Group items tagged

'Significance' is a magazine published by the UK Royal Statistical Society and American Statistical Association. E.g. this article discusses the data we might use to analyse food security, thinking about what sources of data are available and the questions they might help us answer. "data on food insecurity is biased towards the environment in which it was created and the priorities of those who collect or commission it. Data from schools is concerned with reimbursement; government data might be focused on budgetary constraints or accountability; grocery stores could (if willing) tell us what food is bought, but not how it is used; meanwhile, non-profits are most interested in demonstrating impact to funders. There is a wide variety of data sets available, but very few are created with the intent to understand the real drivers of hunger and poverty. The data may be repurposed, but modellers must be especially careful to moderate the assumptions of each data set. ...... It is also especially important that those experiencing hunger and poverty are consulted when designing any data analysis project. The input of domain experts is crucial to the success of data science endeavours, and those experiencing poverty know the right questions to ask."

The conceptual, practical, and ethical issues surrounding "big data" and data in general begin at the very moment of data collection. Particularly when the data concern people, not enough attention is paid to the realities entangled within that significant moment and spreading out from it.1. Data sets are the results of their means of collection. It's easy to forget that the people collecting a data set, and how they choose to do it, directly determines the data set. An illustrative example can be found in the statistics for how many hate crimes were committed in the United States in 2012. According to the FBI Uniform Crime Reporting Program (UCR), the number was 5,796. However, the Department of Justice's Bureau of Statistics reported 293,800 hate crimes.

Citizen-generated data is especially important for women's rights issues. In many countries the lack of women in positions of institutional power, combined with slow, bureaucratic systems and a lack of prioritisation of women's rights issues means data isn't gathered on relevant topics, let alone appropriately responded to by the state. Even when data is gathered by institutions, societal pressures may mean it remains inadequate. In the case of gender-based violence, for instance, women often suffer in silence, worrying nobody will believe them or that they will be blamed. Providing a way for women to contribute data anonymously or, if they so choose, with their own details, can be key to documenting violence and understanding the scale of a problem, and thus deciding upon appropriate responses.

A great data-story on the gender imbalance worldwide, using data and research to investigate and highlight key issues. This piece in the economist takes a different approach to using the data https://www.economist.com/node/15636231 "In 1960, the earliest year the World Bank provides data for, the world was within 0.002 percentage points of a perfectly equal distribution. Ever since, the gap has widened; now men outnumber women on the planet by more than 66 million. When this piece was first published in early 2014, the gap had already been the widest ever - the trend continues."

On the importance of having data, in order to understand and tackle an issue... The most recent U.S. estimate...concluded approximately 513,000 women and girls at risk of genital mutilation...But that number should be taken with a big grain of salt...the data doesn't account for immigrants from countries where female genital cutting isn't studied or widely practiced...."You also can't assume that people who come to the U.S. are a representative sample of their country of origin," Clark said. That's especially problematic for estimating rates of female genital cutting, since it's not practiced uniformly within countries. It's also possible, he said, that some immigrants abandon the procedure as they assimilate....some advocates point out that although the estimates focus on immigrants, ...female genital cutting isn't new to the U.S. Female circumcision was performed as a treatment for masturbation by American physicians as recently as the mid-20th century...

Really good video illustrating how (a) gathering data, and (b) understanding the shape of that data can give us insight onto real world issues, and help us to target approaches to tackling them "Britain has carried out one of the biggest data-gathering exercises on the gender pay gap, exposing large disparities between the average pay given to men and women in some of the country's best-known companies. We dispel some of the myths around the gap, and explain what it really means and why it matters"

Ourworldindata is a great website discussing lots of different datasets about global issues. This example data-story discusses the issue of global health, giving an overview (and lots of great visualisations), and discussing how we actually measure 'health' (life expectancy, quality of life measures, etc.).

A complex look at an important issue and what data can tell us about it, and how missing data complicates the picture.

journalists often write about public opinion polls, which are designed to measure the public's attitudes about an issue or idea. Some of the most high-profile polls center on elections and politics. Newsrooms tend to follow these polls closely to see which candidates are ahead, who's most likely to win and what issues voters feel most strongly about. Other polls also offer insights into how people think. For example, a government agency might commission a poll to get a sense of whether local voters would support a sales tax increase to help fund school construction. Researchers frequently conduct national polls to better understand how Americans feel about public policy topics such as gun control, immigration reform and decriminalizing drug use. When covering polls, it's important for journalists to try to gauge the quality of a poll and make sure claims made about the results actually match the data collected. Sometimes, pollsters overgeneralize or exaggerate their findings. Sometimes, flaws in the way they choose participants or collect data make it tough to tell what the results really mean. Below are 11 questions we suggest journalists ask before reporting on poll results. While most of this information probably won't make it into a story or broadcast, the answers will help journalists decide how to frame a poll's findings - or whether to cover them at all.

The data seems to overwhelmingly point to a criminal justice system riven by racial bias. But, remarkably, it could be even more overwhelming than some studies make it seem. That's because of a statistical quirk called "collider bias," a kind of selection bias that means that the crime data that shows racial bias is, itself, biased by racist practices. If you thought crime data showed clear evidence of racism before, understanding how collider bias affects these analyses might make it even clearer.

When we launched Beyond the blade earlier this year, we wanted to know how many young people and children were being killed by knives in the UK. Who are these young people being killed?, Where are they dying? Is the scale of the issue changing, and if so how? We spoke to experts about the number of children and teenagers affected in Britain and Northern Ireland. We checked with the Office for National Statistics, the Home Office, politicians, academics and thinktanks. But the answer to how many young people are dying every year, it seemed, was that nobody knows. So we started trying to find out. Until now, there has been no publicly available information about the demographic profiles of those who have died from knife attacks in the UK

An introduction to a set of deep dive articles an important issue....When we tell people that we work at the intersection of statistics and human rights, the reaction is often surprise. Everyone knows that lawyers and journalists think about human rights problems … but statisticians? Yet, documenting and proving human rights abuses frequently involves the need for quantification. In the case of war crimes and genocide, guilt or innocence can hinge on questions of whether violence was systematic and widespread or one group was targeted at a differential rate compared to others. Similar issues can arise in assessing violations of civil, social, and economic rights. Sometimes the questions can be answered through simple tabulations, but often, more-complex methods of data collection and analysis are required.

Advancement toward gender equality at work has slowed since the 1990s for three major reasons: people's attitudes stopped becoming more gender egalitarian, occupations stopped gender integrating, and the gender wage gap began decreasing at slower rates. Sociologist Paula England has called this phenomenon an "uneven and stalled" gender revolution, and there have been dozens of studies showing how the progress in gender equality experienced during and immediately after the feminist movement of the 1970s has not been sustained through the 1990s and 2000s. Does this stalled revolution play out in management positions, too? And if so, how? To explore this, I used data on full-time managers obtained from the U.S. Census and American Community Survey for the years 1980 and 2010 to examine three major factors that contribute to gender equality in the labor force: women's representation in management, the occupational gender segregation among managers, and the gender wage gaps that vary across managerial occupations.

Data aren't dead, contrary to what some pundits stated post-election [2], rather the limitations of data are not always well reported. While pollsters will be reworking their models following the election, what can media journalists do to improve their overall coverage of statistical issues in the future? First, discuss possible statistical biases, such as errors in sampling and polling, and what impact these might have on the results. Second, always provide measures of uncertainty, and root these uncertainties in real-world examples.

A great video example discussing a contentious issue (coal consumption) using the data!

Nice discussion of the data around contraception choices. Shortly after Donald Trump was elected president I started noticing an interesting trend on my social media newsfeeds. And no, I'm not talking about the near-constant bickering of people with differing political opinions. I started seeing post after post from friends publicly asking one another about their experiences with different forms of birth control. The motivation for these kinds of conversations centered around the pending rollback of copay-free contraception, but have since been re-kindled every time reproductive rights come up in the political arena. And it's not just talk. Many of these conversations centered around the use of long-term contraceptives like intra-uterine devices or IUDs which can protect against pregnancy for 3 - 12 years. In the months immediately following the 2016 election, AthenaHealth reported a 19% increase in IUD-related doctor's visits and Planned Parenthood reported a 900% increase in patients seeking IUDs. Cait, 27, recently switched to a copper IUD, and said that she made the switch due to convenience and "because now in light of our current administration I'd like to have something that will continue to work and be affordable even if I end up without health insurance."

An excellent visual data story describing the gender pay gap (in America) and debunking the claim that there is no real difference in the amount men and women get paid. The pay gap varies depending on the occupation, working hours, education attainment, experience, and geography. That explains part of the difference in pay between men and women, but not all of it. And even though most economists agree that after adjusting for age, education, experience and other variables there's still an unexplained gap, there are voices who argue that the gender pay gap is a myth. Pay gap deniers purport that women's choices, rather than discrimination, cause the pay gap between women and men. But those choices are actually consequences of the social forces at play.

Roughly put, building a machine-learning model involves training it on a large number of examples and then testing it on a bunch of similar examples that it has not yet seen. When the model passes the test, you're done. What the Google researchers point out is that this bar is too low. The training process can produce many different models that all pass the test but-and this is the crucial part-these models will differ in small, arbitrary ways, depending on things like the random values given to the nodes in a neural network before training starts, the way training data is selected or represented, the number of training runs, and so on. These small, often random, differences are typically overlooked if they don't affect how a model does on the test. But it turns out they can lead to huge variation in performance in the real world. In other words, the process used to build most machine-learning models today cannot tell which models will work in the real world and which ones won't.

Alcohol and Other Drug (AOD) Media Watch is based on the same premise as the ABC show Media Watch. It aims to highlight poor examples of journalism regarding AOD-related issues in the hope that we can assist journalists to report more objectively using science and evidence rather than perpetuating myths, opinions and moral panic. Research has found moral panics in the media can actually be detrimental. Moral panics in the media can actually be detrimental by counter-intuitively leading to increased drug use since it increases the perception that more people are using the drug than actually are. It has also been found to found that moral panics reduce the degree to which some people believe that the drug being reported on is harmful. It also reduces the credibility of AOD information in the media.

Interesting links to different sources discussing data on teacher pay, and what it shows in relation to different things that stakeholders might care about. educator pay varies significantly across states, from an average of $44,921 in Oklahoma to $77,957 in New York. Why should school administrators and government leaders care about teacher pay - beyond wanting their employees to be able to afford their living expenses? Below, we present research that examines this issue. What scholars have found is that teacher salaries are linked to employee retention and that better pay seems to draw smarter people to the field and into the classroom. It's not clear, however, whether higher salaries result in higher student achievement.