Friday, March 19, 2010

Korea's robot English teachers won't go away.

"Tiro," or teaching robot, in Daejeon in 2007. I wonder if it will be any more effective teaching English conversation to 35 students than humans have so far been.

The story of South Korea soon using robots to teach English is one that won't go away. In January the Korea Times had an article, "Robots to Replace Native English Teachers," using the projections of a single economist as the basis for an ominous headline. I didn't touch it at the time because I was taking most of January off, but the news came up again a few weeks later when western media picked up on it. The Korea Times has another article today, looking at the results from a few trial runs.
Classes using robots developed for educational purposes have proven to be effective in enhancing English classes, the Ministry of Knowledge Economy said Thursday.

Students of English classes using robots as teaching assistants showed better learning achievements in speaking, as well as greater confidence and motivation, it said, citing a survey carried out by the Korea Education and Research Information Service (KERIS).

The eight-week project was conducted from late December last year at three elementary schools in Daejeon and Masan, South Gyeongsang Province, as part of an initiative to make South Korea a leading robot developer by 2018.

Using robots to teach English in Korea isn't news. You can find articles from 2005 talking about it, and even further back if you're looking at Japan. Foreign media, always eager to pick up on a "news of the weird" item out of East Asia, reported on it in February. The blog Gusts of Popular Feeling did a nice job last month looking at some of the robots in use today, and the social climate created by some in the media that might make robots a safer alternative to foreign barbarian teachers.

Today's Korea Times article gets closer to the reason these robots are employed deployed:
"Using teaching robots in classes is expected to raise the quality of public school education, thus leading to less dependence on the private education," said Kim Hong-joo, a ministry official.

Also, an early start in teaching robot projects will be helpful in leading the new global market as the nation aims to be one of the top three global leaders in this field by 2013, he added.

Decreasing spending on private education is something you see in the papers all the time---Koreans spent 1.12 million won per household per month in 2008, and the country spent three billion more last year---though it's not clear how robots in the classroom will improve English education or make parents not want to seek out an actual human being for instruction.

The second part seems to make more sense, and is in line with South Korea's goal to become a worldwide leader in practically everything. South Korea will use English-teaching robots because it wants to demonstrate it can. As I'll try and show below, native speaker English teachers are basically lame ducks anyway, so sacrificing a little quality for the next couple years isn't really a concern.

Getting back to the January article, I'm not sure if you can believe "a good number" of native speaker English teachers will be replaced by robots by 2018, necessarily, because projections for the use of NSETs change all the time. I recall reading articles years ago that said NSETS would be out by 2012, and 2013, and 2015, and 2016. Before the 2008 presidetial election here you had one candidate advocating English taught by NSETs, another who wanted more money spent on training Korean English teachers, and others who wanted NSETs phased out.

At first glance the progression shown in the January article confirms my suspicion that the NSET experiment is nearly over in public schools. But the source for the January Korea Times article, "an economist at the Hyundai Research Institute," isn't a guy in the Ministry of Education, isn't a guy in place to shape policy, and certainly isn't a guy whose forecast warrants a headline "Robots to Replace Native English Teachers," but this is the Korea Times. Nonetheless he says:
"Before such sophisticated English-speaking robots debut, teaching by native English speakers will be conducted by video-conferencing with teachers in their home countries," he said.

This is something we've seen happen the past couple years---I wrote about it in 2008 and again last week---on a small scale with students on islands and in remote counties who otherwise wouldn't have regular meetings with a native English speaker.

There are other signs that schools and policymakers are looking to move away from native speaker English teachers, just a few years after they were imported in large number. The government has been hiring thousands of Korean "lecturers" to teach practical and conversational English in public schools. The government has created "Teach English in English" certificates for Korean English teachers who can conduct an entire English class in English, a departure from today where most classes are grammar-based and have little to no target language use. And, the government is looking to replace the TOEFL with a domestic English exam that tests "practical English," rather than the inappropriately difficult material you find on standarized exams now (take a look at last year's college entrance exam).

I can't say native speaker English teachers are by and large efficient or effective, but I do know little preparation or thought has gone into their implementation, and little planning goes into how they're used in school, practically setting them up to fail. It's remarkable that all the interest in "practical English" comes at a time when NSETs are projected to be on their way out, because "practical English" and English conversation of course certainly play into the strengths of native English speakers, and right into the weaknesses of Korean English teachers. Currently NSETs are used in schools where English is taught entirely because of standardized exams, where the actual use of the language is an obstacle to the test material, where conversation classes are of no importance and receive no significant grade, and where native English speakers exist at best as a novelty and at worst as a nuisance. When foreign teachers arrive in school their principals and co-teachers often have no idea what to do with them, give them little guidance beyond "teach speaking," and have no goals for their use or ways to measure success or failure. You often have Korean English teachers who don't do their jobs as co-teachers, don't assist in lesson preparation, don't attend the mandatory English workshops, and don't even show up to class.

It doesn't really matter because there's no incentive or accountability, and this is something I wrote at length about in a December post about a government blacklist of "incompetent" foreign teachers. NSETs aren't given the opportunity to anonymously evaluate their Korean co-teachers---except on blogs or Facebook---so nobody really knows if Mr. Park doesn't show up for class, or Mrs. Kim doesn't participate, or Ms. Lee beats the students. It's up to these co-teachers to decide if NSETs have their contracts renewed, but you have to question the qualifications of evaluators who don't attend or participate in class, or don't know anything what NSETs are brought in to do. Teachers are required to attend training sessions in language and methodology, but since everyone passes regardless of ability, attendance, or effort, there is again no accountability. And, since all we hear about in the papers are that foreign teachers will be out by 20__, there's really no reason to change to accommodate a temporary intrusion.

None of this is new to readers of my site, but to first-time visitors I'll direct you to other posts where I've addressed the problems NSETs face:
* (12/3/2009) "Are native speakers part of English here? Your thoughts on the 2009 GETA International Conference."
* (12/2/2009) In the Korea Herald, writing about mandatory culture classes for foreign teachers.
* (6/26/2009) Korea Herald: Just what makes a teacher "qualified"?
* (6/15/2009) Not enough applicants for those "English Lecturer" jobs.
* (6/5/2009) Seoul wants English classes to be taught in English, will give TEE certs out.
* (5/13/2009) Korea Herald: The media bias against foreign teachers.
* (5/6/2009) 12% of native speaker teachers in Ulsan not retained.
* (5/1/2009) Korea Times: Foreign teachers wrongly portrayed in Korea.
* (4/7/2009) Korea Herald: Stop the scatter-shot approach to English.
* (12/30/2008) Half of foreign teachers leave after one year? GREAT! That's an article that should be brought up every now and again, because a MOE official in charge of native speaker English teachers says
``They are neither regular teachers nor lecturers who can conduct classes independently. They are `assistant teachers,' hence their teaching experience doesn't matter much,'' he said. ``Rather, it's better for students to have more new teachers so that they can meet various kinds of foreigners,'' he added.

* (12/10/2008): Poor guy.
* (11/24/2008): EPIK in the news some more.
* (11/21/2008): 4,000 "English Lecturers" coming in 2010.
* (11/14/2008): A must-read: an account of teaching English in South Korea in the sixties.
* (10/6/2008): More money going into English education next year.
* (9/11/2008): More English-Only classrooms, more gimmicks.
* (6/23/2008): Pronunciation matters.
* (11/28/2007) A reaction to Kang-Eun-hee's "Korean English Teachers."

Those posts all have links to other posts and articles, and the "English in the news" category has more information as well.

In that Gusts of Popular Feeling post last month he links to an interesting comment on another site:
I'm currently teaching in South Korea (and yes, there are always job openings... though less than usual, with the recession on). I teach at two public elementary schools, one of which is on the extreme outskirts of the city and only has 46 students. For some reason, this tiny school got an English robot called the Cybertalker, which uses voice recognition and some kind of face recognition to tailor pre-made conversations to students. The only time I've seen the thing turned on was in the frantic lead up to a school inspection, when my English classes were cancelled in favour of registering all the students in the system and trying to make it perform for the school board officials. Even with days of practice, the students couldn't make it respond - even the almost fluent teachers couldn't make it recognize their English. These are the crappiest teaching robots in existence. A Speak and Spell would be more useful.

Indeed when I first heard about the robot story and how they were cast by the Korea Times as a replacement to native speaker English teachers, I thought about something my former school tried.

In October 2008 I wrote about my school getting a brand new AMC-200, a piece of machinery that, as explained to me, works like what you'd find in a 노래방 (singing room). It came with dozens of books, on a wide variety of topics, and when you select the lesson you'd like to do, you punch in a particular number and it calls up a small video, or a recording of a dialogue, or a particular language activity, or a song, or a number of other little tasks.

I never saw it used, and matter of fact it wasn't even plugged in during the spring semester. I liked the use of such technology, and also the overuse of powerpoint and multimedia in the language classroom in Korea, as using a sledgehammer to kill a mosquito. There's no need for fancy displays and expensive equipment when students just need to learn and practice the basics, how to construct sentences and basic meaningful utterances.

South Korea teaches English exclusively toward standardized tests, yet achieves some of the lowest scores on the world on them (a function of every student in the country taking them, but still). Spoken English and "communicative competence" has never been a priority, and it shows. You'll never have improvement there without an overhaul of the goals and methods of English education, and whether you have a native English speaker, a Korean English teacher who can speak some English, or a robot with a funny voice, you'll still have the same results unless you change how English is taught and what it's taught toward. Korea talks about using robots in school just to demonstrate it can, unlike many other countries in the world. However there's nothing to indicate a fancy robot will achieve any more than this piece of equipment:


kushibo said...

"Tiros"? They call them Tiros now, but they will be called Engbots.

Stephannie said...

I say bring on the robots! Robots wont complain with late or missing paychecks, wont complain about substandard housing, wont complain about being assaulted, raped, or murdered. Robots will work overtime for free, gladly work through vacations doing desk service and wont expect the yearly 5manwon a month raise. AND Korea will be number one in use of robots in the classroom. Meanwhile, all foreign devils previously working in Korea will go to ____(insert local)___ and offer outragious tuition rates for illiterate (in English of course) korean students to come for home visits & one on one intensive emmersion lessons.
These vast hordes of freshly educated Korean students will also pick up novelty ideas like: equality, democracy protected by civil rights, racial tolerance, intolerance for corruption and a thirst for western style justice. They will then all return to Korea as enlightened becons and lead their nation into the modern era with progressive social change and political integrity.

Hey...why not? the robot fantasy is good enough for the masses to buy into... why not a social revolution led by foreign educated intelligencia? 'cause you know THAT's the real fear... we'll infect the lil ones with *IDEAS* that foster *indepentant cognition* HA

Andy said...

One word...


sonagi92 said...

Raised achievement in speaking in 8 weeks? I'd LOVE to see the assessment data.

The anecdote from the teacher about the robots not working was hilariously predictable. Gee, I wonder if the company that makes these robots is providing some incentives for all this glowing coverage.

BTW, off-topic request for Brian or any other NSET. I'd like to view the national English curriculum standards at the MEST website. At the ministry's website, I found a file called 2008년 개정 영어과 교육과정 on a page called 영어교육 질제고 및 격차해소 in 2010년 주요정책 - 초·중등교육 분야 on the front page. I couldn't open the file, probably because it was made using a Korean application. If anyone can either direct me to resources that I can read or is willing to convert the file and send it to me, I would be extremely appreciative.

Brian said...
This comment has been removed by the author.
Brian said...

sonagi I can read it b/c I downloaded a .hwp viewer. Looks like you can download the file here:

Hope that helps.

L Tron said...

When the robots start using the children's tears to power their death squads, they'll wish they'd just let the NSETs stay.

yehjee said...

Using ROBOTS to teach English???
Wow... I have to admit, that IS really rice-tarded.

Erik said...

Predictions for future headlines:

- Unqualified Robot English Teachers a Danger to Our Children

- Robot English Teachers Caught in Illegal Gambling and Sex Ring

- "Robot English Teacher Gave Me AIDS," says Korean Woman

Colin said...

Robots have been in the classroom for years. They're called students!

3gyupsal said...

It isn't any use to worry, or complain about reports in the Korea Times about what the government of Korea says about Native English Teaching. It is somewhat obvious that the government is using the Korea Times as a voice to express it's position. The message that I am getting from this article is:

"We didn't plan our budget well enough to pay for the NET's that we intended to hire a long time ago, so now we are trying to get rid of them in public schools yet still offer a viable alternative to hagwons."

The hagwon industry and its customers dictate the market, not the public schools or the government. Right now the government sees its self in competition with hagwons because it feels that it can help Korean citizens save money, when in fact, if the citizens choose to pay to send their kids to a hagwon that is their choice. Also if having an NET at a hagwon is a selling point for parents, then NETs in public schools jobs are safe because the government will continue to want to compete with the hagwons.

3gyupsal said...

It is also strange that nobody seems to have the time to come up with a comprehensive system for NETs since according to this marmot's hole post, Koreans work the most of any OECD country.

Really? So there are people in the government who work 120 hour weeks, and the best that they can come up with for guidance of NETs is to, "I don't know just teach about chapter five."

Brian said...

I posted about those stats in 2008, and I do think "productivity" is a loaded word, and rather ethnocentric. But, it's clear that Koreans work long but not necessarily hard.

Anyway, I think there are two big things going on:

(1) The rush to communicative competence wasn't thought through, and
(2) Korea has so much technology, and ambition for that technology, it doesn't know what to do with it all.

I would have to dig through my old papers and links, but I remember that national curriculum was implemented rather suddenly, and it was decided that communicative competence was going to be a priority . . . even though none of the teachers in place had any exposure to that coming up, were qualified to teach it, were able to speak English, or could even fit spoken English into their test-drive classes. So the quick, easy answer was bring in NSETs, with the idea being . . . um, teach speaking. Now, upon some reflection, they're seeing that didn't work out, and when you add in some hostility that's been brewing over the last decade, and an overall aversion to native speakers in the field of TESOL, you're seeing them trying to get NSETs out before they really had a chance to do anything.

I'm a low-tech guy, who prefers a whiteboard, flashcards, and worksheets to powerpoint, video, and the goddamn AMC-200, so I certainly think robots are over the top, and I think this is about proving to the world that it can build and deploy tons of robots, not that it's necessarily a smart move.

Really, if you can't get a workforce of Korean English teachers who speak enough English to teach their damn class, who don't have enough classroom management skills to handle a group of students, and who don't have work ethic enough to try, then basically you're fucked, and you'll have the same problems if you use white people, brown people, AMC-200s, and robots.

Brian said...

You'd have to go through and interview Korean English teachers about why they became English teachers. As we saw in the post about Bill Costello last week, many people choose to be teachers b/c it's a stable job. Hell, I'd become a teacher, too, if it weren't such a pain in the ass in the US. But why did they choose to be English teachers?

I've met teachers committed to improving themselves and helping their students, but I've met plenty who just floated along (again, no accountability). Say what you want about shitty NSETs, and there certainly are a fair share, but don't forget who's doing the lion's share of teaching.

rostafnk said...

I LOLed when I began to read the comments. My co-teacher came over to ask me what was funny. How can I say that despite failure or success this motion is so purely Korean it's laughable? Also, why does he not agree with me that a robot would shoulder more of the work load than me [who clicks between sections of the elementary cdrom]? Perhaps he's being nice as I am not the most worthless teacher in our school by my own volition. Cheers to Stephanie; summed up what I was thinking.

Mike said...

I wonder if they'd stare at the robot as much as they stare at me... or if it would be less novel.

Chris in South Korea said...

Getting back to the post...

Of course robots won't go away. Korea has some MAJOR electronics companies and know-how - God and Goddesses forbid some OTHER country make their robot English teachers first. I'm sure there's more than a few people in said companies working to make robot teachers a reality. Of course, it's worth noting they're only as good as their programming - guess who did that? That's right - the kids that went to human teachers are now ADULTS, baby!

In all seriousness, this will blow over - as it always has. If they can't afford to pay their teachers what they're worth, what makes you think they'll be able to afford one of these things for the school? Native English teachers need only worry about getting paid and not getting screwed by whatever school currently employs them.

Puffin Watch said...

There are always going to be rich people in Korea and more wealth you have more advantages you have. Korea seems to, at time, try to legislate this unavoidable reality away by trying to, say, limit private education fees.

The problem lies in Korean culture. If you don't get into SKY, you're toast. If you don't do well on a test, you don't get into SKY. Merit takes a back seat to where you went to school, who you know, etc.

So all parents try to spend the maximum to make sure their kids get into SKY. Some parent's maximum is a lot larger than another's maximum.

And you need your kids to do well because your kids tend to be your safety net in your old age. That ramps up the mania to make sure you raise high earners.

So, a better social safety net and government/major corporations that count talent over pedigree are true areas to improve.

3gyupsal said...

@ Puffin Watch "If you don't get into a sky school your toast." I hope you were referring to the cultural feeling in this country and not the reality of the situation.

Middle schools are set up to either send students to a college track high school or a technical high school.

My wife's brother went to a technical high school then a technical college. He worked at LG for many years until Hyundai stole him away and made him an executive.

I have heard of other cases where kids go to mechanical high schools then graduate, collect certifications, and then open their own shops. You are right, the college track culture is often myopic, but not getting into a sky school doesn't necessarily make someone "Toast," people just don't realize that a doctor who didn't graduate from SNU med school is also called doctor nonetheless.

3gyupsal said...

The same problem will exist with robot teachers as they do with NETs, the Koreans won't know what to do with them, and no one will try to fix the system, until Korean education ministers learn about modern management, they will continue to make the same mistakes. For all of the problems with the U.S. Educational system, I think that it is interesting how at the last two EPIK training courses that I attended, two guys at the different places cited studies on co-teaching that came from U.S. schools, since the only scholarly work on co-teaching came from the U.S. Due to statistical studies done on "No Child Left Behind." EPIK on the other had has been around for over ten years, and there seemingly haven't been any studies done on it, and no attempts to make it better other than the cosmetic idea of trying not to hire drug addicts.

Peter said...

Very interesting post, Brian.

Using robots and other fancy gadgets, hiring so-called "Teaching English in English" lecturers, attempting to switch the focus of English classes to practical English ... these are all classic cases of trying to "cure a disease" by addressing only the symptoms.

Korean parents don't spend tons of money on hagwons because they're dissatisfied with the quality of public school education. They do it because hagwons serve a cultural purpose of their own, one that's quite distinct from the cultural purpose that public schools serve.

Korean students don't fail to learn practical, conversational English because of specific issues with the English curriculum or with teaching practices. Korean culture simply doesn't require or motivate Koreans to learn practical, conversational English. The Korean government seems loathe to admit this, so they make a lot of noise about "improving the quality" of English education, and engage in stunts like this talk of robot teachers. But really, the public English curriculum in Korea isn't "failing" to teach conversational English competence, because that's not what it's meant to be doing in the first place.

No matter what specific policy changes are made, what's really at issue here are fundamental aspects of Korean culture, which are probably not going to change any time soon.

Justin Kraus said...

Certainly Korean English education curriculum is not designed to develop student's conversational ability. Nor will Robots help. We can certainly argue that this is a problem. But I would caution us not to underestimate the counter argument which might go something like this. First Koreans may simply value an ability to read English over an ability to speak it because they figure (probably correctly) that being able to read English will be of greater practical value in their life than speaking because few will ever leave the country or develop relationships with foreigners in Korea. One the other hand reading English will allow them to experience, in the comfort of their homes, a larger world, particularly through the internet.
Second lets imagine that the Korean curriculum was completely redesigned to focus primarily on improving Korean children's converstion skills. I am still quite skeptical that they would improve all that measurably. Becoming proficient in a language simply takes A LOT of time, more than a public school, no matter how the curriculum is designed, can give. Does anyone know of any country where English is taught in public schools and in which the majority of students reach a conversational level of fluency?

Mike said...

India, for one. The Philippines. Germany, The Netherlands, Denmark, Sweden... pretty much all Northern Europeans graduate high school fluent (or at minimum conversational) in English. As do a lot of wealthy people from heaps and heaps of places. The reason why is simple: practicality and general ability in the population. Korea will get both of those things in time...

Justin Kraus said...


Unfortunately those countries aren't valid comparisons.

In the Phillipines all education I believe from middle school onwards (perhaps even elementary school?) is conducted entirely in English. This only proves my point that, everything else being equal, A LOT of time is needed to attain fluency in a language and that and 1 or two a day in a classroom, no matter what the curriculum, isn't going to cut it.

India also conducts all its education in English, so again, not a valid comparison.

As for Northern Europe its simply untrue that nearly all high school graduates are at minimum conversational. I have been there and although certainly the prevalence of English competency is higher than in Korea, it is nowhere near "pretty much all." The other thing about Northern European countries, especially places like the Netherlands and Denmark is that they have long histories of multi-lingualism which makes acquiring new languages MUCH easier. Korea does not.
So again I'd like to hear of a country that, without turning its schools into English-only zones (such as is done in the Phillipines and India) produces a high percentage of conversational competent English speakers.

Puffin Watch said...

@ Puffin Watch "If you don't get into a sky school your toast." I hope you were referring to the cultural feeling in this country and not the reality of the situation.

Of course the last president of Korea was a self taught lawyer. I'm not even sure if he graduated from a university, certainly not a top one. So, right. Even in Korea, arguably, merit can win out.

Of course, the president himself had it in for SKY, feeling the SKY kept too many doors closed for non SKY grads.

So you're right. My reading of Korean society is most parents view it's the SKY-way or the highway.

There are, of course, many anecdotes of self made men in Korea. But given the huge amount of resume fraud we saw a few years ago -- Korean teachers lying about attending SNU etc -- it sure seems to me that a good general rule is if you go to a lesser university it's vastly harder to get onto certain career paths.

One can match anecdote for anecdote but I guess we'll have to leave it to sociologists to decide how much more truth than perception lies behind my claim.

LadyE said...

I'm curious to know who will be chosen to set up the robots...NETS or KETS? :)

jay said...

nah, they can talk back and its just the same as some K-teacher lecturing for a hour...

great, let the parents who know will have to pay more for real teachers....

you know, they!@

Aaron D'Albey said...
This comment has been removed by the author.
celiph said...
This comment has been removed by a blog administrator.