I’m Caitlin Bakker, the new Discovery Technologies Librarian at the University of Regina. Essentially, that means that I implement and manage systems that connect people with the information that they need, whether that need is teaching, learning, research, or anything else.
My favourite part of my job is being able to solve problems, and I love being able to make discovering, accessing and sharing information easier for everyone.
Prior to coming to the U of R, I spent 8 years at the University of Minnesota, where I worked on data science and management, systematic reviews, research impact assessment, research reproducibility, and open science and scholarship.
The Archer Library is looking for innovative, proactive, flexible, collaborative and service-oriented candidates for the tenure-track position of Science, Engineering and Social Work Liaison Librarian.
I’m interrupting the series on surveys this month for something different.
Image from Monty Python’s Flying Circus (1969-1974)
Well not completely different. I’ll still be talking about data, but instead of diving into Statistic Canada surveys, were going to take a little break and talk about books. I’m often asked about what resources to recommend to people who want to learn more about data. Here are some of my recommendations! The titles below are available by searching QuickFind (except The Data Detective) or through Regina Public Library.
How to Lie With Statistics – Darrell Huff
The classic, the oldie, and the goodie. The book that started the exploration of critical thinking and skepticism in viewing data and statistics. Published in 1954, the practices Huff describes are still being used today. For the people in the back: THE PRACTICES HUFF DESCRIBES ARE STILL BEING USED TODAY!! How to Lie With Statistics is a great work for learning about data visualizations and how they can be deceptive. However, don’t let this be the only book you read on data as it’s overwhelming message is “everyone lies with stats” and that’s simply not the case. While critical thinking about data is important, the book can be cynical and dismissive of all data and statistics. So take the lessons about how visualizations are manipulated, but leave the cynicism behind.
The Data Detective – Tim Hartford
Hartford addresses the cynicism of Huff in the introduction of The Data Detective. This book provides a less cynical and dismissive look at data and instead recommends that when looking at data and statistics, the best thing we can do is be curious. Hartford breaks down his message into 10 rules when approaching data and provides examples of why instant dismissal and skepticism towards data can be just as dangerous as accepting all data as truth. One of my favourite recent reads about data, Hartford encourages a balanced approach to viewing and interpreting statistics that isn’t overly daunting or intimidating.
Calling Bullshit – Carl Bergstrom and Jevin West
Another book about being skeptical about data! This work is a hilarious and frank look at data in the time of social media and misinformation. In particular, it looks at how Big Data and how the all mighty algorithm complicates data comprehension. Overall, this work focuses on using critical thinking and skepticism when looking at all forms of information, not just data, but the lessons discussed in the book also apply to data and statistics.
Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are – Seth Stephens-Davidowitz
This book is a fascinating dive into what big data tells us about human nature. Using big data collected from anonymous search histories, the book provides a never before seen look at human behaviours, prejudices, and patterns. The book explores how big data identifies and tells truths that most humans lie about, while also exploring how big data can be wrong and misused.
So that’s some potential beach reads for everyone as we approach summer! Enjoy!
Archer Book Club’s next meeting is June 22, 2022: 12-1pm, discussing The Song Of Achilles by Madeline Miller.
This novel is essentially Homer’s Iliad retold from Patroclus’ perspective, and focusing largely on his romantic relationship with Achilles and the events that lead to the Trojan War. More information about the novel can be found at our libguide here.
Archer Library is offering Zoom at Noon sessions all spring and summer long! Check out our Discover YoUR Library and Fake News: Evaluating Information dates here.
The new Spring/Summer 2022 semester brings the Archer Book Club to celebrating two years of reading and meeting! We are thrilled to continue offering these selections and discussion opportunities to the University of Regina community.
Cool Librarians don’t look at explosions Image: Unflinching walk from tvtropes.org. Image from I am Number Four (2011).
Last month we looked at quick polls with and I promised a second part on public opinions polls and Statistics Canada surveys. So here’s the sequel, where we’ll explore sampling in surveys. Like most sequels, it’s going to be bigger and more intense.
Okay, so there will be no explosions, but this next level of surveys is more intense than a quick poll. There’s even more potential for bias and accurate sampling and representation are needed to make good conclusions. So let’s dive in.
With these types of surveys, more attention is paid to the targeted population and size of the sample. Survey designers will use a formula to determine their sample size by considering the number of the targeted population, the confidence level (usually 95%), and the margin of error (usually 5%). Once the ideal number for the sample is decided, the survey designers will determine how to choose their sample.
Ideally, samples should be selected using probability sampling, but often time and funding require survey designers to use non-probability sampling. Probability sampling uses random selection to choose participants for the survey. For example, in the past, landline phone numbers were frequently used to randomly select participants for national surveys. The randomness of probability sampling reduces selection bias, response bias, and undercoverage bias. However, probability sampling costs more, is more time-consuming, and can be challenging as randomly selected participants may or may not want to participate. As a result, non-probability sampling is becoming more common.
Even if non-probability sampling is used, it’s still vital that the sample be representative of the targeted population and that selection bias be reduced. If the sample is not representative, then the results and conclusions have reduced validity. For example, if a targeted population is Saskatchewan residents, you couldn’t select your sample just from Regina, the city just doesn’t represent all of Saskatchewan.
Public opinion polls are a great example of how representative samples can be selected. If you look into the fine print of public opinion poll, you’ll find a breakdown of the sample by age, gender (usually binary), geographic location, and sometimes political ideology. For example, Angus Reid Institute recently conducted a public opinion poll on how Canadians feel about the monarchy. The main report provides a breakdown of the sample for reporting purposes and at the bottom, you have the option to view a full breakdown of the sample by multiple factors. This show how Angus Reid has strived to gather a representative sample.
Image: Meme made at Imgflip.com Image from Harry Potter and the Philosopher’s Stone (2001)
We now have to consider how selection bias could play a role. Angus Reid Institute clearly states how they select participants for their surveys on their website. There are two things to note. First, participants are selected from the Angus Reid Forum. While Angus Reid Institute states they are non-partisan, media analysis tends to identify them as having a slight conservative leaning on the political spectrum, meaning their forum members could identify as conservative which could create a bias. Second, participants may be entered for draws or paid to complete surveys, which as we discussed last month, could create a bias. It is impossible to remove all bias from sampling and surveys and the practices of Angus Reid Institute are common for most public opinion polling, so this level of analysis does not discredit their polls; rather, it demonstrates how all public opinion polls are subject to these issues.
So here we are, we survived the sequel! But wait, where was the section on Statistics Canada surveys? Well, it looks like this series is a trilogy.
Spring in Saskatchewan is nearly here! With Ramadan beginning today and with Easter and Passover just around the corner, we have updated their respective pages with new recipes and resources in the “Holidays & Celebrations” tab of the Library Leisure Guide.
Ever watched an episode of Family Feud? The show’s premise relies on contestants guessing answers to survey style questions. Surveys are an incredible way to quickly gather opinion data. As a result, we probably encounter surveys on a routine basis. For example, just this week I encountered the following:
a) While shopping three different stores prompted me to complete a user experience survey by using a website and code on the receipt. If I participated I would be entered in a draw for a gift card.
b) While watching YouTube I was asked to complete a short survey on ads I may have recently viewed.
c) While traveling through an airport there were numerous stations asking about the airport experience. This included satisfaction with going through security, cleanliness of the washrooms, and satisfaction with food services.
The HappyOrNot Terminal system. Seen throughout Toronto Pearson Airport.
User experience surveys are only one type of survey you may experience. Other surveys include text based polls through news agencies, internet or phone based public opinion surveys, and Statistic Canada surveys. User experience polls or daily question polls are “quick polls” that measure the opinion of the participant. User experience data gathers your opinion for a specific transaction or single experience. New media quick polls often relate to a story from the day. Quick polls are great for gathering opinions in the moment as they usually take less than a minute to complete and often only feature a single simple question, but are not designed complexly enough to measure long term opinions on the subject. For example, you could rate a shopping experience as positive on Tuesday and negative on Wednesday due to multiple factors. Research surveys, public opinions surveys, and government surveys are much more structured than a quick poll, so we’ll tackle them in more detail next month!
Not everyone will participate in a survey, so sample sizes need to be considered. The sample is the number of participants needed to represent the survey’s targeted population. Quick polls often use volunteer sampling meaning participants have to choose to complete the survey by going to a website, scanning a QR code, or texting a response. These methods are anonymous and non-confrontational which reduces courtesy bias, when participants change their responses in order to be polite.
Volunteer sampling is efficient, but results must always be critically examined. For instance, volunteer sampling is prone to selection bias, in particular voluntary response bias, because the “silent majority” do not respond. People who have neutral experiences and opinions are just less likely to remember the encounter or are not motivated enough to complete the survey. People who have strong opinions are typically highly motivated to complete surveys as a way to express their feelings.
To increase motivation in neutral participants, quick polls may offer small incentives such as prizes or monetary compensation. However, the incentive has to be carefully chosen to increase motivation, but not encourage false or incomplete responses from participants just to get the incentive. However, as participants we should consider and think critically about what personal data we are giving up in order to get a quick poll incentive. In some cases demographic and contact data must be provided to get access to the survey, which could result in spam material. In other instances, completing the surveys may require making a store account or downloading an app which will track our data. Ultimately the choice is ours!
