Grading with my voice

Cross-posted on my blog:

A few people have asked me for some more detail about how I use screencasting to grade papers so I thought I’d post this.

As a physics professor I don’t grade nearly as many writing assignments as my colleagues in other disciplines but I do take seriously my job to help all my students improve their writing.  When I first started grading such papers I would write comments on each draft and provides a grade via some sort of rubric.  The rubrics started really as requirements that could be met in various ways and have evolved to fully fleshed out rubrics with careful descriptions of what it takes to get a check mark in columns like “meets basic expectations”.

I quickly realized that giving the students the type of feedback I really thought they could use took a lot of time and effort, and I found that meeting with the students about their writing was one of the best ways to do this.  However, those appointments are hard to schedule, especially in a semester like I have now where all three of my lecture courses have heavy writing requirements (my First Year Seminar is writing intensive, my junior-level advanced lab course has writing a grant as 80% of the points, and my fully online course for teachers has a lesson plan as a major assignment).

I realized that the screencasting I was doing for my lectures could also be used to simulate the office experience with students.  I use my pen tablet mouse to mark up the documents digitally (using either Jarnal or Adobe Acrobat) and I use Jing to record my voice while explaining my concerns with the paper.  Here’s an example. In that example you can see how I mix in discussion of content, style, and how it meets the expectations.  You’ll also see that the 5 minute limit that Jing holds me too wasn’t enough (so I just did a second one).  Typically I get my comments done in under the five minute limit (though the reading of a typical paper still takes me something like 20 minutes).

I ride the bus a lot and I like to use that time to grade.  What I’ve taken to doing is marking up the papers in regular ink with little notations to myself about what to say.  When I’m back in my office I then open the digital document, grab my pen tablet, and begin.  I transfer the marks that are necessary all the while holding a pseudo conversation with the student.

I’ve gotten a lot of good feedback about this from my students.  They really like having the screencast at their disposal for pausing and rewinding.  One student told me that he opens his paper on his computer before playing the screencast and he makes changes immediately, while pausing the playback.  Others have said that they really feel they understand what I’m looking for after watching.  There’s also been an interesting study on the preferences of students for the type of feedback they’d like, comparing regular comments, track-changes in Word, track changes with audio, and what I do.   The upshot is that my way was strongly preferred in their survey.

I’d love to hear how others use a similar system.  Feel free to drop me a line here in the comments or on twitter @arundquist.


Recording Students

(cross-posted at

This semester I’m trying two approaches in my teaching that both involve recording students. One has students using screencasting to turn in their homework and the other has students making pencasts of their group work.

Screencasting Homework

I’ve been teaching fully online classes for six years now. In the past my homework collection method has involved students scanning their homework and posting it to my Learning Management System (homebuilt using PHP/MySQL). This works pretty well but it was hard to ensure students were doing their own work. In my in-class courses I solve this problem with daily quizzes based on a randomly selected problem from the assigned set but I couldn’t find an easy way to do this online. One option would be to do timed quizzes in Blackboard or something but students don’t have nice pen mice like I have and so they could only type their answers without the ability to easily write equations and draw figures. This year I decided to do things a little differently.

At the beginning of the week I provide the students with screencast solutions to six problems from the chapter. I make myself available until Friday morning to answer questions in the discussion board and in my online office hours about the concepts of the chapter and the posted problems. Then on Friday morning I post a single homework problem that is due Monday. The students need to solve it, scan it, and then do a screencast of their solution to turn in.

Here are some of the benefits of this method:

  • I only have to grade one problem per student per week.
  • I can hear the students thought process about the problem.
  • Even if they work together or cheat somehow they still need to put it in their own words.


It’s been very interesting to see how a screencast often gets a different grade than the plain scanned document would have. It goes both ways. Sometimes I see a paper that seems technically correct but I hear them describe certain aspects incorrectly. I’ve also heard a student say all the right things while what they have written isn’t technically correct.

I’ve gotten some good feedback from the students doing this (who happen to be teachers working on their physics teaching license) so I think I’ll continue the practice. I’ve also branched out to in-class students, offering this method as a way to make up for missed in-class quizzes.


Group work pencasts

My newest toy this semester is a LiveScribe smartpen (actually eight of them). These pens are incredible! They record both what you write on the page and the audio happening at the same time. When you go back and click on a word it’ll queue up the audio from that moment. You can also post “pencasts” that work the same way only on a web page so that students can access them. Even since I got my first one I’ve found plenty of ways to use it in my work. I originally wanted one to help me take better notes in one-on-one meetings with students where, in the past, I’ve found that I sometimes lose track of promises made by both parties. I certainly use them for that but I’ve also used my pen at campus-wide speakers, doctors appointments, department meetings, and yes classes. What I want to write about here, though, is how I use them in class.

Here’s a breakdown of my hour-long general physics class periods:

  • 10 minutes for a quiz on a randomly selected problem from the previous class period.
  • 10 minutes to recap the material for the day (often prompted by a randomly selected summary posted by one of my students).
  • 15 minutes to answer all the questions posted by my students on the material for the day.
  • 20 minutes for groups to work on the problems assigned (one of which will be randomly selected for the quiz next time).
  • 5 minutes for the groups to record a pencast of a roadmap (not a solution!) for the problem they worked on.


After class I post all the pencasts so that all the students have at least a sense of how to do all the problems when studying for the quiz. A typical day’s daily outline will then have links to all the pencasts along with links to screencasts I’ve posted on the material and any resources I’ve found useful.

The students seem to have fun with these pens. They’ve made several suggestions for how best to use them including hitting the record button when I come around to their group as they’re still trying to understand the problems. I now have eight pens in total and I look forward to finding more ways for students to use them in the future.


Group grading

How do you grade group or team work?  I was speaking with Dave Davies today (who hopefully will have a future guest post on using Twitter in the classroom, by the way) about this issue and it seems we have some similar approaches to this.

For me, I tell my lab groups and my FYSEM groups that 90% of their grade will be based on the group work but that the last 10% will be voted on by the group.  For those last 10 percentage points I give the group 8*(number of group members)+1 to divvy up.  So for a group of four they get 33 points, for example.  I tell the group that they need to meet and mutually decide the distribution of those points and write their decision in the final report for whatever the project is.  I have a few basic ground rules.  The points for any individual must be an integer but numbers greater than ten and less than zero are allowed.

The results have been interesting.  I would say the most common breakdown is 8,8,8,9 for a four-person group but some of the outliers have been interesting.  One of my favorites was 8,8,8,8 where the whole group gave up the last point in the interest of fairness.  I’ve also seen 10,10,10,3 and 18,5,5,5.  I don’t think I’ve seen 33,0,0,0 and I’ve rarely seen negative numbers.  The hallway conversations have been interesting too as I’ve heard students say “I already have my C so I can take a zero.”  This last one bothers me a little as grading shouldn’t be a game but in the end I’m glad they’re all talking about everyone’s contribution.

I came up with this back during the Grant Opportunities for Academic Liaisons with Industry (GOALI) grant when some 3M people were talking to us about the differences between group work and team work.  Clearly the notion of needing to be there for each other but also to talk openly about various contributions is big at 3M.

Professor Davies was talking to me about how he gives every participant some points to divvy among the group.  This is a different take but he sees similar upsides to his approach.  So how do you grade group/team work?

Andy Rundquist “Superfly”

Convince your neighbors

Many of us poll our students during class about conceptual questions.  This technique is one of the hallmarks of Eric Mazur’s “Peer Instruction” teaching method initially developed for teaching introductory physics.  Many at Hamline use “clickers” to accomplish this polling.  What I’d like to talk about in this post is what happens after the polling, something I call “convincing your neighbors.”

Mazur gives a guideline of 80% when deciding whether to have students convince their neighbors.  This means that if more than 80% of the class chooses the same answer (not necessarily the correct answer), it’s not worth class time to have them talk amongst themselves.  If no answer reachers that threshold, however, having the students discuss their opinions is a vital ingredient to learning.

I’ve been teaching this way for ten years now and I’ve seen this work well and I’ve seen it work not so well.  Some of the pitfalls include:

  • Sometimes its hard to get students to find someone who disagrees with them
  • Students don’t always fully commit to their answers
  • Students readily agree to change their mind when confronted with someone they think understands the material better

Recently I taught a class that worked quite well and introduced a new twist.  I put students into groups of three or four for the duration of the class period so that they could work together on sample problems.  For the conceptual information I used my concept quizzes the same as always.  However, this time I said “convince your group” instead of “convince your neighbors.”  I found that this caused much more vibrant conversations.  Each group had a white board to work the sample problems on and often they would use the white board to flesh out the details of the conceptual question.

It seems that by randomly forming groups I dealt with the first point above.  Students weren’t sitting by their friends whom they often study with and so it was rare for a group to all vote the same way.  While students didn’t necessarily commit to their answers any more than in the past, the group dynamic seemed to get them to focus more than I’ve seen before.  As for the last point, because the groups work together on many different things during the class period it seems they are a little less in awe of the “smart” students.

As a side note, I don’t use clickers for this.  This is how I do it.

How have others seen the “convince your neighbors” aspect of conceptual polling work?

Google Reader

I use Google Reader all the time and I find myself telling others about it in all sorts of different situations.  I thought I’d write up some of the things I do with it.

First, what is Google Reader?  It’s an RSS aggregator which means that it pulls in information about articles from websites I’ve shown an interest in and it allows me to browse them all in one place.  Beyond that it lets you tag, organize, and search them as one would expect from a Google product.

What kind of websites?  Here’s a screenshot from my Google Reader showing a portion of what I read:

Andy's feeds

some of my rss feeds

I have a few different newspapers, several research journals along with probably 20 different technology, physics, and teaching blogs that aren’t shown in that screenshot.

What I see when I log in to Google Reader is a list of all unread articles from whichever folder I’ve chosen.  Typically I just look at them all at once.  I get about 150 different articles to read per day and if I fall behind I tend to do a mass “mark as read” just to get them off my to-do list.  It’s easy to navigate through them using all kinds of different keyboard short cuts that Google provides.

Instead of digging too deep into how I use Google Reader, I’d like to focus on why I think it’s a good tool for teachers.  Whereas I started (and continue) to use this tool to stay on top of news and information that I care about, I’ve come to really appreciate the ease with which I’m able to share interesting information with my students and colleagues.

First “sharing”.  Hitting shift-s while looking at an article marks that article to be shared.  All this means is that those marked articles get collected in one place. That becomes a page you can easily share with others.  What’s cool is that it also generates its own rss feed that you can use with little widgets you can add to your other web pages (here’s a link to my home page with such a widget).

Next “tagging”.  While the sharing feature has enabled me to quickly provide a current list of interesting articles to my colleagues, students, and friends, I wanted a way to tailor those articles to particular audiences.  In Google Reader you can do this with tags that are set to be public.  When viewing an article of interest you hit “t” and then start typing a tag.  Google will automatically start to suggest tags that already exist which you can choose from or you can make a new tag.  If the tag is set to be public, it produces its own rss feed and public page.  That one just linked is the list of articles I publish to the Hamline University Physics Department Facebook page.

What about surfing around?  Another cool tool that Google Reader provides is a bookmarklet that allows you to turn any page you visit into an article organized by Google Reader.  In other words, if I’m surfing around and stumble onto a cool web page that I’d like to share, I hit the bookmarklet button in my browser and a window pops up like this:

google reader bookmarklet

Google Reader bookmarklet

If I want to add it to one of my public pages I just hit the “add tags” button and choose the appropriate public tag.  If I “add a note” that text will also show up on the public feed and page.

I’ve found Google Reader to be a great way to stay on top of things and to organize how I provide new information to my students.  I strongly encourage you to check it out.

I’ve also posted a screencast showing how I manage all of this.  Don’t hesitate to drop me a line if you have any questions about how to set this up for yourself.  -SuperFly