LingQ Certificate of Achievement

Open Forum
1

The subject of a LingQ certificate for various languages has been raised on another forum. This is something we have been discussing internally here at LingQ. We are not yet in a position to implement this but we can start the discussion to get some ideas. I look forward to input from everyone and in any language. Here are some initial ideas.

There are two kinds of certificates or recognition possible.

  1. Recognition for activity. (for personal use only)

We can acknowledge people who meet certain milestones at LingQ in a given language.
Total LingQs created
Known words
Hours listened
Words of reading
etc.

We can have individual electronic certificates or a composite index, and recognize these people in the community and even issue certificates. This is relatively easy to do but is not of much use in the job market.

  1. Certificate based on testing (for third party use)

Here we could have a level that would be a number between say 100 and 800 or something. It would consist of the following elements weighted in some way.

  1. A test of vocabulary where the learner’s known words are tested using what is known as a cloze test. (cloze procedure: a test for diagnosing reading ability; words are deleted from a prose passage and the reader is required to fill in the blanks from a drop down list).

We are going to implement this kind of test in any case for our members to practice. This is on our to do list.

I see this as a pass/fail test, on 25 to 100 words at at time, where a high percentage correct answers, at least 95%, and for at least 10% of the total of the learner’s “known words” will confirm that the learner knows those words. A confirmed known words total is worth more than just reaching the milestone as is the case now.

  1. Each confirmed “known words” will correspond to a potential skill level in the language. The next step is to look at comprehension, speaking and writing.

We are going to introduce, at some point in the future, some form of recorded oral interview and questions. These will modeled after the questions on leading tests like TOEFL etc… There will be listening comprehension tests and question and answer tests, and oral essay questions. These will be recorded and filed in the learner’s portfolio with time constraints built in.

Written assignments, as now, will be retained n the learner’s files.

The learner should be able to select the best 5 or best 10 from among these written and oral submissions or test for display in his or her Portfolio.

  1. If a member wants an evaluation, he or she will ask for it, and will have to pay points. A panel of 3 tutors will review the learners portfolio, and then schedule an oral exam on skype between the panel and the learner. The learner will be evaluated on his or her portfolio and the oral exam.

The evaluation will therefore cover

Words known
Writing submissions from the Portfolio (the original uncorrected versions)
Recorded oral submissions from the Portfolio
The final Oral exam.

2

I think the Certificate based on testing is a good idea. I would do all this tests just for myself, I don’t need it in evidence. I also suppose it could increase the attractiveness of LingQ

3

Yeah the tests of the caliber you’re talking talking about will be absolutely necessary for Lingq to sort jump the hurdle between the 'people who are interested in learning languages" to the “people who must learn languages” crowd. It’s a good to focus on testing and the detailed plans you have sound like you’re heading in the right direction.

My only question is who will be the ‘native speaker’ panel for Esperanto? :wink:

4

Maybe we can convince George Soros

But seriously, standardised test construction is a challenge. Generally experienced teachers put together 100’s of questions of various types and field test them over a period of months among demographically relevant groups to find out which questions are the most valid and reliable. It is perhaps the only part of language teaching that can be actually called a science.

5

Perhaps LingQ can gather data about the correlation of LingQ activity and established test (TOEFL etc) scores and see if any patterns come up. Maybe offer those who have taken test X free memberships in exchange for access to their test scores and a contract to perform a specified amount of LingQ work over a period of months. I know it sounds backwards but getting current LingQ people to do established standardised tests would more problematic logistically, and financially.

6

We will develop this gradually. I think that in member profile we should show the member’s level in LingQ and ask members to list their results in other well known tests. This would give us a sense of the relationship between these tests and LingQ’s levels.I do not think we need to compete with established tests, but develop our own approach and let it slowly gain acceptance.