Assumptions about online class size are often based on “traditional”, face-to-face (f2f) class experiences. Past studies suggest that the effectiveness of f2f classes is negatively tied to the number of students in the class, and that a range of 12-25 students may be an acceptable plateau for many classes. The ideal may in fact be a 1-to-1, mentor-like relationship between student and instructor–such as described by Bloom as the 2-sigma problem. We can also comfortably state that more students can equal more work for the instructor (more questions, more grading, more interaction)–potentially at the cost of instruction. At the very least we can assume that there are some class sizes at which learning will inevitably suffer (Drago & Peltier, 37), for all courses require some degree of student-instructor interaction, don’t they?

We therefore may be inclined to believe that there must be some correlation in online courses as there is in face-to-face courses. Indeed, a number of commentators on the issue recommend the same class sizes for online courses as have been recommended for traditional courses, even though there is still little research on the effect of class size in online courses. Toth & Montagna conclude that in ten years of research on the subject there is no consistent evidence of a predictable connection between student achievement and online class size. Why might that be? It may be that best practices in education are changing (informed by evolving learning theories and “21st Century” learner needs) such that student-instructor interaction is no longer critical. I think more likely that f2f and online courses are very different animals (though they are not always treated as such by instructors or students). Online courses have a number of specific dimensions of design and instruction that relate to their effectiveness. It may be that through careful course design and effective teaching strategies, online courses can reflect increased class size differently than face to face class size, reducing or even eliminating negative effects depending on the course.

Determining Optimal Class Size for Online Courses

Class size is not the sole predictor of teaching effectiveness in online environments; in “The Effects of Class Size on Effectiveness of Online Courses” Drago & Peltier cite five factors of teaching effectiveness: course structure, course content, instructor support and mentoring, instructor-student interaction, student-student interaction. Of these five they recognize only one as significantly and negatively impacted by class size: instructor-student interaction (31).

Though only one of five factors, instructor-student interacton is by no means a negligible dimension in online courses. Indeed, it may be more important to course effectiveness due to online learning’s larger transactional distance. The fact that students and teachers are rarely if ever in the same place at the same time may lead both to students and instructors to feel isolated, thereby reducing motivation. For students, transactional distance may have the added threat of making students feel unnoticed or unseen, a sense that may increase inclinations toward academic dishonesty. Transactional distance can be counteracted by increasing “presence” of instructor and students in the online environment.

The importance of presence and instructor-student interaction may vary from course to course. Every course is different, and courses and departments have different goals and outcomes to consider. Once we recognize this, we find questions that can help us determine optimal class size:

• Mentoring and feedback. Is learner success reliant on instructor-student interactions? Some courses thrive on learner interaction with informed and knowledgeable experts. If so, decreasing class size may increase instructor-student interaction.
• Teacher load. Do assessments require direct instructor review and feedback? Multiple-choice questions do not, but essay questions do. A course that can be accurately assessed with well-design multiple-choice questions may allow for a greater scale of students than one that needs one that requires more subjective review of individual assessments.
• Classroom community. Could an interdependent learner community support course objectives? If so, large class size may not be an obstacle, but a boon.
• Course curriculum. Are so-called “lower order” thinking skills (remembering, understanding) the focus of the course? This may allow for more objective assessments that can be automated, reducing instructor load.
• Course design. Is the course designed to provide rich, engaging content, relevant learning activities that help learners to take ownership of their learning? A well-designed course that encourages and supports learner independence is probably less reliant on class size for its effectiveness.
• Technology training. Is the instructor fluent with the technology? If s/he is capable of navigating the system rapidly and interacting with students using the best available tools, this may allow for more instructor-student interactions at less cost to their load.
• Course management. Has the instructor good online course management and time-saving strategies? Such strategies can reduce the load of online teaching, allowing instructors to facilitate more students.

Though these questions aim at the core issues of curriculum, course design, and teaching of online courses, they unfortunately are too subjective to predict real numbers for online class size. One must therefore base actual enrollment caps on experience, and it is here that we might as well look to f2f class size as a starting point. It is probably less harmful to students if class size starts as small as possible, increasing semester by semester as informed by both the student and the instructor experiences.

In doing so, here are some ideas to support student success in large online classes without overly taxing the instructor:

• Understand realistically how the online course impacts your faculty load.
• Do not write-off the students as “on their own” or think of Instead, build-in extensive support and maximize contact with students using discussion forums and mass e-mails.
• Invest in developing or implementing cohesive, dynamic, and engaging online course materials as a means of improving the online course overall.
• Train instructors to use the technology to their best benefit, thereby decreasing fear and frustration with the technology while reducing wasted time.
• Teach instructors course management strategies (Turoff & Hiltz) aimed at large courses.
• Shift responsibility for non-critical tasks (i.e. answering e-mail) from the instructor to graders or teaching assistants if possible.
• Leverage class size to the benefit of the students by growing communities of learners, e.g. through groups, peer assessments, etc.

I’m sure there are more ideas out there, and from folks with more experience with large class sizes than me. The real point of this posting is to foster the dialogue through which we might come to practical solutions.

References

• “Class Size – A Dangerous Discussion?” The TLT Group. Retrieved from http://www.tltgroup.org/ProFacDev/DangerousDiscussions/ClassSizeHome.htm
• Colwell, Joy. (2004). “The Upper Limit: The Issues for Faculty in Setting Class Size in Online Courses”. In Proceedings of Teaching Online in Higher Education 2004. Retrieved from http://www.ipfw.edu/tohe/Nov10.htm
• Drago, W & Peltier, J. (2004). The effects of class size on effectiveness of online courses”. Management Research News, 27(10) pp 27 – 41. Retrieved from http://www.emeraldinsight.com/10.1108/01409170410784310
• Online class size: Balancing quality and cost. (2000). Distance Education Report, 4(17), 3.
• Toth, L.S. (2002). “Class size and achievement in higher education: A summary of current research”. College Student Journal, 36(2) pp. 253-261.

Instructional Design & Development

1. ADDIE, Dick & Cary, and other ID models useful to revisit and reflect upon, especially for planning of critical stages of course design. Models do not necessarily inhibit rapid prototyping. UVU IDS might rethink how its course design process fits into such models (instead of immediate prototyping do we first analyze needs, course objectives, then outline content in design? Prototyping should be mid to late design phase).
2. Build instructional design models into project mgmt software as template. We in UVU IDS do something like this with dotProject, but could more explicitly label phases to provide data documentation and data.
3. Log instructional patterns (e.g. activities, interactions, assessments) and create “templates”.
4. Brainstorming ideas: aim for quantity not quality (no negations), stock conference room with treats and toys, limit to 1 hr/break & flow, focus everyone on the task, use online shared docs.
5. In design phase, instead of lists of info, make visual representations like wireframes & site maps. Let it be basic, ugly.
6. Tame your use of media to reduce cognitive load (read Clark’s Efficiency in Design again).
7. Avoid novelty for novelty’s sake (bad example: wave runner game with multiple choice questions, our Frogger game).
8. Try rapid e-learning dev tools like Captivate, Articulate, and Rapid Intake, but beware of outputs that aren’t accessible, web standards format.
9. More often than not Flash is just flashy.
10. PowerPoint continues to be abused by presenters from around the world, resulting in mind-numbing, soul-sucking internments. Folks talk about PPT design in terms of how many bullets, words, fonts per slide, with nearly no mention of “Presentation Zen”.
11. Also, a PPT designed for a (classroom) presentation is not elearning, even if you convert that PPT to Flash with Captivate. @BrianDusablon says #1 problem in e-learning is PPT. (#2 is systems that put obstacles between learner and learning.)
12. However, you can use PPT as an effective e-learning prototyping platform, if not a complete and robust authoring platform.
13. Surprise! Corporate training wants demonstrable outcomes, not rubrics that assert generalized ideals of quality based on theory. How well do our distance learning “quality” rubrics (QM, Chico) measure the actual effectiveness of learning experiences at helping students meet learning objectives? Isn’t that more important than, say, the 7 principles?
14. Though the session was canceled, it appears possible to use Google Spreadsheets as a data source for live e-learning content generation (think flashcards, quizzes)
15. UVU IDS should create all lesson discussions, assignments, and assessments as plain text files WITHIN the lessons folder, until Common Cartridge XML becomes useful on Moodle and Bb Vista.
16. URLs from augmented reality games session: http://argology.org, http://argn.com, http://unfiction.com

Questions

1. Does UVU IDS have enough accurate dotProject data yet to estimate the hour-cost of a new online course?
2. Non-linear ID is still popular, but is it effective? Have we passed the point at which hypertext inhibits learning and focus?
3. Where are communities that share DTDs and schemas, esp for e-learning? Does our XML DTD match any other known e-learning DTDs? (Really only Common Cartridge)

Reading List

1. Goal-based learning design a la Robert Schank
2. Humanizing education articles
3. Videogames and Education: Humanistic Approaches to an Emergent Art Form by Harry J. Brown
4. Chico State’s latest vers. of ROI
5. Survey of Instructional Development Models by Kent L. Gustafson, Robert Maribe Branch
6. Any empirical research on linear vs. non-linear (hypertextual) learning outcomes

Software Demo List

• Articulate (rapid e-learning development)
• Captivate (rapid e-learning development)
• ProForm / Unison (rapid e-learning development)
• Jing (screen capture)
• Agilix BrainHoney (LMS/learning system)

Anyone who knows me knows about the Teaching with Technology Idea Exchange, aka TTIX, a small, free conference dedicated to the open exchange of ideas and best practices in distance learning and technology-enhanced education. If I’m not soliciting proposals, I’m soliciting keynotes, or I’m soliciting sponsors, or I’m soliciting participants–and really, the participants are the most important ingredient for a successful idea exchange!

So let me formally thank and welcome everyone who’s planning on coming to Orem, Utah June 3-5 to present, participate, sponsor, exhibit, help or support. We know some people could not travel this year due to tightening budgets, but we hope to offset that with a sepcial announcement:

Instead of taping and archiving this year’s sessions, we are going to try to live stream each and every TTIX session via UStream. This means that participants from around the world will have a chance to watch and engage with TTIX live audiences via the backchannel. I’ll post more on this soon at http://ttix.org

Finally, of course, Marc and I will be tracking your participation in TTIX via social media to crown this year’s Grand Poobah, and give out some sweet prizes. Every blog post, Twitter update, Flickr photo, Delicious bookmark that you tag TTIX (#TTIX on Twitter) will count towards your effort to claim this coveted award, if you sign up.

(For those of you who have been in a coma for the last 13 years, MP3 is the most popular digital audio file format. It uses a lossy encoding and compression technique to ensure small file size while preserving audio quality.)

A lot of the “MP3″ legal concern is over MP3 encoding or decoding technology, but the patents also affect the distribution of the .mp3 file itself. Emusic.com is just one of many who pay a patent royalty for the distribution of it’s MP3s. But what about educational institutions? For instance, MIT OCW uses MP3 technology for compression of audio lectures; do they pay off Thomson for each file?

Not likely. Thompson’s web site declares…

A license is needed for commercial (i.e., revenue-generating) use of mp3/mp3PRO in broadcast systems (terrestrial, satellite, cable and/or other distribution channels), streaming applications (via Internet, intranets and/or other networks), other content distribution systems (pay-audio or audio-on-demand applications and the like) or for use of mp3/mp3PRO on physical media (compact discs, digital versatile discs, semiconductor chips, hard drives, memory cards and the like).

However, no license is needed for private, non-commercial activities (e.g., home-entertainment, receiving broadcasts and creating a personal music library), not generating revenue or other consideration of any kind or for entities with associated annual gross revenue less than US$ 100 000.00.

This appears to exempt non-profit educational usage, so I e-mailed Thompson directly. Reeder Grant of Thompson replied swiftly and courteously:

If you are not receiving, or you do not anticipate receiving more than $100,000 annually from mp3 distribution activities, you do not need a license. From your described use below (distribution of non-profit educational MP3s), it would seem that you would fall into this latter category of not needing the license.

Good news, right? Based on this information I suggest the following options for educational creation and distribution of .mp3 (which in no way should be considered legal advice):

1. Proceed. Create and distribute .mp3s without a license if they are clearly for non-profit educational purposes. But make sure you use a legally purchased and licensed, or, even better, an open source MP3 encoder like LAME (open source MP3 encoders have been specifically ignored by the legal actions). And consult a lawyer.
2. Contact the patent holder(s). I would go with Thompson for the above-stated reasons. Then consult a lawyer.
3. Break into the open. Use a different, widely supports audio compression format. OGG seems to be the most popular “open” lossy audio compression technology. OGG is not natively supported by iTunes or Windows Media Player (two of the most popular desktop media players), but plug-ins are available for both (make these available to users on the file download pages). No lawyer required.

Hardcore edupunk’s might lean toward option 3, but while nearly everyone is ready to play an MP3 in some form or another, few are ready to play an OGG, and less are comfortable with the change of file format.

(I will add that though I’ve never played with OGG, I am fascinated by the claim that OGG can “contain” lossless audio file information from other formats such as FLAC. FLAC is an open “lossless” audio compression codec, and I’ve always ripped my CDs as FLAC and MP3 simultaneously using EAC, Mareo, and LAME [FLAC for archive, MP3 for portability], but this FLAC + OGG marriage may be more suitable in the longterm.

(If you’re thinking of trying OGG, audiophiles recommend that you not “transcode”, that is, convert from MP3 –> OGG. Convert from CD, original WAV, or other “lossless” format.)

P.S. With respect to the original ITForum question, I should point out that Audacity is not a MP3 encoder as implied, nor does it contain an MP3 encoder, thus exempting it from the patent licensing bane. Audacity is an audio file editing software that uses it’s own file format. Audacity does, however, work with external MP3 encoding software (like LAME). Basically Audacity triggers the external encoder to compress the audio file.

Here’s a higher quality flv version:

Intro to OpenShare for Moodle 1.9 (flv)

We are planning the workshop along two tracts: one for Web 2.0 “beginners” and one for “veterans”, though that may change depending on audience needs. We’ll be covering topics such as blogging, researching online, productivity anywhere, and mashing up your learning environment.

You can take a look at our early drafts of these tracts on Scott’s wiki, and you can register for WCET and the pre-conference workshop right now, at an extended Early-Bird rate.

Certainly if you plan or want to participate in this workshop, send us feedback on what will be most valuable to you.

• LilaTov Cocktail: “Creepy Treehouse” effect: Twitter & Facebook suck when they’re required by your professor
• Marie-Jose Klaver Met je professor in een boomhut (In a Treehouse with Your Professor) (Dutch — does “kippenvel” mean “goosebumps”?)
• AnneC Trouver l’équilibre (French — relates personal experiences with student resistance to instructor-implemented social media)
• History-ing: Creepy Treehouse

The persistence of this discussion should be encouraging for John Krutsch and Marc Hugentobler, who will be presenting at this year’s WCET Annual Conference specifically on the creepy tree house effect in a session titled “Taking the ‘Creepy’ Out of ‘Creepy Tree House’”. I look forward to seeing educators and administrators engage in discussion and debate on the meaningful/meaningless-ness of the term, any deleterious effects it might have on teaching and learning, and how we can leverage technology without wasting our time.

George put his slides for this keynote online on SlideShare. Here are my fast-and-furious, almost-at-George’s-pace notes (which I hope to come back in and edit);

Task of education is to “combat” for lucidity

Knowledge is in the connections
more college students in china than in any other country
we are not in control of where education is going
we are not in control of these tbs of information

Complexity
putting together a puzzle
metaphor of a weather pattern – that’s why we can’t predict (Photo)
education is meant to be more like a puzzle
too much information
we end up with extra nuts and bolts
fragmentation
(I remember reading EVERYTHING in a book, in a newspaper, in a magazine, in a comic—hungry for knowledge. Now there’s too much)
“Fragmentation requires re-creation”
Fragmentation challegene coherance”
freedom of creation = abundance
(how do we filter)
There’s something else I need to read.
Need to filter out the noice, but that’s beyond the capability of our tools
fast-paced deep stuff. I feel like I’m a smooth stone George has skipped across a deep water
Brings up Kerr’s challenge
“Something is happening.
“But is it sufficient to warrant a reconsideration of learning theory?”

Web 2.0 is hype. “I never thought I’d hear myself say that blogs are hype.”

oh shiny object slide (George should use more of these—great response, great illustration)
Long timeline ofslwo change: Information (great slide showing transition upwards)
what do the tools allow us to do that they didn’t before
reminds me of the idea that technology returns us from individual thought (intraspersonal/intraspective) to collective though, or thought heavily influenced by the sometimes rash opinion of others (interpersonal/extraspective). Can we have a balance of these when everything is published open, for everyone.

Gutenberg press was one of those technologies that spilled blood

Let’s look at this; don’t look at the tools. It’s about those bigger factors of openness, access, creation, control.

Connectivism.
Tagged his critics on his del.icio.us account—great modeling of the true scholarly approach toward getting at truth.
How is this unique?
(pause. Man, he’s a bullet train barreling down the track)
a unity of learning and knowledge
not a significant difference between learning and knowledge
learning != process; knowledge != product
Abundance
I say overabundance. Scarcity of quality may remain proportional? Of course not exactly, but there will be a quantity of crap that may equal the proportional quantity of silence we had before the Internet. Now instead of not having enough I have too much information. Instead of being hungry and savoring the crumbs of information, I am overfed and nauseous at the sight of more platters of information.

Levelsof networked learning
Neural-biological
Conceptual
External-social

neural
connectionism and ai
what fires together wires together
biologically learning is creating a network

conceptual

when we make a concept map it makes explicit what we know

the occurrence of words reveals connectedness of concepts to create meaning

do network properties exist at a conceptual level?
We do have network attributes to knowledge seems intuitively right
PERSONAL BRAIN
novak on concept maps (see his delicious)
our concepts are understood by filtering through networks
simulations dont teach us steps, they teach us sequences of patterns
enable individuals to form patterns

external and social
we are connected to each other

Presentation Slides

cheatability_factor.ppt

For this session I added several slides that illustrate my gut reaction to a number of the new technologically-based approaches to inhibit cheating in assessments which I hope you will find amusing.

We had a lively and interactive discussion of the problem of cheating in online courses, and possible approaches to inhibit it. We took one participant through our cheatbility rubric explaining criteria and concepts along the way.

For the first time John administered Buzzword Bingo live in-session with bingo cards printed with key terms from our presentation. We did this not (only) as a self-deprecating joke, but as a means of focusing participant attention on the presenters and the dialog. I believe at least 6 participants scored a prize during this session while playing Buzzword Bingo, though John and Marc had to coax more than one participant to simply shout out BINGO instead of raising their hands!

Presenter: Dam Lim

Learning games are built for mobile devices using SWF and XML. Mr. Lim talks about how he would like a single interface, and comments on possibilities for inter-institutional collaborations. I’m sure John Krutsch will be talking to him afterwards.

How does this work cross-(mobile-)platform? I know you can play YouTube videos, but you can’t generally run Flash content on iPhone or on Treo’s Blazer. Mr. Lim is using Windows Mobile on a PocketPC to demo this.

Plays a YouTube.com videos that show how making of the game works.

YouTube.com video is too small to see. Should make it full-screen.

Quickly demonstrates how the game is made and loaded in PocketPC.

I’d like to see the game itself in greater depth, and discuss how these are implemented into a course structure, and at least the predicted impact on learning.

Sums up with plenty of design challenges–so many broad and disruptive challenges leave me pessimistic.

Shows what Krutsch labels “simulated screenshots” of what a game might look like on iPhone (if it supported Flash).

Exit Mr. Stein.