Full Discussion - Class Size -Strategies for Innovation in Community College ESL, February 4 - 8, 2008

Class Size; Defense Language Institute English Language Center (DLIELC) Described

A discussion of how class size affects performance and outcomes; a description of the Defense Language Institute English Language Center and how it addresses several variables including class size and instructional time.


In the long run, this may be all that I know.

Ted Klein


  1. Were there 10 or fewer students in the class?
  2. Was the classroom comfortable in terms of environment and learning atmosphere?
  3. Did the instructor have a pleasant and supportive personality?
  4. Were the lessons communication centered, rather than informational, most of the time?
  5. Was the instructor a native-speaker or near native-speaker of the target language?
  6. Was the target language used as a medium of instruction all or most of the time?
  7. Did the students do most of the communication, rather than the instructor?
  8. Did the instructor maintain control of the class in a non-threatening manner?
  9. Did members of the class seem compatible with each other and the instructor?
  10. Did the students seem closely matched in their target language proficiency?
  11. Did all of the students participate?
  12. Were students enthusiastic?
  13. Did the instructor use a variety of techniques to elicit communication activities?
  14. Did the instructor assist students, rather than push them?
  15. Did the instructor use normal, rather than exaggerated speech?
  16. Were training aids used to enhance or reinforce results?
  17. Were new learning objectives reinforced adequately?
  18. Was correction applied moderately and positively so that it wouldn't inhibit communication?
  19. Was there a balance of language skills (listening, speaking, reading and writing?)
  20. Were students dealt with appropriately for their ages? (e.g. adults treated like adults).

Hi Ted,

Thanks for this. This is a great list - did you generate it yourself?

I guess I have a bunch of questions for you about it:

How do you use it? As a general guide, or do you deliberately try to address each item? Are you the only one who uses this, or do others you work with also use it?

Do you find that if you adhere to these principles, that the students advance?

Marie Cora


I did this list years ago based on literally decades in and out of the U.S.A. teaching, training teachers, supervising, coordinating, etc. in ESL. It is based on what seems to work or not work. I'm proud to say that I'm back in the ESL trenches after, among other things, twenty years with the Defense Language Institute English Language Center (http://www.dlielc.org/welcome_dli/index.html). I've been teaching immigrants part time for the last eight years for the Adult Education Department at Austin Community College. Getting back in the trenches has reminded me of what language teaching is all about. I feel sorry for anybody who has to work at a higher level, because that's really not as much fun! I truly hope that I apply everything on my list daily and don't fall into any of the "easy traps." I have distributed this list over the years to anybody who seemed interested and it is published on my website at http://www.tedkleinesl.com/ESL/20questions.html Feel absolutely free to use it in any way that will make life easier for students. Thank you very much for the input. Questions are welcome.

Ted Klein


I like the list very much too, and hope you can answer Marie's questions. Also, how large is your program, how many hours/week do classes meet, and how would you "grade" your program using this list? What percentage of your students advance a level (or more) in a year (or more)? You don't list intensity and duration of instruction - two factors that Jodi Crandall and I have found important. Do you find these important too, or do you think those factors "wash out" if the principles you list are followed?

Forrest Chisman

I have not been a part of this discussion and I really liked the tool that Ted has shared with us. However, I have question and I hope that it is not something that has already been asked and answered.

The first item on Ted's list is a little confusing. In most of our ESL classes we enroll more than 10 students because of fiscal constraints and the need for ESL in the community. So is it a negative or a positive to have fewer than 10 students in a class? In our case, we expect to see more than 10 students in a class and for the teacher to sustain the numbers.

As for the achievement gap, it is huge issue in all literacy programs because of many socio-economic factors.

In our area, part of the Bay Area, the boom in the housing market (in past several years) and high rents made it difficult for people to stay in one neighborhood. Therefore they constantly move (this is seen more in people who do not have high levels of education from their native country).

People with a certain level of education (college degrees from their countries are more likely to find stable jobs and have some kind of community support). Most other people hold two or more jobs, go in and out of classes, change schedules, and finally drop out because of various constraints. More than likely, they lack study skills and have no time to practice.

The achievement gap stems not only from the differences in educational levels of immigrants, but also due the huge difference in the availability of community resources.

Usha Narayanan


I'm not sure where Ted came up with that number. Ted, is it your experience that with more than 10 students, learning decreases? If so, how do you fund that number? I think a lot of people would be interested in ways to decrease class size.

It's more typical to have larger classes because of the funding constraints you talk about. With more students, it becomes more important to involve them in activities in which all get to participate, which means less teacher talk and more student interaction. But even small classes need that.

What do others feel about the "ideal class size"?

You have also identified some of the major reasons adults drop out of classes (or opt out, only to return at a later date). Do any of you keep records of your students that would identify those who do return? Do you have any idea of whether they have tried to continue learning English outside of the classroom and how they did this? I don't know of any research about adult English Language Learners in this area, but there is an ongoing study by Stephen Reder and others at Portland State University (see The Longitudinal Study of Adult Learning at http://www.lsal.pdx.edu/index.php) following adult literacy level students for several years. They have identified some ways in which adults continue learning outside of the classroom and also that some of these learners come back to classes after being out of them for some time.

Providing support services is always a challenge. Have any of you been able to partner with other organizations to reduce the cost of these services to your program? What kinds of partnerships have been most effective? If you teach in a community college, have your students had access to the various support services provided to other students?

Several of you have talked about the differences in progress made by students with more advanced education and those who are at literacy level. Because literacy level students take longer in making progress, most programs provide separate classes for literacy level students and literate beginners. Those learners with limited formal schooling and literacy will need more time to make progress. I'm going to ask Forrest to talk about what he and Steve Spurling and Sharon Seymour found out about persistence of literacy level students and their learning gains.

Students with advanced education in their own language may be able to have a condensed program since they are already experienced as students and often have high motivation to get through English so that they can take courses related to their previous or future career. City College of San Francisco offers an "accelerated course" in which 2 semesters worth of work is taught during one. Do any of your programs offer something along these lines?

Jodi Crandall

Hi Jodi,

Greetings from Lake Travis in Texas. We met a couple of times in the past, I think at least once when you visited the Defense Language Institute English Language Center (see http://www.dlielc.org/testing/index.html) in San Antonio and at TESOL. The reason I mention DLI is that it represents a language program with little leeway to fail. I spent 20 years from 1968-1988 with them. Their mission was/is to train allied and friendly military personnel worldwide in general and specialized English. Most students start ESL in their home countries with DLIELC personnel advising, and in some cases teaching, in these overseas military language centers. It may still be the largest language program in the world. Students after reaching certain levels go to Lackland Air Force Base in San Antonio, complete their general English, usually go through specialized terminology and then on to whatever training their country needs with the U.S. military. Students have run the full gamut from recruits and NCO's up to generals and admirals. This is a very tightly organized "well-packaged" language program with predictable training times and results.

Here's why I bring you this background. The rule of thumb while I was at DLI on class sizes was 8 students optimum and 10 maximum. This was rarely broken. However, once in a while a higher headquarters' bean counter would calculate that if a mere two or three students could be added to a class, voila, look at the money we would save! This money came both from foreign governments and Uncle Sam. DLI would argue and then try it, I believe several times over the years. However, in an organization that has a very effective testing system; both achievement and proficiency, it was soon noticed that the scores were going down, just enough so that they could prove that no money was being saved on teacher salaries and other expenses. I spent three years with the Royal Thai Navy for DLI as language training advisor and remember having to twist arms with the RTN Navy Education Department with the same problem. They wanted 15 in a class.

Here's what I suggest. I accept these numbers and know during my last eight years of teaching adult immigrants that my best classes have consistently been smaller. My students average around 9-11. If an organization is stuck with a low budget, make the hours of training per week lower, but keep the class sizes within 10 or so students. Fewer hours of really effective training are certainly better than large classes where the student attention level and collegiality are reduced. I remember Mary Finocchiaro saying years ago that she didn't care how many students were in her classroom, she would teach them! Unfortunately, most of us just aren't THAT dynamic.

Ted Klein


This is fascinating stuff. Jodi will probably have more questions about it than I do. (Jodi take note!) My major question is how far one can generalize from the DLIELC experience. After all, military personnel are a "captive audience" and it's part of their "job" to learn English via DLI. As a result, one would assume they are a very motivated bunch. Do you think this skews your finding about class size, instructional time and other variables, or not? Also, how large were the test differences when you "squeezed in" a few more students/class? Do you think the tests were good enough to make those differences meaningful?

Forrest Chisman


The person who started the DLIELC testing program, not too long after WW II, was Sydney Sako. Sydney was a mathematician with a strong interest in language. The people who developed the test items were well-educated language people. I'm a right brained/creative person, who was the second oldest person in his high school class to graduate, because of math. Therefore, I really don't understand the whole process. However, I recognize what works. The English Comprehension Level (ECL) test, was amazingly successful. Each form, and there were many, was normed against other forms and the numbers were very steady. It was a combination of around 120 listening and reading multiple choice items. If you tested students using different forms, the numbers were quite close. Each specialty required a specific score; 65 for admittance to on-the-job training, 90 for Staff College, 85 for pilots, etc. When I was in the program, the failure rate among students who went on into their specialty because of language problems, was very small...something like 5%. In the 1980's the federal Oral Proficiency Interview (OPI; see http://www.dlielc.org/testing/opi_test.html) was added for persons who needed specific speaking levels, such as air traffic controllers, pilots, Puerto Rican ROTC graduates to get their commissions and some others. I understand that the ECL is now done on a computer, but the principles remain basically the same. You can get up-to-date information on the current system at http://www.dlielc.org/testing/ecl_test.html Students who had received a good balance of training; listening, speaking, reading and writing skills and who performed well on the ECL test were generally also proficient orally.

There is a second option, the American Language Course Placement Test (ALCPT) (see http://www.dlielc.org/testing/ALCPT.html) which consists of expired ECL tests that DLIELC can sell to recognized educational institutions. Several years ago I tried to get it adapted for Adult Education here, rather than the current system. The effort was almost successful, as a lot of people liked it. It was much quicker and cheaper to administer than what happens now and doesn't depend on opinion. However, acceptance of change is always a problem in anything. I have an article about the ALCPT on my website at http://www.tedkleinesl.com/ESL/lang_tests.html The ALCPT does not test persons who can't read English, but there are alternative ways to place students in that situation.

Military personnel are much more reflective of the population as a whole than one might imagine. Don't count on the accuracy of my memory on all of the details, but if I recall there was something like a ten percent drop in average proficiency gain monthly, when two or three students were added to a class of ten. DLI Training Management had wheels to predict how long it would take to gain X many points and they were surprisingly accurate. Students were in class four hours a day and two more hours in a supervised lab, five days a week. Scores were used for placement as well as graduation. My personal experience of this whole English Language Training package was positive. There were always complaints, but the proof was in the pudding.

Now I live in the "do your own thing" world which is much more fun. However, I respect other ways of operating. The main difference is that students were in their national uniforms, had to be on time and the system was more "regimented." Occasionally persons from the academic community find teaching jobs there and some of them run away at the first opportunity! Others stay for a career. I managed...and have some fond memories. I have been in ESL since 1961, so DLIELC represents less than half of my career.

Ted Klein

Back to Transcripts Table of Contents

Please note: We do not control and cannot guarantee the relevance, timeliness, or accuracy of the materials provided by other agencies or organizations via links off-site, nor do we endorse other agencies or organizations, their views, products or services.