One thing you might try is asking them to decide which option is correct-e.g. "We need teacher/ a teacher to help us." Another possibility would be to specify what changes can be made when correcting incorrect sentences: for the questions on articles,you could allow the addition of words and not allow any morphemes-such as the plural -S morpheme;  with the questions on verb forms, you could allow morpheme changes but not the addition of words.A third possibility is to treat this test as a pilot and make a note of all the answers provided by students which do not conform to the test focus and try to assign a cause to each one. This would enable you to change the problematic sentences in the final version of the test. 

Another thing has occurred to me (apologies if you are already doing it): are you including any tests to see how your students use articles and verb morphemes in spontaneous language use?  As I see it, their ability to interpret if sentences are correct or not could be based on learning of theoretical grammatical knowledge as opposed to genuine acquisition of the items that are being tested.

Dear Yang Li,

Thank you very much for your answer.

My subjects are EFL learners who have been studying English only in formal setting for about 6 years now, as  part of public education. In fact, one of the objectives of the research is what you have mentioned, their acquisition of grammatical morpheme. But the errors they show were different than expected and they were also unsystematic. 

Dear Chris Turner,

I do appreciate your valuable notices.

They are really helpful. 

You are right about your final remark, my subjects learn English Grammar theoretically and they have no real English language experience. Do you mean that I should elaborate speaking-based task like interviews? Won't they use the same strategies?

Regarding assigning standards for answers, I understood that I should show them the erroneous words, because I didn't do that before. I just present ill-formed sentences and ask for rewriting them correctly.

I like the idea of considering it as a pilot test and I think it will work.

However, I am a bit skeptical about assign options, I think that  giving them hints may not reflect their actual knowledge of morphemes. 

Thank you very much

Hi Abdulaziz,

With regard to your reply: 

When students do grammatical tasks,  they are focused on the structures that they use. This will lead them to draw on their conscious knowledge of grammar "rules". When they are performing spoken tasks, they may also draw to some extent on such rules but will also show which language patterns they have already acquired and are thus able to use spontaneously. 

I find there is a huge difference between the errors  my students make when doing grammatical exercises and those they make when speaking spontaneously.

Thank you for your meaningful input in your example. I visualize that your students' basic grammar framework needs more practice and maturity. They need to have strong baseline of present (real life tasks and activities), real situation of what is happening now (present continuous) (V 'ing'), 'it is', 'it has', 'I am', 'I have', simple past, stress and intonation and morphemes (especially inflections). Please see my postings. I hope it helps!

re avoidance: The easiest might be to give them two tasks: one in which they can easily avoid certain constructions, one in which they cannot. An unconstrained text production task (Write X words on topic Y.) and a written translation (a short text or text passage) can both be embedded in a meaningful context and should give you the right data.

Over-generalization is more difficult. I would set up a stimulated recall scenario. Students do a suitable language-learning task. Immediately or very soon after the researcher goes through the task output with them. Asking questions for example on verb morphology and article use (referring to both accurate and inaccurate examples), such as why did you use this ending? why did you not use an article here? Both the text production and the stimulated recall are recorded, coded, and analyzed.

Dear Abdulaziz Assanosi,

I have also experienced something very similar to yours. I could get out of the situation by utilizing what is called a plausible interpretation. In point of fact, I tried a retroactive commentary technique in which the participants taking the grammatical judgment task were called for negotiating and interpreting the sources of errors they had produced on the test. The effectiveness of the approach has been supported by scholars like Pit Corder . Utilizing an emic perspectivization to fathom out the causes of language errors is rewarding. Wish you lots of luck with your research.

Best regards,

R. Biria

Thank you very much Chris Turner, Mathias Schulze, and Reza Biria for your valuable helpful replies. I really benefit from them and I do appreciate that.

 Dear Chris Tuner,

How are you

I am grateful for your valuable contribution to my question.

I checked the ALC, but I found that the target language for the learners is Arabic which is different form my students who are Arabian students study EFL. However, the other site is very helpful and I found the corpus by Randal and Groom (BUiD Arab Learner Corpus). I am checking it now and I plan to email the developers shortly.Thank You very much.

Yes, I was completely wrong about the arabic learner corpus. I misread the information completely. Sorry about that.

It's Ok. Your suggestion was very helpful, I found many corpora deal with grammatical errors by Arabic speakers. You opened the door for that!.

Thank you very much.

It looks like your students are coming up with discursively acceptable versions, but not the ones you are looking for, so the first thing I'd ask is this: is this for a research project or a classroom assessment task?  If it's for a classroom task, in your next lesson you might bring the notion of the "grammar of choice" - i.e. that there are more than one correct ways to say things, but the meanings and implications may differ.

If it's for a research project, you may want to add language to constrain their choices.  For example, giving "Now I am cook" gives them too many options (well, at least two), and they don't seem to be choosing the one you want to elicit.  You can try this:  "We are getting ready for a family dinner.  My sister is in the dinging room setting the table.  My mother is roasting chicken. And I am cook vegetables."  You may want to highlight the whole sentence you want them to rewrite.  Of course, this strategy gives them quite a few hints on what they are supposed to do (other present progressive verbs in the passage), but at least you'll know they are not avoiding the structure.

I hope this is helpful.

Hello Dear Olga Griswold,

I am doing this as a part of a pilot study for a research project investigating the accuracy order of grammatical morphemes of EFL learners whose Native language is Arabic.

Your suggestion are very helpful, I will try them next time. However, I wonder, Won't these direct hints affect the validity of the test? Bearing mind the low levels of the subjects, they may answer just by comparing with previous structures without true realization of the context. If so, this will affect the validity of the test, particularly when the research objective is to check the accuracy order which is considered by most researchers as a clue to acquisition order.

Thank you very much.

They might, since you are providing other present progressive forms.  However, some way of restricting context so that the most desirable choice from the point of view of communicating a certain meaning through a certain grammatical form may be needed anyway if you are to elicit the form (and, therefore, rule out the issue of avoidance).  You can try other ways of restricting the context through adding language so that students are pushed to use present progressive whether they know how to form it or not.  Maybe something along the lines of "There are two women in this picture.  One has a beatiful red dress.  The other is wear a very cute green blouse and jeans."  I'd suggest playing around with the context creating/restricting sentences and seeing what works.  I know designing instruments for eliciting data can be the biggest pain in the neck, but you want  yours to be solid. :-)  Try something similar for the indefinite article.  Maybe "There are two pears on the plate and apple, but only one." Etc.  

I guess there is always a bit of risk in providing more context to your research participants that restricts their grammatical choices.  But then real language is used in real context, and the context drives our grammatical choices. :-)  So, if your students are able to provide proper forms in proper contexts, you'll find something about their acquisition. :-)

Oh, and just Olga is fine. :-)  

Thank you very much, that is very helpful. 

Hi Abdulaziz,

In our lab, we have used an elicited production task where we show participants a picture and probe them to produce sentences using different verbal inflections, e.g., tense/agreement. For example: 'The girl likes to run every morning (in reference to the picture of a girl who's running). She does this every day. So what did she do yesterday? or What does she do everyday? or What is she doing now? You can give them examples of sentence frames you want them to use in response to the question to make sure they produce the targeted inflection, e.g., Yesterday she ......

I have successfully used this procedure with older children and bilingual adults. try it and let me know if it works.

Cheers,

Thembi.

 Hi Sithembinkosi,

It seems interesting. I will try it and inform you about the results.

Than you so much

Hi, I think that you have to limit your students' choices. Moreover, you can try to focus on a single grammar pattern or grammar rule and try and look for corpora about the same grammar structure or morphological aspect. Then you will be able to compare your data with the data about your students. To sum up, firstly you need to have consistent data with a low number of variables. Secondly you can collect data and compare them with other corpora on the same morphological problem. Good luck! 

Best regards