Evaluation 101

I just emailed one of our subscribers about an evaluation of learning spaces that they're beginning to plan. We've got materials for the kinds of inquiry (focus groups, surveys, trips) that can guide planning. I also suggested he look at "The Flashlight Approach." I concluded with an old thought, which I think is framed more clearly than I usually say it!

"I always start designing a study by trying to identify the most important choices that people need to make, choices about which they are still reasonably uncertain. And I mean “reasonably” in both senses of the term:

1. In the minds of these folks, the uncertainty about what to do is significant, and
   
2. the uncertainty can be reduced by looking into evidence. Before the study, they may be leaning one way or another, but they could still be convinced to change their minds and make a different choice if they were to see evidence that my study would be capable of producing.
   

Grant-writing advice(2): Grant proposals as an example of teaching over distance

As the author of a grant proposal, is this the process you're hoping will happen?
1. Two or more people working for the funding program separately read your proposal
2. A day or more after reading it, the group discusses a big stack of proposals. Buried in that stack is yours. One of them says to the other(s), "Which proposal are we going to talk about now?"
3. Someone replies, "Oh this is the one that (and they summarize it in only 2-3 sentences, concluding with what they like about it, compared with the others in the stack.)
4. The first person replies, "Oh yes, I remember that one now! Here's what I think."
5. Hopefully the discussion advances your proposal toward funding.

Focus on step #3. How likely is that to happen? Is what you're proposing so clear and compelling that a day later, your 'distant learner' will remember what they liked about it?

Try an experiment. Give your draft proposal to a stranger who knows nothing about it (tell them nothing except that this is a grant proposal), have them read it, give them a day, and then ask them to summarize briefly what you've asked for and what they think of the proposal. If they can't do it to your satisfaction, consider rewriting the proposal.

If you like this idea, don't do what most people do -- don't let the writing process lag so near to the deadline that there's no time to get this kind of review and then (if need be) reorganize and rewrite the text. Hint: if you have time for nothing else, focus on the title and the abstract.

If you like this note and would like to see more on how to write successful grant proposals, please let me know!


If you missed Advice #1, here it is in a nutshell: Don't use the saying I invented, "So far as I know, no one has ever done this"),

We're experimenting with a Ning site. My post on Ning makes the argument that 'putting a course online' is a goal that's likely to lead to problems and that, instead, we should think about redesigning courses so that, in some ways at least, they're dramatically better than the same topic taught in a traditional campus-bound way. Take a look and try using Ning to engage in a conversation. And let us know what you think of Ning, versus this blog. Thanks!

Can "meaningful" higher education embrace technology AND personal interaction?

"In an age when higher education is threatened with a relentless technology that threatens to dispense with human beings altogether, Professor Van Doren exemplified a tradition of inquiry that celebrates personal interaction as the path to a meaningful education — one shaped by spontaneity, emotion and, yes, reverence." … "Mark Van Doren believed that young people were intuitively capable of grasping even the most complex literature."

Above from: "Notebook: Kerouac Got an A," by Adam Van Doren, NY Times, "Education Life," July 27, 2008, p. 38. "A grandson remembers the professor who inspired a generation of literary giants."

For more excerpts and links to full text, see the following and below.
"Seminar over, the students would follow their professor out of the classroom, conversing as he crossed the campus. Down into the subway they would go, even onto the train, not stopping until they reached the entrance to his townhouse in Greenwich Village. Finally, their professor, genially and perhaps reluctantly, would say, “Well, that’s it, boys, I’ve got to call it a day,” and then he would close the door, leaving those impressionable young men dazed on the sidewalk, wondering how they had landed so far from Morningside Heights." …

"My grandfather maintained bonds with students that lasted long after they graduated, and they in turn revered him." …

“Mark’s questions were very good and if you tried to answer them intelligently,” Merton wrote, “you found yourself saying excellent things that you did not know you knew, and that you had not, in fact, known before. He had ‘educed’ them from you by this question.” …

"The professor kept corresponding with these college youths years later, suggesting career paths, critiquing their manuscripts, promoting their work — even writing poems about them. “Death of a Monk (T. M.)” was written shortly after Merton’s death. He contacted publishers about promising students, encouraging Kerouac to publish “The Town and the City.” Kerouac quit the football team after getting an A in “Shakespeare.” (It should be noted that though the ledgers show my grandfather was a tough grader, those who would go on to make a literary impression on the world also did so in his classroom.)"

FOR FULL TEXT:

Permalink as of 20080727

Technique for breaking a workshop or class into small groups

Problem: you're running a large meeting or workshop, or teaching a large course. You break people into parallel groups, each working on the same task. Someone keeps notes for each group, on a pad or on an easel. How can you facilitate all those groups at once? After some minutes, it's time for 'reporting out.' How can you keep the reporting from taking too long?

Here's one way to do it. The University of Queensland in Australia (UQ) tried this approach for an ePortfolio workshop in mid-May. (We were helping them with planning and running the workshop.) They planned to divide 70 people into about 5 working groups, each sharing experiences on the same topics. (And, to complicate things slightly, I was assigned to pull the threads together but I was in the US.)

Know about Google Spreadsheets? It's a free web service that allows you to create spreadsheets just using a web browser. Better yet, more than one person can see, and write on, the spreadsheet at the same time.

So the UQ team created a Google spreadsheet with 5 identical worksheets, one for each group. Each group had a facilitator and an 'eScribe' with a laptop, who had already briefed. When the groups started their work, the eScribes already had open spreadsheets with the topics already written down the left hand column of their worksheet. (Each small group was assigned a differetn worksheet - a different tab - in the same spreadsheet; when the groups started, these served as templates for taking notes.)

I had been asked to facilitate all these groups. As it happened, I could not be in Brisbane that day. In fact, I was doing some work at the University of Alabama in Tuscaloosa. At the time of the workshop in Brisbane, I was sitting alone in a classroom with laptop. I could flick from one worksheet to another almost instantly by clicking on the tabs at the bottom of the page, watching the eScribes' notes appear cell by cell.

Each time I saw a note that I wanted that group to "report out," I highlighted that cell in yellow. And they could instantly see my highlights appear! By highlighting only 2-3 cells per group (and making sure there were no duplicates across groups), we could assure brief reports that would probably interest the other groups. After each reporter concluded (just a minute or two per team), I got on speaker (via VoIP), summarizing to the whole workshpo what I'd learned from reading all the notes from all the groups.

Advantages over traditional techniques: because the comments were typed online rather than handwritten, I didn't have to decipher their writing. I didn't have to spend time walking from group to group , and disrupting them as I walked. Instead, I was clicking from one group to the next about every 15 seconds, watching their notes appear. If I had a question or noticed a problem, I could use Skype or a chat window to communicate with Brisbane. If necessary someone could walk over to that small group and relay my question.

I'll never do breakouts the same way again, even in person. Try it! And thanks to the folks at UQ for coming up with this technique!

Strategic thinking for programs

A conversation with Mace Mentch of Case Western Reserve shed some fresh light on my recent blog post on a multi-faceted strategy for supporting improvement of selected teaching/learning activities.

Mace and I talked about how it's becoming somewhat more common to talk about the goals of academic programs in terms of learning outcomes. Because these learning outcomes are often defined as skills (e.g., design skills), a logical next step is to ask how often students are practicing, and receiving coaching, in skills. In effect, attention shifts from the outome-as-noun to the outcome-as-verb.

It's a shift people have talked about for decades:

• From teaching as explaining, with IT providing an explaining tool for faculty, toward a second, and perhaps greater emphasis on
• Learning as practicing, with coaching from experts, peers and clients, with IT providing tools for practicing, coaching and other forms of communication.

Big switch, and one that's not possible unless students have, and can use, appropriate technology for that discipline.

I'm not saying anything new when I point out that this sometimes enables faculty to exploit parallels between their own (technology-enabled) research and the learning of students, using comparable tools to learn by working on comparable problems. We studied several examples of that in our study of the iCampus program at MIT.

So here it is in a nutshell:
1. Identify one or more skills of graduates that some faculty want to improve (poor to good; great to world class) (e.g., writing, design, research, composition, academic argument, performance, ...)
2. To develop those skills over 3-4 years, what activities do students need to practice more, have coached better etc.? What tools and resources do they need in order to practice?
3. Develop accessible, modular resources and services that faculty can use to learn additional ways to:

• develop assignments and teaching skills to help students develop the target skills.
  
• assess how well those assignments are doing, in order to improve them;
• develop wisdom for dealing with the problems they're likely to encounter when their courses spend more time on helping students practice and improve these skills.

That set of bullets was the topic of the last post.

Many institutions, like Case, are making progress on item #2 in the list above: faculty are developing or getting access to digital libraries, easy-to-use software tools and other technologies needed in order for students to practice and develop sophisticated academic skills.

It's time to attend to #3: the many small steps that faculty will need to take in order to successfully build their courses more around the practice, coaching, and assessment of such skills. That's an especially exciting challenge with large courses!

Watch the doughnut, not the hole

The TLT Group's work has several hallmarks: taking low threshold steps forward; collaboration; and making a pluralistic, constructive approach to assessment. But arguably the most important, most counter-intuitive, and most pervasive theme of our work over the last decade has been our focus on 'activities.'

In contrast to our approach, the typical technology initiative focuses on -- surprise! -- technology. Buy new, fast, and inexpensive technology. Help faculty learn how to like and use that technology. Evaluate that technology (occasionally). And, all too soon, regret that the technology is now so old, slow, and expensive, and replace it with the technology that is newer, faster, and less expensive.

In contrast, The TLT Group usually focuses on what faculty, students and staff are likely to do with technology: how technology can potentially enable them to alter specific patterns of teaching/learning activity, e.g.,

• discussion,
  
• faculty-student contact,
  
• peer instruction,
  
• thinking through complex problems,
  
• calling up visual images and pointing to elements of those images while talking about them,
  
• integrative thinking,
  
• active learning,
  
• teaching in ways that take instructional advantage of diversity, ...

Our approaches to faculty support and course improvement, to cost modeling and time-saving, and to formative evaluation all focus on helping educators and institutions improve teaching/learning activities over time: small steps and, ultimately, larger changes.

Focusing on activities has several advantages:

• Activities drive outcomes, including learning, retention and costs. so the most direct way to change an outcome is usually to alter activities.
  
• Technology is never the only ingredient for improving outcomes; by focusing on the activity, we tend to notice the non-technological changes that also need to be made (e.g., reward systems, policies, partnerships, ...)
  
• Activities change slowly, in part because they are influenced by more than just the technology. It often takes several generational changes in technology before an activity and its outcome can be altered significantly. So plans to change outcomes need to be on a longer time scale; we help programs budget time and money in ways more likely to pay off in real improvement in learning.
  

Many of our materials and services focus on activities:

• Faculty support and course improvement (e.g., ideas for TLT sorted by Chickering and Gamson's seven principles of good practice; assessment workshop materials organized by activities in a course)
  
• Formative evaluation (evaluation questions sorted by the seven principles and other crucial teaching/learning activities, as well as technologies used for those specific activities)
  
• Learning space design and evaluation
• TLT case studies for faculty workshops, each focused on problems that can arise when faculty change course activities
• Our materials about content change are also organized around teaching/learning activities such as (digital) writing, research by students, and learning about other cultures.
  

Here's the point. In the past, our development of materials for professional and faculty development, evaluation, learning spaces, etc. has been opportunistic. While developing materials about the seven principles in one arena we might be focusing on diversity or visualization in another.

In the next stage of development of materials and services we are consider picking one or more activities and then developing a suite of faculty development workshops, evaluation templates, learning space examples, etc. needed to improve that activity. And, in true TLT Group fashion, we'd look at this 'warts and all,' studying the dangerous discussions and dilemmas that can arise in the process of making such changes.

Does that make sense? If so, which activities would be most important for The TLT Group to support at your own program or institution? if we wanted grant support for a consortial approach to such development, which activities might draw funder support?

• Teaching diverse classes and workshops?
  
• Using evidence to improve practice, in courses and services? ("culture of evidence")
  
• Digital writing Across the Curriculum?
  
• Learning communities?
  
• The 'seven principles?'
• something else?
  

Where should we work first in 2008-2009?

Organizing workshops for participants who differ in needs, excitement, ...

The TLT Group is well-known for its faculty and staff workshops. Here's one technique we've used to organize workshops for faculty who differ in many ways: in skills, in degree of excitement, and in reasons for attending.

Several years ago, the Butler University College of Pharmacy and Health Sciences decided to plan a laptop requirement for its entering students. In the 18 months leading up to the entry of the first laptop cohort, faculty would redesign their courses in order to take advantage of the new technology. The University subscribed to The TLT Group to get some help for those 50 or so faculty. We planned a day-long workshop for Butler staff. As part of that workshop, we wanted faculty to learn from the findings and experience of an expert at another university.

Usually, a workshop has to pick one of two strategies to help participants learn from an expert. The expensive option is to bring the expert to campus to be a featured speaker in the workshop. The inexpensive alternative is to have participants read an article written by the expert.

We used a technique that combines virtues of those two approaches, while also doing better than either at meeting the needs of a diverse group of faculty.

1. Before arriving at the workshop, participating faculty read an article by the expert.
2. At the workshop, we spent about 10 minutes planning questions to ask the expert. As they read this long and provocativ research report, our faculty had noticed different things, liked different things, and objected to different things. So there were lots of questions and comments. (At the time we had picked the article, we had asked the author if he would be willing to take our call at the time of the workshop so that our faculty could ask him questions. He was delighted to do so.)
3. Then we called him and began with the questions we'd planned, as faculty handed a microphone from one questioner to the next. We could hear our expert over a speaker. He didn't give a talk; he just responded to our questions and comments about his article. We actually spent far more time quizzing the expert then if he had given a talk.

Cost: just the price of the phone call.

Time-saving: People were able to read the article (which was quite long- 15 or 20 pages) in far less time than a speech would have taken. Even with the time for discussion, it was a great time-saver. : each person could focus just on those points that had excited or provoked them the most.

Impact of Class Size on Learning Outcomes?

Someone posted that question on the Professional and Organizational Development (POD) listserv this morning. I thought you might like to know how I responded:

There is no general relationship between class size and learning outcomes. (Think of a self-help book that sells millions of copies, or a PBS program on the civil war, if you'd like an analogy. It's possible for people to learn relatively sophisticated lessons even when "class sizes" are enormous. And the learning outcomes of presentation, whether delivered by book, video or by uninterrupted lecturing, aren't affected much by class size).

The research only gets interesting when you begin by specifying some teaching/learning activity and then asking about

1. the impact of class size on its feasibility, and
2. the outcome when that pedagogy is in successful use.

I call these three elements of resource/technology (class size), teaching/learning activity, and outcomes a "triad." The resource or technology offers choices. The activity is what people actually did with the choices. And the outcomes result from those activities. (That simple grammar suggests some powerful questions for the scholarship of teaching and learning in any course or set of courses.)

Some important teaching/learning activities do far better in small classes, and some learning outcomes are much easier to achieve with those activities. (e.g., seminar style discussion; activities that require personal coaching by an expert) But the threshold (the class size where the pedagogy breaks down) varies: I used to work at Evergreen where seminars of 20 students worked pretty well. But a master class in violin would certainly work much better with a 1:1 ratio than with 20:1.

Of course, if you summarize studies of class size that include all sorts of pedagogies and all sorts of outcomes, some of which involve lots of interaction, you'll find a small relationship between class size and outcomes. Which is what Pascarella and Terenzini report in their synthesis of research on learning, How College Affects Students.

Your question is an example of a larger class of questions about the relationships of inputs or technologies to learning outcomes (e.g., how does the size of the library affect learning outcomes? distance learning v. campus classrooms? highly paid profs v. low paid profs?) The role of resources in learning, even a scarce resource such as the attention a professor might pay to a student, is to create possibilities for action.

The power comes from the teaching/learning activities - what faculty and students actually do in those classes. That's why you'll see relatively significant findings on studies of learning where the independent variable is a powerful pedagogy, but not where the focus is a resource or a technology. For examples of such powerful pedagogies, take a look at the P&T book above or at the 2007 annual report from the National Survey of Student Engagement or reports from the Association of American Colleges and Universities.

Steve Ehrmann

Introducing The TLT Group: Why Subscribe?

As you probably know, Steve Gilbert has been doing pioneering work on brief hybrid workshops - workshops that:

• Are only 5-15 minutes long (short enough that people will be able to find a moment to do the workshop, even on a busy day);
  
• Involve a little learning by doing and interpersonal interaction; and
  
• Are well enough supported with eClips and other materials so that peers can use them to help peers to learn something useful.

The other day Steve challenged me to create some brief hybrid workshops about The TLT Group. The notion is that these materials would be used by someone who would like to tell colleagues about The TLT Group so that they could discuss whether or not their institution should subscribe. Here's my first attempt at such a workshop: http://www.tltgroup.org/about/intro.htm

I've tried it yesterday while visiting the University of Alabama (where we just did a great workshop, one of the goals of which was to spread the word about all the different benefits of their subscription). Ideally people at our workshop can now use these materials if they want to tell other people (who didn't attend) about what their subscription offers.

What do you think? Would you use these materials to help your colleagues learn about The TLT Group? If the workshop design or materials needs to be tweaked before you'd use it, what would you like us to change?

Thanks for your help!

Faculty Learning Communities: A New Strategy

Imagine a faculty learning community - a community of inquiry and evidence - that is doing collaborative research in order to improve its members' success in 'teaching online' (or teaching with ePortfolios, or teaching with student response systems, or fostering digital writing, or teaching in a computer classroom, or any other technique or technology you choose). The TLT Group and Washington State are now beta testing a new kind of survey tool that opens up new avenues of research for such communities. The tool is Flashlight Online 2.0 and the strategy it supports is called a 'matrix survey.'

Let's imagine that the faculty learning community is going to do research together on its members' uses of ePortfolios, in order to discover ways in which its members could improve learning in their courses.

Up until now, if the community were interested in using a survey or interview protocol, members might have assumed that all students in all classes would need to be asked the same questions, so that faculty could pool their data:

• "How satisfied are you with the use of ePortfolios in this course?"
  
• "Is the ePortfolio easy to use?"
  
• "How much is the ePortfolio helping you learn?"

These are 'lowest common denominator' questions. After the community gets answers to questions, would they really understand student learning any better? would they have learned something useful for improving learning? Probably not. The questions are so sterile because they have to be conceptually broad enough to cover the wide variety of uses of ePortfolios across courses.

And there are many activities for which ePortfolios can be used, but only a few are likely to be important in any one course. Here's a list that Susan Kahn (IUPUI) and I developed:

1. Reflection as a means of deepening learning
2. Integrate/synthesize prior learning and course learning
3. Student academic self-assessment, guidance (within degree program)
4. Building a sense of professional identity
5. Personal/developmental: Each student develops/describes own goals & abilities
6. Audiences and assessors for the student’- more, better
7. Learning communities, support of
8. Department reframes major in terms of competences across courses
9. Faculty share practices, perspectives
10. Job and school applications
11. External accountability

An activity that specific is much easier to study: do the students understand the goal? what are the software's strengths and weaknesses when used for that purpose? what factors are helping and hindering the student in doing this particular thing? and so on.

Research would be much easier for a community if all its members used ePortfolios the same way. But the needs of real faculty learning communities is seldom this neat. Different faculty are likely to use ePortfolios in different ways. And that's where the matrix survey can play such an important role in supporting more focused, flexible research.

Imagine members of the community working together to develop a useful set of student feedback questions for each of those eleven activities above (assuming each of those activities was a goal for at least one of the participating faculty).

Flashlight Online 2.0 enables each faculty member to fill in a menu specifying which of those eleven activities are important for her course, automatically a student response form that contains only questions about those particular activities.

Once the students have responded:
a) the individual faculty member can get a report on how her students have responded; and
b) the faculty learning community can get a report on how all their students have responded, activity by activity, for all eleven activities. Imagine that there are 25 courses being studied. The report on activity 1 might come from 17 of those courses, on activity B from 4 of those courses, and so on. The community could use this pool of data to study each activity, searching for insights from the larger pool of data. I emphasize activity F in my course and saw something that surprised me from my own students: is that typical for courses that emphasize activity F? or unusual?

Such a survey is not just a one-time thing. The faculty learning community could use its matrix survey term after term, even if different members of the community were on different academic schedules (or even at different institutions), accumulating data.

To read more about matrix surveys, with examples of such survey questions for different uses of ePortfolios, click here.

To comment on this idea or ask questions, please post a comment below.

And, most importantly, if you are interested in working with us on developing such a matrix survey strategy for a faculty learning community (or any other purpose), please contact me.

Fostering Incremental, Cumulative Improvement

For over a decade, Steve Gilbert and I have been working on ways to help institutions foster incremental, cumulative improvements in teaching and learning with technology.

It's easier to attract attention to the opposite: big grants and top-down redesign. What's wrong with that? Nothing, but that 'big push' approach is quite limited, especially in times of tight budgets:

• It's hard, and rare, to get such chunks of time and money
• What do you do a few years later, once the initial capital has been expended? does the innovation that stagnate, regress, or what?
• Most change in higher education can't be achieved through such an approach. Most faculty, most courses, and most student learning experiences would remain untouched.
  

To foster an incremental, cumulative improvement in teaching and learning (with technology) here are some ideas, materials and tools that The TLT Group has developed for its subscribers.

As you'll see, many of these items have a common working assumption: most faculty are interested in improving their teaching, and the learning of their students, to the extent that each such improvement can be achieved without undue time or risk.

• Identify "low threshold" ideas for improving teaching and learning, and circulate them (e.g. via frequent, brief, read-at-a-glance e-mails.)
  
• Offer brief, hybrid workshops (BHWs) to faculty - each just 5-15 minutes -- to help them learn low threshold ways of improving their teaching. These workshops should be well-documented (e.g. with brief online tutorials, brief testimonials from faculty who have used the ideas in the past). If faculty participate in such a workshop (e.g., as an agenda item in a departmental faculty meeting; during a brown bag lunch), try the idea in their own courses, and like the results, they should be able to use the same workshop materials to help other interested colleagues to adapt the same techniques. In other words, a good BHW can be offered peer-to-peer, without always requiring a highly skilled leader.
• Train students to help faculty improve teaching, especially teaching and learning with technology. For years, The TLT Group has offered subscribers help in upgrading their programs for student technology assistants so that the students can provide this kind of aid to faculty.
• Help faculty use student feedback in order to figure out how to improve their teaching. The TLT/Flashlight program to develop such brief hybrid workshops is called "Asking the Right Questions" (ARQ).
  
• Support faculty who act as 'compassionate pioneers' (e.g., helping their colleagues use ideas that the first faculty member has already tried)
• Stretch and coordinate the use of scarce faculty support resources by developing a virtual Teaching, Learning, and Technology Center. In strategy for encouraging incremental, cumulative improvement, faculty have to do much of the work: identifying a need and making an improvement. But that doesn't mean they're completely unaided. So how can the institution speed help to all interested faculty when, at almost every institution, support resources are splintered among faculty development, information technology, the library, distance learning, individual departments, and other units? By having as many of those support staff as possible work (in many ways) as though they were one unit. Such a virtual TLT Center can have a single help line, a single calendar, co-planning of events, shared development of its (their) own staff, etc.
• The study done by TLT/Flashlight of the MIT/Microsoft iCampus program revealed more ways to support incremental, incremental change. For example, it is far easier for faculty member B to understand, appreciate, and implement a teaching idea from faculty member A if A & B teach the same course, perhaps with the same text, perhaps with the same academic calendar, with a similar approach, and similar concerns. In that case, B can 'grok' what A is saying almost instantly, decide quickly whether to try the idea, and (the first time) try it by copying what A did. It's a relatively low threshold process. So the trick is to find a way to pair A and B. Interestingly, the larger the universe of faculty (e.g., a big system of institutions, a statewide effort, a national professional association), the easier it is to create such pairs and small groups of like-minded faculty.

Everything I've written so far focuses on how to invest in and support incremental improvement. What about making enough such small changes, and congruent changes, so that the results are 'cumulative?" Over time, we'd like the results of all these tiny changes become apparent to instructors, and also to students, departments, alumni, accreditors, etc.

At most institutions, the job of larger entities (departments, professional associations, institutions, libraries) is to work to support all kinds of improvement (and thereby offend nobody). The job of the faculty member is 'merely' to choose.

There are real problems with that working model. Few of those larger entities can afford to support all conceivable incremental improvements. And few faculty have the time to scan all those opportunities, and then choose one.

Doesn't it make more sense for those larger entities (departments, associations, and institutions) to choose some directions for improvement, and focus on incremental improvements in those directions.

I'm indebted to Chris Alexander of Berkeley for that insight. Decades ago, in his book The Oregon Experiment, Alexander rejected the idea of campus master plans and instead argued for focused support of incremental change in the university's physical infrastructure. The goals would be agreed on and adjusted through community governance.

As it happens, The TLT Group also helps institutions organize Teaching, Learning, and Technology Roundtables. One role that such TLTRs can play is to pick some goals for incremental instructional improvement, and then help the institution maintain its focus for enough years so that the results become visible, valued, and rewarded. What kinds of goals are worth the candle? How about "learning community," "undergraduate research," "writing across the curriculum," or "building a culture of evidence"? The important thing, for departments as well as institutions, is to pick a few such goals of real importance, where systematic support of incremental, cumulative improvement can really pay off for students, faculty, the department, and the institution.

Your comments? What strategies have I missed? What resources?