More than a century of research shows the workbook exercises and
multiple-choice response items used in traditional grammar practice
does nothing — zero, zip, zilch — to improve students'
writing skills.
Even if poorly-designed studies are set aside, the bulk of the
research says traditional grammar teaching has no value in
improving students' writing.
Yet many teachers, particularly in private schools and homeschools,
not only use traditional grammar practice, but appear to turn out
a higher proportion of competent writers than the anti-grammar group.
What's going on here?
Reasons for negative results
The anti-grammar group assumes the only possible explanation
for the research results is that formal grammar instruction via
the usual workbooks and grammar exercises is worthless for writers.
Several other explanations are possible. For example:
a. The grammar might be taught poorly.
b. Teachers' grammar instruction methods might be inappropriate
for their students.
c. Students might lack preparation to benefit from the
instruction.
d. Students may not learn the material well.
e. Students may lack opportunities to write, which
makes the writing they do look shabby.
In other words, the problem could be lousy teaching rather
than any problem with formal grammar practice per se.
Reasons for positive results
The pro-grammar group have assumed the only possible explanation
for their success is that the exercises improve student writing.
However, other explanations could account for the success of the
grammar-teaching group. For example:
a. Those teachers might be more thorough in their teaching.
b. Those teachers' instructional methods may be appropriate
to grammar and/or to the students.
c. Those teachers may be teaching grammar when students
are ready to benefit from the instruction.
d. Those teachers may attribute the writing improvement to
grammar instruction when the improvement is produced by some
other learning experience(s).
Another possibility: a combination
It is more likely that the value of grammar instruction and of
traditional methods of grammar practice depends on the interaction
of several factors rather than on a single factor.
The research I've read and my own observation leads me to the
following seven conclusions:
1. The items on traditional grammar tests typically
deal with rules that are easy to evaluate rather than
rules that writers routinely use in their writing.
2. Grammar instruction does not improve students' ability to
identify, gather and organize content, which is the fundamental
task of writing.
3. Students who are native English speakers master the
fundamental structures of English grammar without formal teaching.
4. The first task of formal grammar instruction is to
teach terminology for discussing language, starting
with terms for the grammar students learned unconsciously by
hearing language. Such teaching can be done prior to writing
instruction.
5. The timing of formal grammar instruction is important.
For students who haven't learned how to write competently,
concurrent formal grammar instruction hinders development of
writing skill by putting too much stress on rules they
don't understand.
6. Once students become competent writers, traditional
grammar study may help them become fluent writers, but
other methods may work just as well.
7. The best way to improve the grammatical correctness of written
work by competent writers is for teachers to flag one specific
type of error in their work and have them study that grammatical
issue until they master it in their own written work.
Applying these seven conclusions to teaching is probably less
difficult than traditional grammar teaching methods because teachers
have less material to teach.
What makes this approach unpleasant is that it takes time.
Instead of an English teacher approach, you something closer to
clarinet teacher approach: a 30-minute lesson followed by 30 minutes
of practice six days a week.
