* Reducing Avoidable Errors

2002 unpublished

Aims:  Like most ESL teachers, I have been frustrated on numerous occasions by elementary errors which students repeat over and over again.  We revise the simple past and leave it on the board – affirmative, negative and interrogative – and then set the students to write a recount, only to find their writing liberally splattered with I was went, they going, he go…  Or we go through the wh questions and practise the spelling and pronunciation of [θ] and are confronted a few minutes later by almost every Slavonic speaker in the class happily writing wen and whit.  Or we list words with the digraph au and see because spelt with the vowels randomly rearranged yet again.

Oh, for a way to reduce these elementary avoidable errors!

Overview of idea:  After years of tearing my hair out trying to find some solution to these sorts of problems, I attended a session entitled Structured monitoring of second order errors presented by Eugene Mogilevski of Monash Univ. at the AFMLTA (Australian Federation of Modern Language Teachers Associations) Conference in Canberra, July 2001, where he outlined a strategy he had used successfully with his second year university French students.  He reported a 40% reduction in the sorts of errors I had been trying to eliminate in my students.  This inspired me to try to follow his example and to modify his LOTE technique to suit ESL students.

Procedure:  As recommended by Mogilevski, I spent a short time (less than one hour) discussing with the students the effects of their errors of native-English readers: that errors which are more or less inconspicuous in speech are quite confronting to a native speaker when they appear in writing; that the students therefore make themselves appear stupid in the eyes of their readers; that, since they were not stupid, they would not want to make themselves appear to be stupid.  I then went on to explain to them the procedures I intended to follow in an attempt to help them reduce their errors.  This consisted primarily in making them conscious of their frequent problems and recording their progress in eliminating the avoidable errors;  that is, those errors where they actually knew the rules and which they nevertheless repeated frequently through carelessness or inattention.

The first piece of writing they did in their usual manner.  At the end they counted the number of words they had written.  When I marked the work, I counted their avoidable errors, the errors where I knew they were well aware of the grammatical rules, and the spelling errors in common words they had seen and used many times before.  I converted this error count into a percentage of their word count and recorded that figure in my notes and on their marked work in red pen.

When I returned their work, I projected onto the white board a list of avoidable errors they had collectively made and the students themselves corrected the errors and explained what was wrong in each case.  The most common avoidable error initially was using the wrong tense (mainly present simple or continuous where the simple past was required) although this very elementary error dropped out very rapidly.

I repeated this procedure with each piece of writing they completed throughout the term and found they took a very active interest in the progress of their own scores and, in a number of cases, became quite competitive about their results.  At their prompting, I told the class who had scored the lowest proportion of errors each time.  In my marking I emphasised the improvements each student had achieved and gave a chocolate frog as a reward to each student who halved his/her error rate.  I also pointed out that students who achieved zero or almost zero errors on one piece of work could expect occasional reversals, since no-one can remain perfect all the time.  I did this because I was aware of the danger of pessimism creeping in for some very highly motivated students if improvement was not absolutely consistent.

Evaluation of trial:  My class was comprised of students who had been in Australia from 5 months to 13 years; about half of them came from Eastern Europe and the other half from South East Asia.  They were studying at level III in the Certificate in Spoken and Written English, Community Access focus.  There were 17 enrolments and, of these, three were absent for a significant proportion of the course due to illness and two others found employment before they finished the course.  Over eight weeks they completed four pieces of writing under test conditions with bilingual and Oxford Advanced Learner's dictionaries freely available.

The greatest reduction in errors came between the 1st and 2nd weeks.  Although there was some minor backsliding, the general trend was downwards as shown in the table below. 

Avoidable Errors

Student

Wk 1

Wk 2

Wk 3

Wk 4

Average

1

8.0

6.0

3.0

0.02

4.3

2

4.2

2.0

3.0

3.0

3.1

3

2.0

0.8

1.0

0.6

1.1

4

1.6

1.0

0.0

1.0

0.9

6

7.0

3.0

1.6

3.0

3.7

7

11.0

2.0

1.4

2.0

4.1

8

4.0

2.0

0.7

0.6

1.8

9

8.0

3.9

 

2.8

4.9

10

3.0

3.0

2.8

0.9

2.4

11

3.6

2.0

2.0

1.0

2.2

12

4.5

0.7

1.6

 

2.3

Class Av.

4.7

2.2

1.6

1.4

2.5

The gaps in the grid are where students were absent.

Follow-up:  With this very small sample over such a short period, the results of the research project have no statistical significance, of course.  However, they are interesting and possibly warrant more protracted investigation with a larger sample.

It would be interesting to be able to track these students at intervals after the project finished to discover whether the technique produced lasting results.  The psychological impact of the study appeared to lie in focussing the students' attention on a numerical value for their errors and in their perception of the shrinking list of errors which appeared on the whiteboard as the weeks passed

It is interesting that teaching theories over the past generation have decried marking errors in red pen and giving old-fashioned numerical marks for student work as damaging to student self-esteem.  In this small experiment, however, the reverse seems to have been the case: students were delighted to have some actual measure of their improvement and were greatly encouraged by this. 

However, some caution is necessary here: older teachers have all seen the negative effects of constantly poor scores on less able or less well motivated students.  The technique, to produce positive results, must give every student the probability of demonstrated improvement.  If the errors to be avoided are too complex and sophisticated for the students' level of language development, then the technique is likely to discourage, rather than encourage, improvement and heightened motivation.

On the whole, the experience was very positive for both teacher and students.  Despite the small-scale nature of the project, the technique shows promise, a conclusion backed up by vocal and very positive student feedback.



Comments