to apply the skills I learned in class to real-life data and to see results that have actual meaning and implications,” says Wang. “The thought that this project might lead to more research and make a real difference is really motivating.”

��˾��ֱ�� research shows dramatic rise in U.S. opioid deaths as epidemic spreads to eastern states

noreen.rasbach — Mon, 11 Mar 2019 15:54:42 +0000

��˾��ֱ�� research shows dramatic rise in U.S. opioid deaths as epidemic spreads to eastern states

noreen.rasbach Mon, 03/11/2019 - 11:54

Cutline

A lethal dose of the synthetic opioid fentanyl: ��˾��ֱ�� research shows that the rise of illegally produced fentanyl is the main reason behind the sharp increase in opioid-related deaths (photo via Wikimedia Commons)

Dee Keilholz

Topic

Global Lens

Faculty of Arts & Science

Global

Research & Innovation

Sociology

Statistical Sciences

United States

The opioid epidemic ravaging the United States has claimed a staggering number of lives, but a clear picture of how the crisis has changed over the years – and who it currently affects the most – is surprisingly hard to find.

One popular narrative paints the opioid epidemic as a rural, low-income phenomenon, concentrated among Appalachian states – a claim that might be largely rooted in stereotypes, says Monica Alexander, an assistant professor in the department of statistical sciences and the department of sociology.

She decided to put the theory to the test.

Just a couple of months earlier, Alexander had published a paper that unravelled another myth surrounding opioid deaths in the United States.

“I was inspired by a New York Times article on how the opioid epidemic was mostly affecting white people. That turned out to be false. The data showed that the current number of opioid-related deaths is increasing at a higher rate for Black people,” says Alexander (pictured left).

“We were curious to see if there are other standing narratives that aren’t quite accurate, and to find out how things have actually changed.”

In search of a clear picture of how opioid-related deaths are distributed across the United States, Alexander teamed up with three other researchers at Stanford University and Harvard University. The team wrangled close to 20 years' worth of data on causes of death in the U.S., revealing the changing face of one of America’s biggest public health crises in the process.

Their study appeared in JAMA Network Open last month.

Here are some of the most important findings:

The crisis is growing – and fast

The opioid crisis has grown exponentially, the research found, with the number of opioid-related deaths quadrupling over the last two decades.

Within 18 years, more than 351,500 men and women died from opioid-related causes in America. In states with high opioid-related death rates, such as New Hampshire and West Virginia, the opioid crisis has reduced overall life expectancy at age 15 by a full year.

“What that means is that if – hypothetically – you would get rid of all opioid deaths in a state, people’s life expectancy would increase by one year,” says Alexander. “Just to put that into perspective: Changes of 0.1 years over a year are considered normal.” That makes life expectancy loss from opioids greater than that from either firearm or car accidents in most of the U.S.

The crisis has spread

Long believed to be concentrated among Appalachian states and parts of the Midwest, Alexander’s research shows that the crisis has spread. Eastern states have seen a sharp rise in opioid deaths – with Connecticut, Illinois, Indiana, Massachusetts, Maryland, Maine, New Hampshire and Ohio among the states with the highest rise in opioid-related deaths.

Synthetic opioids are the main culprit

Synthetic opioids now by far outnumber heroin deaths in most eastern states. In 28 states, opioid-related deaths from synthetic opioids doubled every two years. In half of those states, deaths doubled every year.

While prescription pain killers, such as oxycodone, led the first wave of the opioid epidemic, illegally manufactured synthetic opioids, mainly fentanyl, have taken over as the leading cause of death.

Fentanyl is a particularly potent synthetic opioid. Up to 100 times stronger than morphine and many times more potent than heroin, a barely visible amount of fentanyl powder can be lethal. Among drug dealers, lacing drugs with fentanyl to save money has become a popular practice, and it is putting drug users at risk for accidental overdoses.

There are many reasons why some U.S. areas are harder hit than others, says Alexander. Differences in drug supply between western and eastern states might be one of them.

“West of the Mississippi, you see heroin sold mostly in the form of black tar heroin, which is hard to mix with other drugs. East of the Mississippi, you mostly see white powder heroin, which is very easy to lace with synthetic opioids,” she says. “It’s impossible for drug users to just look at it and determine if this is just heroin or heroin laced with something much deadlier.”

To help policy-makers take active steps, Alexander and her team highlighted several public health interventions in an open access paper released in late February, such as expanding access to naloxone, putting needle exchange programs in place and increasing support for people suffering from mental health issues and addiction.

“Prevention is obviously the goal but looking at the data I don’t think we can just rely on prevention. In my opinion, there needs to be a widespread focus on treatment and easy access to treatment,” says Alexander.

“We are at a point where this is an emergency. Something needs to be done if we want to prevent even more people from dying.”

'There’s inequality in death': ��˾��ֱ�� statistician analyzes mortality data to help level the playing field

noreen.rasbach — Fri, 16 Nov 2018 20:03:28 +0000

'There’s inequality in death': ��˾��ֱ�� statistician analyzes mortality data to help level the playing field

noreen.rasbach Fri, 11/16/2018 - 15:03

Cutline

"I know this sounds pretty cliché, but I wanted to do something that could help people, especially marginalized people," says Monica Alexander about her statistics work

Dee Keilholz

Topic

Global Lens

Faculty of Arts & Science

Global

Research & Innovation

Statistical Sciences

Life is not a level playing field, but if you ask Monica Alexander, assistant professor in the University of Toronto's departments of statistical sciences and sociology, she will quickly point out that death isn’t either.

Marginalized populations are disproportionately vulnerable to conditions or events that can cause an early or unnatural death. By analyzing data on who dies, when and how, researchers like Alexander can learn a lot about how to improve or even save people’s lives.

For example, insights gained from data on child mortality can lead to more effective social policies. Likewise, recognizing patterns in the spread of a lethal epidemic might help contain future outbreaks.

Unfortunately, it’s not quite that easy. Reliable information on issues that affect vulnerable populations is often sparse – and without the right tools, prying trustworthy knowledge from datasets riddled with holes can quickly turn into guesswork. Filling in those gaps by using the right tools is at the core of Alexander’s work. Her research focuses on developing statistical methods that produce reliable results, particularly when there is little data available.

Her passion for serving disadvantaged people has led her to contribute to research projects that have informed the work of organizations such as UNICEF, the Bill and Melinda Gates Foundation, the World Health Organization, and the Human Mortality Database.

Dee Keilholz of the department of statistical sciences in the Faculty of Arts & Science spoke to Alexander about her work at the intersection of statistics and social sciences and her passion for making a difference.

What motivates you to do the kind of work you do?

I know this sounds pretty cliché, but I wanted to do something that could help people, especially marginalized people. There’s a lot of inequality in death because vulnerable populations tend to be just so much more affected.

I knew I was never going to become a doctor. I’m terrible with blood, but I realized that I can use my skills in analyzing health and mortality outcomes to impact social policy.

Your work focuses on developing statistical methods to overcome data gaps. Why is it that data on disadvantaged populations seems especially sparse?

Populations that have the highest death rates unfortunately also have the worst data. Child mortality, for example, is substantially higher in places like sub-Saharan Africa than it is in Canada, but because we don't have a good registration system for deaths in sub-Saharan Africa, we don’t have a lot of information to work with. That makes it difficult to get a sense of how mortality is changing or what efforts to improve health are successful. That’s where my work comes in. I might have some data from surveys or small-scale surveillances and then I try to figure out how we can get a good picture of what is going on, despite the data not being perfect.

How do you do that?

By applying different statistical methods on sets of data and using the trends that I see to inform my estimates – in combination with a demographic approach.

A classical demographic approach is to use strong empirical regularities. We humans are actually quite predictable about when we’re going to die. When you look at mortality by age groups, there’s a very distinct shape that looks like a “J”. The rate of deaths is a bit higher in the first years of life, but then it dips down. With old age, it increases again. You can use prior knowledge like that to adjust and model biases in the data.

You recently co-published a paper on the opioid crisis. What did you find?

That paper was inspired by a New York Times article on how the opioid epidemic was mostly affecting white people. I’m always skeptical when it comes to narratives about health or death that focus on white populations when things tend to be usually worse for minority populations.

We looked at data from 1979 onwards to figure out how this epidemic has evolved over time for both white and Black populations in the U.S. We found that opioid deaths are increasing at an alarming rate for both. However, while white populations have a higher mortality rate today compared to 15 years ago, the current number of opioid-related deaths is increasing at a higher rate for Black people. That suggests that the narrative of the opioid crisis as a white, middle-class epidemic, driven by bad prescription practices, isn't entirely accurate.

What’s a project you enjoy working on right now?

I'm currently working on a project using Facebook data to get a better sense of the migration wave that happened in Puerto Rico around the time of Hurricane Maria. How many people left? Where did they go? Did they go back? What are the demographic characteristics such as age, sex or education? We can learn so much from not just looking at the overall number of people who left but also at who are these people who were affected.

As luck would have it: ��˾��ֱ�� statistician's new book explores the meaning of luck in everyday life

noreen.rasbach — Fri, 05 Oct 2018 13:52:36 +0000

As luck would have it: ��˾��ֱ�� statistician's new book explores the meaning of luck in everyday life

noreen.rasbach Fri, 10/05/2018 - 09:52

Cutline

"We want things to happen for a reason," says ��˾��ֱ�� Professor Jeffrey Rosenthal. "It’s a universal human need. I completely sympathize with that notion, but I also think that isn't how the world always works"

Dee Keilholz

Topic

City & Culture

Faculty of Arts & Science

Global

Research & Innovation

Statistical Sciences

When Joe Parker offered to take a photo of a vacationing family at a beach in Waikiki, he had no idea at first that he’d just met his half-brother. Rick Hill, the man he’d just photographed, had a noticeable Massachusetts accent, and Parker commented on it. They soon discovered that they grew up in neighbouring towns. Parker started throwing out some names, including his dad’s, Dickie Halligan. To Parker’s great surprise, Hill replied that Halligan was his father too.

Was this impromptu family reunion just a lucky coincidence or did fate intervene? Or are surprise encounters like this one a lot more common than we think?

The story of Joe Parker is one of many delightful anecdotes that the University of Toronto's Jeffrey Rosenthal, a professor at the department of statistical sciences, shares in Knock on Wood: Luck, Chance, and the Meaning of Everything. His new book explores the concept of luck in its various forms through the lens of a statistician.

After the release of his first book on randomness and probability, Rosenthal says friends, colleagues and readers kept asking him whether he believes in luck or not. Unable to give an answer, he realized that there must be more to the topic than meets the eye.

He embarked on a two-year research and writing journey that led him to collect countless stories, read piles of research papers and watch hours of Netflix – who said research can’t be fun? (He also coincidently discovered that he was born on a Friday, the 13^th.)

The result is a light-hearted, yet insightful analysis of the meaning of luck, its different faces, and how to tell the difference between “dumb luck” and the things we can control.

Dee Keilholz of the department of statistical sciences spoke to Rosenthal about some of his findings.

Do you believe in luck?

I do. And when I say I believe in luck, I mean random things that are outside of our control. Luck plays a huge role in everybody’s life, myself included. However, I realized after a while that when people ask me if I believe in luck, they often mean different things. Some people believe in fateful events and destiny and then there’s lucky charms and astrology. There are these various forms of luck, depending on whom you talk to. I realized that it’s hard to come up with a simple answer that explains what luck is. That’s partly what my book is about – trying to figure that out.

If luck is this random, highly influential force in our lives, how do you suggest we deal with it? Should we let go of the illusion that we are in control of our life or should we “shape our own luck” and try to take things into our own hands?

I think there are things in life that are in our control and others that are not. For example, somebody might t-bone your car, and that’s just bad luck. Wrong time, wrong place. But there are also things you can do to drive safely: wear a seatbelt, drive within the speed limit, and so on. I made up a “luck version” of a popular serenity prayer: “Grant me the serenity to accept the random luck I cannot control, the knowledge to change the luck which can be modified, and the wisdom to know the difference.” Knowing the difference is key.

How can statistical thinking help us understand that difference between random luck and the things we can change?

In my book, I use the term “luck traps,” meaning times when people think that an event is more meaningful or surprising than it is, and there are certain ways of thinking that can help us understand why that is.

One of the many principles I talk about in Knock on Wood is what I call the “Out of How Many” principle. Or, in other words, how many different chances were there for a certain event to happen? Let's take the half-brothers who met at that beach in Hawaii. There are more than 300 million people in the U.S., and a percentage of those have estranged relatives. Each of those people make countless decisions over the course of their lifetime about where to go and what to do. So, you could say that there were millions of “tries” and out of all these tries, one time a happy reunion happened. I mean, it’s rare but not that incredible if you look at it like that.

Why do you think we are so prone to seek greater meaning behind random events?

M. Night Shyamalan, the director of the movie The Sixth Sense, once said that every great movie has some element of magic in it. That’s certainly true for a lot of the films and literature we love. We want the good guys to win against all odds and the boy and the girl are destined to get together in the end. We want things to happen for a reason. It’s a universal human need. I completely sympathize with that notion, but I also think that isn't how the world always works. Bad things can randomly happen to good people and love doesn’t necessarily conquer all.

Don’t you find that sad? Doesn’t this make the world a bit less magical?

I don’t think so. Our world in all its randomness is quite wonderful as it is. There are more than seven billion people on this planet and most of them are perfectly nice and fascinating to get to know. I believe we need to appreciate the world the way it works. It leads to better logical thinking and decision-making, while leaving plenty of room for incredible adventures.

��˾��ֱ�� biostatistician uses big data to shed new light on chronic diseases

Christopher.Sorensen — Thu, 27 Sep 2018 18:25:22 +0000

��˾��ֱ�� biostatistician uses big data to shed new light on chronic diseases

Christopher.Sorensen Thu, 09/27/2018 - 14:25

Cutline

Nathalie Moon recently joined ��˾��ֱ��'s department of statistical sciences as an assistant professor in the teaching stream (photo by Diana Tyszko)

Dee Keilholz

Topic

Our Community

Big Data

Faculty of Arts & Science

Medical Research

Statistical Sciences

Big data has the potential to create new life-saving knowledge in medicine and health, but only if we know how to make sense of large data sets.

And that’s what biostatistician Nathalie Moon has set out to do.

Moon’s latest research empowers medical professionals to gain critical insights by helping them decide how much data to collect – and which kind. A recent University of Toronto hire with a focus in teaching, she's also making an impact by passing her knowledge along to the next generation of statisticians.

For the last five years, Moon, who joined ��˾��ֱ��’s department of statistical sciences as an assistant professor this summer, has been working with researchers at the Centre for Prognosis Studies in the Rheumatic Diseases (CPSRD) at Toronto Western Hospital. The centre closely monitors patients suffering from rheumatic disorders – conditions causing chronic pain of the joints and connective tissue – to gain insights into the progression of disease and to improve treatment. Moon leveraged her statistical skills to design research studies that produce accurate results while making efficient use of resources.

Dee Keilholz of the department of statistical sciences spoke to Moon about her latest research project and her passion for teaching students.

What does a biostatistician do?

A biostatistician develops statistical methods and applies them to address problems in areas such as medicine, biology and health. A big part of my work revolves around designing research studies and, to some degree, analyzing data related to biological processes.

Tell me a bit more about your research with the centre. What was the focus of your research and what problems were you trying to solve?

My research looked at three distinct problems, all under the umbrella of understanding the progression of chronic diseases such as psoriasis and psoriatic arthritis – conditions that cause inflammation of the skin and joints respectively.

One part focused on patients who develop a pretty painful, disabling complication called arthritis mutilans. CPSRD follows patients over time and tracks if they develop this complication. However, some patients stop showing up for follow-up appointments, which leads to loss of information. We just don’t know if those patients developed this complication. So, CPSRD launched so-called “tracing studies”, to track down these “lost” patients to see what happened to them. But what if you have hundreds of patients who dropped out, and you only have resources to track down 50? That’s where my research comes in. I basically looked at the data we had on these individuals before they dropped out to figure out who we should focus our resources on.

Another question was around composition of study cohorts – the group of people who participate in these studies. If you want to track the progression of a disease over a patient’s lifetime, that might mean you have to follow that patient for 20 or 30 years. That’s very expensive, and you have to wait a long time to get results. That’s why I looked into designing studies where you don’t have to track people for decades but still get good estimates by recruiting some patients in the early stages of a disease and others from later stages. I showed that this can lead to large savings without compromising the quality of the estimates.

What motivates you to do this type of research?

I prefer working in an area where I can see the link between what I am doing and its usefulness in real life. I was hooked on biostatistics when I took my first course in survival analysis at Queen’s University, which led me to pursue graduate studies in biostatistics at the University of Waterloo. To give an example, survival analysis looks at questions such as cancer patients and the impact of tumor size on patient survival. I just had that “aha” moment where I realized how relevant these questions are, and I wanted to learn how to answer them.

You’ve just started working at ��˾��ֱ�� as an assistant professor with a focus in teaching. What interested you in taking on this role?

I’m an assistant professor in the teaching stream, which means I will spend most of my time teaching students. I first taught a course during my PhD and it made me realize how much I enjoy helping students learn and encouraging them to try different things. There’s such a variety of career opportunities in statistics, and I like being part of that journey of figuring out what’s right for them. It’s making a different kind of impact than the research I did during my PhD.

I really look forward to teaching my first- and fourth-year courses this term. I think my students are going to benefit from me being a researcher but also a teacher, because it means that I can bring that research experience to the classroom.

'Statistics anxiety': How this ��˾��ֱ�� lecturer is using art to help students overcome their fear

noreen.rasbach — Fri, 17 Aug 2018 17:13:10 +0000

'Statistics anxiety': How this ��˾��ֱ�� lecturer is using art to help students overcome their fear

noreen.rasbach Fri, 08/17/2018 - 13:13

Cutline

Asal Aslemand, a ��˾��ֱ�� lecturer, has adopted a unique format for her courses to fight statistics' "bad rep" and lower students' anxiety

Dee Keilholz

Topic

City & Culture

Faculty of Arts & Science

Statistical Sciences

��˾��ֱ�� Scarborough

Undergraduate Students

Psychology and neuroscience student Farnaz Vahadat waited until the last year of her undergraduate program to enrol in a statistics class she needed to finish her degree.

“I knew I had to enrol at some point, but I put it off until the very last minute. I was just so terrified of statistics and that I would fail,” she says.

Vahadat is not alone with her “statistics anxiety”. No one knows that better than ��˾��ֱ�� lecturer Asal Aslemand. Since she started teaching statistics courses to students pursuing a degree in social or life sciences, she has come across many who would gladly choose a visit to the dentist over signing up for a statistics class.

“Statistics has a bad rep,” she says. “It's just something a lot of people perceive negatively. It makes me sad, because I know how much students in almost every discipline can benefit from learning statistical methods."

To help students overcome their fears, Aslemand, who earned a master's degree in statistics and a PhD in statistical education, decided to combine her two loves, statistics and education, while also not forgetting about her background in visual arts.

In 2016, when she taught her first statistics course for non-statisticians at the ��˾��ֱ�� Scarborough's department of computer and mathematical sciences, she let students pick data sets for their final project that lined up with their interests. For extra credit, students could also let their creativity run wild and turn the results of their data analysis into a piece of art.

“I was absolutely blown away by what the students came up with: string art, mini models, posters. Some students even composed songs [see video below] and poems about their final research,” says Aslemand.

“I’m not saying, we need to use art to make science better," she says. "But I think that providing students from various backgrounds with an interdisciplinary approach and a way to visualize their research helps them connect with and communicate their work – and maybe even fall in love with statistics.”

That certainly came true for Tasnia Elahi, a second-year city studies student, minoring in geographic information studies and statistics. It was Aslemand's class, she says, that convinced her to choose statistics as a minor.

“Before, I hadn’t even considered statistics, but Asal’s class made me realize how much statistics and social sciences are connected,” she says. “Being able to visualize my research at the end of the term was really helpful, too. It helped me connect to my findings on a different level, and it also made it easier to explain to others why these numbers matter.”

To illustrate her analysis of a data set on homeless shelters in Ontario, Elahi created a 3D model out of paper and Christmas decorations – including twinkle lights to illuminate the buildings in her “shelter city”. Her data analysis revealed a shortage of shelter space in Toronto compared to other cities across the province.

Life sciences student Al-Amin Ahamed created a model to illustrate a link between obesity and cognitive function, based on a study in mice.

Since her first class, Aslemand has worked on fine-tuning her teaching methods, while teaching statistics classes at the department of statistical sciences and the ��˾��ֱ�� Scarborough's department of computer and mathematical sciences. She is planning on collecting some of her students’ art pieces in a book. She is also eager to conduct a study of her “teaching experiments” to analyze students’ experiences and see if they embrace statistical thinking long-term – in their work, but also in life. After all, she says, statistical thinking is needed more than ever in times of fake news and black-and-white thinking:

“Statistical sciences are important because they teach a mindset that doesn’t assume but critically questions what is happening. We need to be careful to not just conclude things but to pay attention to relationships and the things we don’t know. It’s necessary for producing good academic work but also for understanding the world around us.”

How to look at polling data and vote strategically: ��˾��ֱ�� statistician offers insight

ullahnor — Mon, 04 Jun 2018 17:30:12 +0000

How to look at polling data and vote strategically: ��˾��ֱ�� statistician offers insight

ullahnor Mon, 06/04/2018 - 13:30

Cutline

Ontario Premier Kathleen Wynne takes questions after the most recent debate. She had hoped for a bump in polling, but on Saturday she acknowledged the election has been lost (photo by Steve Russell/Toronto Star via Getty Images)

Dee Keilholz

Topic

City & Culture

Faculty of Arts & Science

Ontario politics

statistics

On June 7, Ontario voters will cast their vote and choose a new government, and based on Kathleen Wynne's comments over the weekend, the Liberals may lose official party status.

In April, polling data indicated that roughly half of Ontarians have made up their mind. That leaves the other half still undecided and with an important decision to make, especially given the projected neck and neck race between Ontario’s Progressive Conservative Party (PC) and the New Democratic Party (NDP).

Public support for Wynne has nosedived since the campaign started, and recent polls have suggested even longtime strongholds could be lost for the Liberals. In Ontario, parties need at least eight seats in the legislature to be formally recognized. On Saturday, Wynne acknowledged that she will no longer be premier after the election, encouraging voters to elect Liberal candidates to prevent the NDP or PCs from securing a majority.

Jeff Rosenthal (below), professor of statistical sciences in the Faculty of Arts & Science, talks to ��˾��ֱ��'s Dee Keilholz about whether we can trust the latest polls and whether there's ways to maximize the impact of our votes.

Can we trust the polls – in this election and generally speaking?

Generally speaking, I’d say ‘yes.’ Polls are not as far off as many people think. If you can take a truly random sample of people, even if you just sample 1,000 people out of 9 million voters, you still get quite accurate results. If you poll 1,000 people, the margin of error is down to about three per cent.

However, the challenge with polls and their accuracy is that it is really difficult to get a truly random sample. What if, for example, the people who respond to pollsters are also more likely to vote for a certain party? What if certain age groups are more likely to respond? Do people tell pollsters the truth? That’s what leads to biases and that's when polls can run into trouble.

Polls are a great illustration of the power of randomness. In fact, my main research area, Monte Carlo algorithms, uses this same kind of randomness to estimate quantities which are too complicated to compute directly.

What have the polls been predicting for the Ontario elections?

In this election, the polls project a close race between the NDP and PCs. The Conservatives will probably do better in their seat count than their vote count, so it's quite possible that the NDP wins more votes but the Conservatives get more seats. It is true, though, that seat projections are complicated, and there have been cases in the past where the forecasters got it wrong.

In the 2011 federal election, for example, forecasters almost unanimously predicted that it is extremely unlikely that the Conservative party would win a majority – but they did. I wrote a paper about this, suggesting a more accurate seat projection method.

So what should voters do to get a clear picture?

It’s smart to look at a bunch of polls, instead of just one. The easiest way to do that is to visit polling sites that take multiple sources into account. I recommend Calculated Politics or the CBC Poll Tracker.

What kind of polling data should voters pay special attention to if they want to vote strategically?

In Canada, we have a first-past-the-post system. Whoever wins in a riding takes that seat. That means strategic voting can really have an effect.

If you want to vote strategically, you should pay attention to the projection for your riding. Let’s say you want party x to lose and party y to win, but you’re also OK with voting for party z. You should find out if party y or z is most likely to win in your riding and vote for whoever is ahead.

Unfortunately, projections for ridings can be a bit tenuous, because they are based on the latest provincial polling and results of past elections in that riding, rather than actual polling data from that riding. However, I still think this is a reasonable approach – and your best bet if you choose to vote strategically.

What effect is voter turnout going to have on this election?

If the decision to vote is independent of political leanings, voter turnout has no effect because you still get the same relative proportions. However, if groups with certain political beliefs are more likely to vote – or not to vote – than others, that can indeed make a difference.

We saw that in the 2015 federal election. Voter turnout for 18- to 24-year-olds went from 39 per cent in the 2011 election to 57 per cent in 2015 because Justin Trudeau really connected with the younger demographic – and he won.

In this election, it appears that older people are more likely to support the Conservatives and are also more likely to vote than younger people. Younger people are more likely to support the NDP but less likely to vote. So, voter turnout is an important issue in this election.

Can the upcoming provincial election give us any clues as to how the 2019 federal election might turn out?

The Liberals in Ontario are expected to do quite badly in the upcoming Ontario election. It's not clear exactly why this is. Voter fatigue is one possible cause.

The question is will that affect the Liberal Party of Canada? Some of the federal polls suggest that the popularity of the federal party among Ontario voters is somewhat down, and that there is going to be a ripple effect. My guess is that the Ontario election won’t have a lasting impact. By the time the next federal election rolls around, this provincial election will already be ancient history.

��˾��ֱ��'s Nancy Reid elected fellow of the Royal Society

noreen.rasbach — Mon, 28 May 2018 04:00:00 +0000

��˾��ֱ��'s Nancy Reid elected fellow of the Royal Society

noreen.rasbach Mon, 05/28/2018 - 00:00

Cutline

“One of my probability professors once stated in class that, ‘Women are not suited for mathematics.’ We’ve come a long way since then, but we still have a relatively small number of women in leadership positions," says Nancy Reid (photo by Diana Tyszko)

Dee Keilholz

Topic

Global Lens

Faculty of Arts & Science

Global

Statistical Sciences

Few people can say that their signature appears in the same book as Sir Isaac Newton’s. The University of Toronto's Nancy Reid will soon be able to stake that claim – once she signs the Royal Society’s charter book.

The Royal Society recently announced that Reid, a University Professor in the department of statistical sciences, is one of 50 U.K. and commonwealth scientists selected to become the society’s newest fellows. She shares this year’s honour with some of the world’s most eminent researchers and innovators, including Tesla founder Elon Musk.

Over the span of her career, Reid has excelled as an important contributor to the field of theoretical statistics, especially in the area of statistical inference – making sense of large and complex sets of data.

“Whether you are studying health or air pollution or you’re discovering new planets, almost every scientist uses statistical analysis in one way or another,” says Reid.

“A theoretical statistician like me basically studies an abstract version of a problem. I try to find the common threads between different research areas in order to make statistical methods more accurate and efficient so that scientists across disciplines can extract the best possible information from data sets.”

The versatile nature of statistical theory and its application across any field of science drew Reid to statistics in the first place. In 1979, she earned her PhD in statistics from Stanford University and joined ��˾��ֱ��’s department of statistical sciences seven years later.

“Nancy is a powerhouse and a huge asset to our department, the university, the Canadian and international community,” says James Stafford, chair of the department of statistical sciences in the Faculty of Arts & Science. “One common characteristic of everything she does is her immense capacity for work. Another is that when Nancy speaks, others fall silent and listen.”

Through her groundbreaking work, Reid has gained wide recognition. She is an Officer of the Order of Canada, the Canada Research Chair in Statistical Theory and the first woman to receive the President’s Award of the Committee of Presidents of Statistical Societies, one of the most prestigious international awards in statistical science.

Over the years, recognizing and supporting women in STEM [science, technology, engineering and mathematics] fields has increasingly become a priority for her, she says. To draw attention to gender barriers and inequality in her field, Reid wrote an essay in 2014, titled The Whole Women Thing, examining cognitive biases that impact the advancement of women in statistical science.

The Royal Society’s ceremony to welcome new fellows will take place in July in London.

Reid joins Eugenia Kumacheva, a University Professor in the department of chemistry, and Janet Rossant, a professor of molecular genetics, as the third female scholar from ��˾��ֱ�� to become a fellow of the Royal Society.

Emergency doctor goes from rescue ship to newsroom with ��˾��ֱ��'s Fellowship in Global Journalism

noreen.rasbach — Fri, 20 Apr 2018 04:00:00 +0000

Emergency doctor goes from rescue ship to newsroom with ��˾��ֱ��'s Fellowship in Global Journalism

noreen.rasbach Fri, 04/20/2018 - 00:00

Cutline

“When you read about people drowning in the Mediterranean, you normally read about 5,000 people. What’s missing are those individual voices to really make people care," says Sarah Giles (photo by Dhoui Chang)

Dee Keilholz

Topic

Global Lens

Global

Munk School of Global Affairs & Public Policy

If you ask Sarah Giles what a normal work week looks like for her, she will tell you that for most of her career, there has been no such thing. Before she joined the Munk School’s Fellowship in Global Journalism at the University of Toronto last September, Giles had been working as a family and emergency doctor with various organizations, including Doctors Without Borders/Médecins Sans Frontières (MSF).

For one of her assignments with MSF, Giles spent four months on board MV Aquarius, a rescue ship jointly operated by MSF and SOS Mediterranée – and when you’re stationed aboard a ship in the Mediterranean Sea, pulling refugees out of ramshackle boats to safety, a regular schedule is wishful thinking.

“Some days you’re bobbing around in a flat sea, perfectly blue skies and nothing is happening, and other days there’s waves and storms and you’re doing rescues. There was no ability to predict what was going to happen, which was hard because you had no sense of control.”

Throughout her 15 weeks at sea, Giles and her team rescued more than 4,000 people – some of them in life-threatening conditions.

“I remember rescuing a severely malnourished pregnant woman,” Giles says. “She was in terrible shape when she went into labour and we had to get her off the boat into a hospital. Otherwise, she and the baby wouldn’t have had a chance to survive.”

What struck Giles during her time at sea were the many personal stories people told her and her team – and the increasing realization that these stories needed to be shared with the world.

“When you read about people drowning in the Mediterranean, you normally read about 5,000 people. What’s missing are those individual voices to really make people care – like when the story of Alan Kurdi broke, everyone took a collective gasp,” she says.

“If people understand that these are fellow humans with hopes and dreams just like the rest of us, it’ll help make us all be a bit more kind, I think.”

Giles realized that as a medical professional, she had a vast amount of expertise and stories to share. More importantly, her profession gave her access to places and people reporters often struggle to reach. Giles decided to take advantage of her unique position. She enrolled in the Munk School’s Fellowship in Global Journalism, a program that recruits subject-matter experts from around the world and teaches them how to work as multimedia journalists.

“Media outlets no longer have enough beat reporters to cover complex stories,” says Robert Steiner, director of the Fellowship in Global Journalism and a former foreign correspondent for The Wall Street Journal.

“That’s why we mentor people who have a deep understanding of the subjects they are covering. They are able to approach their own disciplines with a truly open mind and find fantastic stories that conventional reporters miss.”

Although she had to take eight months off work, Giles hasn’t regretted her decision to enrol, she says. The program has not only taught her the basics of journalism but also provided her with a network of dedicated, savvy mentors.

“I’ve been sincerely impressed with the calibre and the generosity of the people who teach us,” Giles says. “We are supervised by journalists with a ton of experience. Despite the fact that we are basically nipping at their heels, they have been open, friendly and encouraging.”

One of the most important skills she’s learned is the art of media pitching. Over the last months, her writing has been published in various publications such as The Walrus, drawing attention to the staggering lack of health services in the Northwest Territories.

Access to health care in Canada’s First Nations communities is an issue Giles is deeply passionate about. In addition to her medical work in remote conflict zones, her portfolio includes working with First Nations communities. Her work up north has brought her face-to-face with the limits of her own country, she says.

“If you look at the social determinants of health, every determinant is worse if you are First Nations or Inuit. Our government needs to invest in health infrastructure – basic things like clean water and power. I recently saw a lady who broke her hip on the way to the outhouse because her house didn’t have plumbing,” Giles says.

“I don’t need to go overseas to work in a humanitarian crisis area. Unfortunately, I can do that in my own country.”

After finishing her fellowship in global journalism later this April, Giles will head to the Northwest Territories to continue providing health care to people in need – and to write about the issues that impact her patients' quality of care and life.

“We all have to do our part to fight for a more equitable world,” Giles says. “I love to work for people in Canada and around the world, who don’t have good access to medical care. It’s so fundamental to their well-being. If I can help them live a better life, that’s just levelling the playing field.”

Learn more about the Fellowship in Global Journalism

Master of global affairs students look at blockchain as a tool for altruism

noreen.rasbach — Tue, 06 Mar 2018 19:57:28 +0000

Master of global affairs students look at blockchain as a tool for altruism noreen.rasbach Tue, 03/06/2018 - 14:57

Cutline

Vanessa Ko, a master of global affairs student, and Zachary Skeith, an alumnus of the program, believe there is enormous potential of blockchain technology for social good

Topic

Munk School of Global Affairs & Public Policy

Research & Innovation

Startups

Technology

Zachary Skeith, an alumnus of the University of Toronto's Munk School of Global Affairs, had only a basic knowledge of blockchain technology when he started working at Three Lefts, a research and development startup. Regardless, he strongly believed that technology should be leveraged for more than just profit.

In basic terms, blockchains are digital logs of information that are stored across a network of personal computers. Blockchains use cryptography to make sure that individual records can’t be altered or counterfeited. The decentralized nature of blockchains ensures that no one person or institution can control the system, yet everyone can use it.

The more Skeith learned about blockchain – a digital ledger in which transactions of value are recorded – the more he realized the enormous potential of this technology for social good. What he previously only understood as the system behind Bitcoin, the world’s most famous cryptocurrency, he came to look at as a potential tool to alleviate hunger, settle land claims or ensure that humanitarian aid reaches people in need.

“I spent most of my childhood moving around different continents. I was very lucky to see a lot of the world, but it also made me very aware of the many global challenges we’re facing,” says Skeith, a graduate of the master of global affairs program. “It’s really exciting to me to see the immense potential of blockchain to change entrenched power structures.”

While blockchain technology is mostly being used in the private sector for now, blockchain-based initiatives that focus on humanitarian aid have started to take root in the nonprofit sector.

Vanessa Ko, a current master of global affairs student, started researching the potential of blockchains for humanitarian organizations last year during an internship with the United Nations. In October, she published a white paper outlining how blockchains can help streamline aid projects.

“One potential use could be to track the supply chain of aid items. With the help of the blockchain, you could trace exactly where aid comes from and where it goes. That could be a huge help in co-ordinating between different agencies and keeping everyone accountable.”

Because anything of value can be stored in the blockchain, the applications are endless, says Ko.

UN Women, for example, held a 36-hour-long blockchain hackathon, which, among other innovations, resulted in an app that helps female business owners around the globe make secure financial transactions. The World Food Program is another example of blockchain’s potential in global development: In 2017, it used blockchain technology to distribute more than $1 million in aid to Syrian refugees in Jordan.

Three Lefts has also taken steps to leverage blockchain technology to empower people. The company is exploring a blockchain-based foundation that would give marginalized communities access to first-stage business funding. Instead of a board of directors controlling how funding is allocated, this project will leverage the decentralized nature of blockchain to allow every member to vote for business ideas and collectively decide how funds are distributed, says Skeith. Another one of Three Left’s projects is aimed at helping Indigenous groups secure and control their intellectual property. A third project focuses on combating forced labour in supply chains.

Like any disruptive technology, though, blockchains present challenges in terms of ethics and governance – problems that Skeith works to address before they arise. Part of his job at Three Lefts is to spark conversations with clients, colleagues and industry regulators about what an “ethical future” of blockchain might look like. What sets him up for success in his role are the skills he was able to hone as an master of global affairs student, he says – namely lateral thinking and the ability to facilitate between people from various backgrounds.

“Because of my global affairs background, my focus is the political and social impact of technological innovation,” says Skeith. “Whereas most people in the tech space think they need to scale up first and governance will come later, I believe the opposite is true. We need to think about social capital first.”

Reports about the consequences of companies scaling up without building governance and ethical guidelines into their products have become a common occurrence. Facebook, for example, has recently come under fire in the media for spreading fake news, allowing companies to exclude older workers through age-targeting, and letting advertisers racially discriminate in housing ads.

“I think we’re at a point where we need to define how we want this technology to work. If we treat the blockchain like we’ve treated most technologies from the early ’90s, then the blockchain is going to become just another status quo tool,” says Skeith. “And that would be a shame because it has the potential to fundamentally improve our society.”