Icebreakers! (not the gum)

To start off this post, it’s probably fitting to quote a Duran Duran song (1990): “The lasting first impression is what you’re looking for.”

Besides starting with the usual housekeeping on the first day of class, why not set the tone for the course by providing students with a glimpse into the classroom environment as a community of learners, get students to connect with one another, AND do statistics? Look no further than an Icebreaker activity! We present two Icebreakers that can get your class (either in-person or online) off to a great start: Questions on the Back (a classic) and How Old? Visualization

Questions on the Back Activity (Laura Le)

The purpose of the Questions on the Back activity is to allow students to experience statistics in an informal (and fun!) way. And, it can be implemented within in-person (note: does not adhere to the physical distancing guidelines) and online (asynchronous) introductory statistics courses. 

Activity for in-person courses

The start of the activity is to tape a question1 to the back of each student. Students will not know the questions that are being taped on their back, but tell them that the goal of this activity is to collect data (numbers only!) from their fellow classmates to help them figure out what the question is on their back. Now, some of the questions may be easier to figure out, such as “What is your shoe size?”, and other questions may be harder to identify, such as “What is your lucky number?”. 

1Instructor prep prior to class: (1) Create a list of questions where the answer is a number. Here is my running list as a starting point. (2) Print off the list and cut out the questions into little strips of paper. (3) Bring the strips and Scotch tape to class for the activity. (4) Create a slide (or a poster) of the activity’s instructions that can be displayed. Here is my slide for the instructions.

Note: For the remainder of this article, I’m going to refer to the students with the unknown questions on their back as Question Carriers and the students who read the questions and provide an answer as Responders.

At this point, students are asked to walk around the classroom with a writing utensil and something to write on and interact with their classmates. When a student finds a peer, I ask that they introduce themselves and possibly state their program or major. Then, the pair takes turns reading the question on the back of their peer in their head (silently) and providing an answer to the question with only a number and nothing else (e.g., no units). If the question asks “What is your…?”, it is a question about the Responder (and not about the Question Carrier). After collecting responses from all the students in the class (if the class size is less than 20) or after 15 responses (if the class size is greater than 20), students are to find a place at the board (if there is enough room) or use a drawing tool (paper, iPad) and graph the data to help them figure out what question was on their back. 

Once everyone has graphed their data, I ask for volunteers to summarize their results (using the graph as a visual) and to try to guess their question. For those that volunteer, they are the first to be able to see their question on the back. The number of volunteers I have depends on how many minutes are left before class is over. In total, this activity takes approximately 20-30 minutes of class (or longer, depending on how many students you ask to describe their graph and guess their question).

This activity is one of my favorites to kick off the first day of my in-person introductory statistics courses. Why, you may ask?

  1. Speed Meeting: The activity allows students to meet and interact with their peers on a one-to-one basis in a less intimidating, more personal setting than in a large group setting (e.g., going around the room), and they can meet most of their peers (if not all of their peers) in a relatively short period of time. It’s a great way to set the class’s tone as a community of learners.
  2. Element of Surprise: The students are interested and motivated in figuring out what the question is on their back. 
  3. DOING Statistics on Day 1: They are DOING statistics on the first day of class. They are collecting, and possibly organizing, real data, which is a GAISE recommendation and goal for introductory statistics courses. Some are using external cues (variables) beyond the number provided to help figure out their question, such as how long it takes Responders to answer the question or the Responder’s body language. hey are also summarizing data with a graph of their choice, since they are not told what kind of graph to create. 
  4. Informal Assessment: The kinds of plots that are created help me, as the instructor, understand where my students are at in their prior knowledge (specifically, on graphical representations). I have seen all kinds of plots, some more useful than others, from boxplots and histograms to line charts (with the index number on the x-axis and value on the y-axis) and pie charts. 

Activity for online courses

The online version has a similar goal of exploring real data while getting to know other students in the class, but the roles are flipped. Rather than one student trying to guess one question, one student gets a question, collects data on that question from information in their peers’ introduction posts2, summarizes the data with a plot, and creates a one- to two-sentence description of the data (but not the question) to the rest of the class in a Q&A forum. Then, their classmates respond to the post by guessing which question they had from the description of the data. Students are provided a Word document for the Icebreaker activity that includes the instructions for how to complete it. 

2In the Introduce Yourself discussion forum, students are asked to answer five questions (all have numerical answers) about themselves and told that information will be used in a learning activity for that week. Then these questions are placed into the activity and students are “randomly” assigned to one question. 

Since there are only a few questions (approximately five) that are asked of students, the questions will have multiple students supplying a description of the data. However, it is still very insightful to see how each student decides to tell the story of their data. 

___________________________________

Note: I first learned about the Questions on the Back activity from Michelle Everson when I was a graduate student in the Statistics Education department at the University of Minnesota. To be honest, when I initially learned about these “first day” activities as a kick-off to the course, I was not 100% on board. I thought they sounded interesting, they might be fun, but they were a little cheesy (and not just because I’m from Wisconsin). This was before I tried it out in class, thus not realizing its potential for students and for instructors. So give it a go!)

How Old? Visualization Activity (Steve Foti)

This activity was inspired by a conversation I had with Dennis Pearl at USCOTS 2019 about fun things to try in the classroom. He was showing me the Microsoft-powered website, how-old.net, that will try to guess your age from a picture that you take or upload and describing how he has used it before in the classroom. With a brand new data visualization course coming up in the spring semester, I was especially open to ideas that could be adapted to my course. After playing around with the website and reflecting on our conversation, I developed an activity and piloted it in two different courses, Biostatistical Literacy in the fall and Data Visualization in the Health Sciences in the spring.

The Biostatistical Literacy course is a graduate level service course that typically contains between 20-30 students, most of whom are advanced-degree-seeking medical professionals. The data visualization course is a new MS elective offered by our department that is open to both majors and non-majors, and most recently contained a small handful of students studying biostatistics as well as other health professions. These are the courses I have tried this activity in so far, but I believe it could be used in classrooms of any skill level. 

The basic idea of the activity is to collect data by having each student take multiple selfies (the website utilizes your default camera app) and record the age it guesses each time. With their individual data, students are asked to create a visualization to show some important feature of their data using any means they are comfortable with (e.g. pens, colored pencils, Excel, R). Then, in pairs or small groups, they are asked to think about and discuss ways they might be able to successfully manipulate the algorithm (e.g. putting on/taking off glasses, smiling, changing the angle). Using their idea, students collect more data and add it to their visualization in a way that distinguishes it from the original data. Finally, the data displays are shared with the class and we have brief discussions about them. I typically lead these discussions with questions like, what does this graph show? Does anything about the data stand out to you? Does it look like the attempt to manipulate the algorithm was successful? 

The full activity instructions are shared on our resources page

I enjoy this activity because it is a little bit different than your standard icebreaker. The discussion between students has an element of fun and mystery, and is likely something that they have not worked with before. At the same time, the work they are completing allows them to showcase their creativity and their comfort with graphing and communicating about data. Below are a couple of examples that students in my classes have created through this activity. Not all of them exactly follow the instructions, but are still generally on topic and are interesting to see. 

So far, I have only tried this activity twice in a face-to-face setting. This fall, I will be testing it out in an asynchronous, online version of the biostatistical literacy course. Since we can no longer have a live discussion, I have changed the activity submission to a discussion board format. Students will upload their final graphic and post their conclusions about their data, their graphic, or the algorithm used by the age guessing software. I may also require them to respond to at least one post in an attempt to encourage interaction and full participation in the activity. 

I think this activity lends itself a little better to the face-to-face setting. Here, students are able to interact throughout the process and share their thoughts on how they might manipulate the algorithm. It is more fun, in my opinion, when students can share and laugh about the results of the age guesser, compare ideas for manipulating the algorithm, and be present for the concluding discussion. In the online setting, the students are no longer able to have the same level of interaction, so while the activity still offers some benefit as a statistical activity, it loses some of its credibility as an icebreaker. 

Online Strategies due to COVID-19, Part 2

In this series of posts, the StatTLC blog team describes how we are managing with the abrupt changes to our courses. In this, we share some of our decisions (and the thinking that went into them), the tools we are using, and tips. We are teaching a diverse set of classes this semester at institutions with many different technology tools. We hope that you find this useful as you make some decisions for your classes moving forward in the time of COVID-19.

Adam’s Situation

Calendar: My institution is on trimesters, so instead of switching a class to an online-only format midstream, we are starting our spring term courses online. We hope to be in person for the second half of the spring, but I doubt that will happen.

Classes: Introduction to data science (undergrad, 34 students), Statistical consulting (undergrad elective, 15 students)

Switched to: Synchronous online classes on 3/11/2020

Technology: Institution uses Moodle, Zoom/Google Hangouts Meet, Panopto; I’ll also use GitHub and Slack for data science.

Adam’s Thoughts

With my “extended” spring break I am making changes to my two courses for online delivery. Both courses require students to use R extensively. In consulting, students work on a group project all term. In data science, students are learning a lot of fundamental ideas and solving problems for homework. I’m a bit worried about the tech requirements for these courses, but there is no way around it without devising completely new courses, which is out of the question. Luckily, my institution already has an RStudio server that students can use remotely. The IT department is also working to get all students internet access and equipment, but who knows how that will go in practice.

My consulting course was supposed to meet once a week for two hours. This allows students to meet with their groups and check in with me. I plan to schedule 30-minute weekly check-ins with each group. I’ll also be using Moodle heavily to guide project progress and have students submit weekly journals where they outline their progress and reflect on assigned readings.

To adapt data science for remote delivery, I plan to do the following:

  • I need to set student expectations from the start and make sure that we are all in this together. I plan to be very transparent, and openly admit there will be technical snafus and unforeseen struggles in this new format. I also really need to beef up my syllabus with a lot of new statements about these expectations.
  • I am going to shamelessly use existing content, recording new lectures via Panopto only when necessary. These videos will be viewed asynchronously. 
  • Like Laura Ziegler (Part I), I plan to make weekly videos with a recap of last week and a look ahead to this week.
  • I’ll use Slack for discussion outside of class and to answer questions during office hours for students who would prefer that type of Q&A platform.
  • “Class time” on Mondays and Fridays will be similar to office hours, where I answer questions and clarify concepts. On Wednesdays, I will have students work in groups. To allow asynchronous work, after the first week students don’t need to “attend” class, but the assignment will be due the next day.
  • I’ll use Zoom for office hours.
  • I’m abandoning tests in favor of case studies, where students will either write a blog post or record a presentation (using Panopto or similar) that they submit to me.

Steve’s Situation

Classes: Biostat Methods II (MS, about 20 students), Data Visualization in the Health Sciences (MS elective/service course, 6 students)

Switched to: Synchronous online classes on 3/11/2020

Technology: Institution uses Canvas and Zoom

Steve’s Thoughts

After teaching a total of 10 sessions online sessions, here are my thoughts/comments:

My courses are all transitioning to synchronous online sessions at their regularly scheduled times through my university’s Zoom license. So far, Zoom has included enough functionality to allow a fairly painless switch to the online setting. Some of my comments may be specific to my university, so I apologize if not all functionality works at your institution. Main topic of each comment is in bold for easier scanning.

  • Scheduling all Zoom class sessions through my course Canvas pages (Zoom Conferences app) has made it simple to provide students with the appropriate meeting links. One recurring meeting for Monday’s sessions, and one recurring meeting for Wednesday’s sessions. I believe you can also set a static meeting ID so in theory students could use the same exact link for all sessions. Scheduling through Canvas automatically notifies students of the meeting times and saves me the time of announcing the link before every class session.
  • When scheduling Zoom meetings, there is an option to automatically record the class session to either the local computer, or if your license allows it, to the cloud. Once we received permission to record to the cloud, this option was strongly preferred because you can set it to automatically transcribe the audio. While I expect most of my students to show up for the live session, I like the option to upload a recording just in case a student is without internet access or runs into technology issues during the scheduled class time. 
  • I set my Zoom meetings to automatically mute all students upon entering to avoid having a bunch of open microphones as people connect. Students are allowed to unmute their audio/video at any time if they have a question or comment (which works fine in my small classes, e.g. < 20 students). Otherwise, students can interact by using the “hand raise” button or the “clap” emoji which looks like a small hand raise. I can see these indicators by keeping my “manage participants” box visible at all times on one of my two monitors. Additionally, I keep the “chat” box open and visible at all times as well in case students are more comfortable typing than speaking. I am considering transitioning to requiring students to use their video, as sometimes I will ask a question and get no response at all. Two-way video may be important for engagement. 
  • Most of my class sessions involve a mix of verbal lecture, presentation slides, code examples, writing on the board, and exercises for students. Zoom’s presentation tools make it possible to continue using these methods. The screen sharing option in Zoom allows me to share my computer screen, which includes my slides and my R session. While sharing, I typically click the “Annotation” tab and use the “spotlight” feature to highlight my mouse. This makes it easier to track as I move it around the screen to “point” at certain things. 
  • Finally, for a virtual whiteboard, Zoom has a few different options. If you have an iPhone/iPad that you are comfortable writing on, you can share your device’s screen by clicking screen share in Zoom and choosing the iPhone/iPad option (instructions provided by Zoom from that point). Personally, I use the touchscreen on my Chromebook to create a digital whiteboard. To do this, I join the Zoom conference on my Chromebook and share my note taking app with the conference (I use Squid). This allows me to write on my Chromebook tablet with a nice stylus and have the result appear to my class in real time (the delay is very minimal).

Laura’s (L.) Thoughts

For anyone who knows me, I’m very much a people person. I say this because I was a little bummed when I realized that I was only teaching online for the 2019-2020 school year. However, in the wake of recent events (and especially because I’m taking on additional in-person classroom (my kiddos…so is that 2 additional courses?? 🙂 ), I feel I can offer tips and tricks into delivering an online course.

Tip #1: Communication is key!

While this may come as no surprise, I feel it is even more key in the online environment. Communication includes:

  • Instructions on how to navigate the online course, if they aren’t used to doing so (e.g., course overview and orientation). For example, provide a course structure for the rhythm of each week/unit (see this example from my introductory course). 
  • Updating expectations and (possibly) grading for the alternative mode of instruction. Things to think about are should they post questions in the Q&A or via email? How will you handle requests for extensions? Who should they contact first if they have questions: instructor or TAs?
  • As Laura Ziegler said, once a week (at the minimum, twice at the maximum) announcements about the week, upcoming assignments, upcoming assignments, and other important notes. 
  • Providing clear directions on all learning materials.
  • Offering timely, constructive, and frequent feedback on assessments. 
  • Responding to questions or posts in a timely manner (within 12 hours, minimum, and no more than 24 hours, maximum).

For other tips, see the recent StatTLC post by John Haubrick on instructor presence in the online classroom.

Tip #2: Create collaborative keys via Google Docs for activities.

If you have activities in class, move the activity to a Google Doc and have the students create the answer key as a class (we call them collaborative keys). Then the teaching team (instructors and/or TAs) can monitor the key to make sure the responses are on the right track and pose any additional questions. This offers an asynchronous, but effective, method for delivering active learning materials. 

We have been doing this method for a while in our flipped classrooms and for our online courses. It works better in the online environment than in in-person, and it’s actually a beautiful thing to see. There are discussions among the students, students helping other students out, questions being asked that are beyond the question that is asked, etc. We require students to post at least once (although, many go above and beyond that). Here is a document that includes (1) assignment instructions (that has a link to an example Google Doc collaborative key for our Week 1 activity) and (2) “How to contribute to the collaborative key” details on how to participate on the key. 

So, if you do have in-class activities, consider using Google Docs to create a community of learners.

Conclusion

This concludes our editor series on transitioning to the online environment during the COVID19 times. We hope that some of our thoughts and experiences are useful as we all try new things to adapt to the current situation. Feel free to share your thoughts in the comment section below or contact us to contribute a post of your own.

We hope that you and your families remain safe and healthy.

Online Strategies due to COVID-19, Part 1

In this series of posts, the StatTLC blog team describes how we are managing with the abrupt changes to our courses. In this, we share some of our decisions (and the thinking that went into them), the tools we are using, and tips. We are teaching a diverse set of classes this semester at institutions with many different technology tools. We hope that you find this useful as you make some decisions for your classes moving forward in the time of COVID-19.

Doug’s Situation

Classes: Introductory Statistics (algebra-based, about 80 students), Introductory Probability (calculus-based, about 20 students)

Switched to: Asynchronous classes (with synchronous office hours and review sessions)

Technology: Institution uses Moodle with Collaborate Ultra, videos made with a variety of free and open source software that are posted to YouTube

Doug’s Thoughts

  • The introductory statistics courses I teach have a course coordinator – this has added an unanticipated layer of planning to all changes and discussions. 
  • My campus uses Moodle which has built-in integrations with Collaborate Ultra. There seem to be two paradigms emerging: 
    • Use Collaborate Ultra and teach synchronously with recordings made available.
    • Make videos for asynchronous learning and then use Collaborate Ultra for office hours. 
  • I’ve chosen to go with asynchronous videos and then use Collaborate Ultra for office hours because I’m already comfortable with making videos. For colleagues with less experience teaching online at my institution, the synchronous approach with recordings seems to feel more accessible from what I’ve been seeing in emails.
  • I make videos using OBS Studio and upload them to YouTube. It is very easy to embed them into Moodle – just drop in a link. Students are familiar with YouTube, and the process is basically painless. (If I need to edit videos, I use OpenShot Video Editor, but with the amount of videos I’m making I’m just going for quick right now.)
  • If I’m making a video of anything other than PowerPoint, I use PenAttention to highlight my cursor. Remember to zoom in on text and applications (e.g. I normally use a small font in RStudio but enlarge it for videos).
  • All of my YouTube videos are unlisted (anyone can watch, but only with a link). I don’t have a good reason for this instead of making them all public. I make a playlist for each class that I add every germane video to; I share this playlist link with the class often.
  • Some students have logged into my office hours expecting content delivery – establishing norms for virtual office hours seems to be something I need to proactively do. 
  • In changing to online, we’ve also changed our grading schemes. We decided to use two different weighting systems (with students earning the higher grade) because of all the uncertainty surrounding this transition and the lost opportunities for improving grades by doing well on a heavily-weighted final.
  • For many of my students, this is their first online class. Explaining the different options for submitting assignments needs to be explicit. I find myself intentionally repeating the same information in multiple messages and on different platforms (email and Moodle).
  • I’m also teaching an introductory, calculus-based probability course this semester. Many of the students are not math or stats majors and don’t know LaTeX. Being very flexible in terms of how assignments are submitted is key – I’m fully expecting some students to email me photos of their homework, and I’m okay with that. 
  • I’ve decided to try to use Zulip with my probability course (about 20 students). This is similar to the Discord app that Chris Engledowel talks about in his StatTLC post, but also includes the ability to use LaTeX, syntax highlighting for code, and is open source. Lots of pros, but the cons is that it is less familiar to students than Discord. So far about ¼ of the class has signed up, but there has been very little use so far (only been a day or so). Students seem to be using it mostly for private messaging me rather than for interacting with each other.
  • We were supposed to have a project at the end of the probability course. This is still happening, but recognizing that I will be less able to support some students, I have developed a few “canned projects” for students to do that are essentially some readings and problems on new topics. (I would offer the same thing in a statistics class with a few pre-selected datasets rather than having students find their own data.) I am still encouraging students to pursue a more creative project, but I recognize that that is not likely to happen for everyone this semester.
  • This blog post has been circulating among my colleagues and raises some really interesting points. Since reading it, I’ve resisted calls to poll students about their technology availability (e.g. webcams, scanners, printers, etc.) because a) this was never an expectation for the course and b) I know some don’t have them and I will already have to accommodate that. I’m trying to meet students where they are and be even more flexible than usual right now.

Laura’s (Z.) Situation

Classes: Introductory Statistics course (6 sections with about 60 students each, course coordinator), Advanced Regression course (about 28 students, half upper-level undergrad stat majors/half MS non-stat majors)

Switched to: online classes (asynchronous and synchronous)

Technology: (Introductory Statistics course: StatKey and JMP, Advanced Regression course: R)

Laura’s Thoughts

I am currently teaching 2 courses; an introductory statistics course and an advanced regression course. The student audience for these courses are very different, and therefore will teach them online differently. I have been reading an overwhelming amount of tips for teaching online, and I am sharing what I have decided will be my best approach to teaching “online in a hurry.”

These are my recommendations for any course, which I will use for both of my courses:

  • Keep it simple, not just for the students’ benefit, but for your sanity as well.
  • Make videos to share with students.
    • Videos should be short for two reasons. First, students will lose interest if they are too long. Second, if you make a mistake, you won’t have to redo as much.
    • Videos should be imperfect. No matter how much of a perfectionist you are, you need to focus on the bigger picture. Don’t worry about having perfect sound quality, saying “um” too much, or having your kids or cat run into the room. Just get it done and out there for students.
  • Try to keep things as similar to what we would do in class with the possible exception of being asynchronous. Students are stressed and if we can keep things similar to what they knew before, that may help relieve stress.
  • Send a detailed weekly checklist to students with recommended dates on when to have videos watched, upcoming due dates, etc…
  • Avoid sending too many emails. We are getting a lot of emails, so we should expect students are also getting a lot of emails. Try not to overwhelm them. Try to write one email per week, ideally on Mondays, providing an overview of what is to come with the checklist that is kind, empathetic, and encouraging.
  • Don’t forget about your TA’s, they are nervous too! Have weekly online meetings with them to ask them how they are doing. Give them the opportunity to ask questions not just about the course but also about life in general.

For a large introductory statistics course, I have additional recommendations. For some background information, the introductory statistics course I work with has 6 sections, each with approximately 60 students. I am the coordinator for the course, and therefore have been in charge of getting it ready to be online.

  • Teach asynchronous with videos. Students are across the country, in different time zones, with different access to internet.
  • Provide software output on assignments in case students do not have access to software.

For my advanced regression course, I have 28 students. Approximately half are upper-level undergraduate students and the other half are Masters-level, non-statistics students.

  • Have spent a lot of time talking to myself creating the online videos for my introductory statistics students and am missing the live aspect of teaching. I am planning to do synchronous teaching for students who want to attend. I will record the lecture during that time and will post it for students who choose not to attend. This is going against most of the recommendations I have seen, but I am going to give it a go anyways!

My plans are not perfect, and will likely change after the first week of online teaching, but that is OK! Be honest with your students and they will appreciate the effort you go through to help them through this challenging time.

Get the p outta here! Discussing the USCOTS 2019 significance sessions

The theme of this year’s United States Conference on Teaching Statistics (USCOTS) 2019, “Evaluating Evidence,” put an emphasis on the current discussion/debate on p-values and the use of the word “significant” when making statistical conclusions. Conference-wide presentations (1, 2, 3) offered updates to the ASA official statements on p-value based on the special issue of The American Statistician and potential ways to move beyond significance. 

Now that USCOTS is four months behind us, we thought it would be a good idea to reflect on how it has impacted each of our teaching plans for this year. Each member of our editing team has shared their thoughts below. What are yours? [Share your plans in the comments section.]

If you are interested in starting a discussion about “statistical significance” in your own classroom, check out this cooperative learning activity that Laura Ziegler used in her department. 

Impact on Steve’s teaching: 

I left USCOTS feeling cautiously optimistic about the p-value/significance debate. On one hand, I was starting to feel like the discussion was spinning its wheels, focusing only on defining the problem and not on coming up with solutions. On the other hand, I learned about new ideas from numerous presenters that not only focused on alternative ways of evaluating evidence, but also on how to teach these new methods in the classroom. Despite my mixed feelings, the front running solution in my mind is covered in The American Statistician editorial: results blind science publishing (Locascio, 2017; 2019). Not only does results blind publishing mean that studies will be evaluated on the importance of their research questions and appropriateness of their study designs, but it will simultaneously remove limitations inherent to -values and other similar measures that result in intentional (or unintentional) misuses of statistics. I think journals that implement this review strategy will be making a big step in the right direction.

In the classroom this semester, I want to actively reduce the emphasis on p-values and statistical significance to make sure my students are grasping the bigger picture in statistical problem-solving. I think instructors of statistics tend to overemphasize the results of models, which causes students to make quick, straightforward conclusions using p-values. In an attempt to remedy this, I will be making a more conscious effort to prioritize the overarching statistical process during class and on homework feedback.  

Impact on Laura Le’s teaching:

After USCOTS, I brought the conversation back to my teaching team in Biostatistics. There were a few courses that were being revised and so it was a perfect time to discuss what to do about this phrase. We decided to continue using the phrase “statistical significance” in the introductory biostatistics courses, because it is a phrase our students will frequently encounter in the medical and public health literature. Instead, we decided to add some discussions and/or material about what this phrase does and does not mean. For example, in the online course that I redesigned, I incorporated possible implications when a result is or is not statistically significant.

Impact on Laura Ziegler’s teaching:

I attended USCOTS with three colleagues interested in statistics education. As a group, we decided that changes needed to be made with regards to how we teach hypothesis tests at our university. We have many flavors of introductory statistics in our Statistics department, nine to be exact! All instructors have their own methods of teaching, but we decided as a group that we wanted to be unified on how we approach significance. We held multiple meetings open to anyone (faculty or students) to discuss our plans. Participants included people who love p-values to those who did not necessarily think that they needed to be taught in an introductory statistics course. In our first meeting, we participated in an Academic Controversy cooperative activity to start the conversation about p-values. Approximately 50 faculty and students, including statisticians and non-statisticians, attended.

In our next meetings, we all agreed there is a larger conversation to be had about statistical significance, but we decided on the following changes that could be easily implemented this semester in the short term.

  1. Put more effort into teaching practical significance.
  2. Avoid teaching hypothesis tests as a stand-alone statistical method. Emphasize other analysis and discussions should occur along with hypothesis tests such as effect sizes or confidence intervals.
  3. Use a grey scale for significance. We adapted a scale from the CATALST project with a minor change, adding grey!

I personally love these changes, and look forward to hearing more discussions and suggestions on the topic.

Impact on Adam’s teaching:

Since I started teaching introductory statistics as a graduate student, I have taught hypothesis testing via the interpretation of p-values using the sliding scale Laura Z. outlined above, while mentioning the dogmatic p < 0.05 and “ranting” against its use. So why teach the dogmatic interpretation? Well… I usually tell myself it’s because students will see that usage outside of my course and that I am making students statistically-literate citizens… but upon further reflection, that’s simply a justification for writing problems that are easier to grade. Since USCOTS I made a resolution: I will still teach how to interpret a p-value as strength of evidence, but I will not lean on them to help students make overly-simplistic statements about that evidence. Yes, some test questions will be harder to grade, but having students express what evidence actually is/means will be powerful. Further, I will continue to recommend the use of confidence intervals, where possible, as students can see borderline situations—e.g. does the interval (0.001, 0.067) support a meaningful difference? Finally, I resolve to think about how to effectively discuss effect sizes in class. I admit that I am not familiar with how these are used across multiple disciplines, and I am leery of simplistic statements of what effect size is “big” or “interesting”, since these seem dangerously close to “significant”, but they do seem to be better tools. If you want to write a blog post on the topic, let us know!

Impact on Doug’s teaching:

I initially taught p-values with a narrow approach: alpha levels and dichotomous decisions. While I initially used a variety of alpha levels (more than just 0.05), there wasn’t much emphasis on when different alpha levels would be used – it was more procedurally focused. I then broadened my approach by considering different alpha levels for different contexts and emphasizing the relationship between alpha levels, Type I and II errors, and power. My next major teaching shift was to teach significance using the strength of evidence approach (as discussed by Laura Z. above). At my current institution, we emphasize the strength of evidence approach but also teach alpha levels for rejecting/failing to reject the null hypothesis. I present multiple ways to make a conclusion from p-values because it is plausible that students are going to encounter a variety of correct and incorrect uses of p-values after their introductory courses and preparing them as much as possible is key. 

I have also started asking students a follow-up question after they have interpreted the p-value such as “If you were the company in the problem, would you choose to [discuss the different options here]…” Again, this is no panacea, but connecting the evidence back to a (hypothetical) real-world decision seems to make the idea of strength of evidence easier for students to grasp.

After the USCOTS keynote and ensuing discussions, my colleagues and I discussed where we wanted to go with this. We currently have plans for iteratively improving our introductory statistics courses over the next few years, and making changes with regard to p-values is on our list. We don’t know how we will be teaching evidence in a few years, but for now we are planning on having more common assignments that emphasize a variety of different ways of interpreting results. It’s a manageable first step toward something more. 

Referenced USCOTS presentations: 

  1. Opening Session (Beth Chance, Danny Kaplan, Jessica Utts)
  2. Keynote by Ron Wasserstein and Allen Schirm
  3. Keynote by Kari Lock Morgan

Welcome to StatTLC!

Our editorial team welcomes you to the Statistics Teaching and Learning Corner (StatTLC), a virtual place to chat about statistics education. While there are many opportunities for educators to interact and disseminate research at conferences and in academic journals, there are fewer opportunities to informally discuss and share ideas and experiences. We have decided to launch this blog in an effort to share our own ideas and experiences teaching statistics and biostatistics at the college-level, but to also provide a platform for the statistics education community to share their ideas and experiences.

You can expect to see relatively short, digestible posts about teaching and pedagogy resources for both face-to-face and online courses, research with a focus on how to implement the findings in the classroom, and teaching experiences from faculty instructors, researchers, and teaching assistants. Be on the lookout for questions prompts and thought provoking statements to inspire further discussion in the comments section of each post!