On Mon, 12 Jan 1998, Martin Nystrand wrote:
> In our ninth-grade study of 11,000+ questions in 54 English classes, full=
y a
> third involved evaluation in the following pattern: teacher asked a test
> question (a question with a prespecified answer), a student answered, the
> question elicited a report, and there was no uptake (incorporation of
> previous answer into a subsequent question). In fact, this pattern was s=
o
> common that we defined such questions as "normal" teacher questions (in o=
ur
> program, we treated it as the default for features of teacher questions).
>=20
> One particular kind of follow up we did study was what we called "high-le=
vel
> evaluation." Teachers often follow up student responses by elaborating
> important implications they see. Teachers sometimes turn some of these
> elaborations into didactic or instructive elucidations=97 little set piec=
es=97of
> points in a prescripted lesson plan=97essentially coverage of important p=
oints
> students should not miss. Others are more serious explorations of lines =
of
> inquiry opened up by students. When the latter occured, we coded teacher
> evaluation as high-level: the teacher noted the importance of a student's
> response in shaping a new understanding, and the course of interactions
> changed some because of what the student had said. That is, we treated
> evaluation as high level when a student contributed something new to the
> discussion that modified the topic in some way, and was so acknowledged b=
y
> the teacher. Specifically, high-level evaluation consists of two parts:
>=20
> =091. The teacher's certification of the response and
> =092. The teacher's incorporation of the response into the discourse of t=
he
> class usually in the form of either an elaboration (or commentary) or a
> follow-up question.
>=20
> For level of evaluation to be high in our coding, the evaluation had to b=
e
> more than "Good," "Good idea," or a mere repeat of the student's answer.
> The teacher had to push the student's contribution further, validating it=
in
> such a way that it affected the subsequent course of the discussion. Whe=
n a
> teacher's evaluation is high level, the student really "gets the floor."
> Hence, high-level evaluation, like authentic questions, directly affects =
the
> dialogicality of teacher-student interaction.
>=20
> On average, only 1% of the classes we observed in our ninth-grade study
> involved high-level evaluation. This compared with 3% in eighth-grade
> English classes.
>=20
> Martin Nystrand
> Professor, Department of English (608 263-3820)
> Editor, Written Communication (608 263-4512)
> Director, Center on English Learning and Achievement (CELA)
> Wisconsin Center for Education Research
> 685 Education Sciences
> 1025 West Johnson Street
> Madison WI 53706
> 608 263-0563 voice
> 608 263-6448 fax
>=20
>=20