<!doctype html public "-//W3C//DTD W3 HTML//EN">
<html><head><style type="text/css"><!--
blockquote, dl, ul, ol, li { margin-top: 0 ; margin-bottom: 0 }
--></style><title>Re: research methodology</title></head><body>
<div>arnold,</div>
<div><br></div>
<div>Me too. 4 groups: Linguists, nonlinguists X native speakers,
non-native speakers (all living abroad). They all rated the same
"questionable" sentences taken from the big Quirk et al.
grammar.</div>
<div><br></div>
<div><font face="Times" size="+2" color="#000000">Dennis R. Preston.
1975. Linguists versus non-linguists and native speakers versus
non-native speakers. Biuletyn Fonograficzny 16:5-18</font></div>
<div><br></div>
<div>dInIs</div>
<div><br></div>
<div><br></div>
<blockquote type="cite" cite>Arnold:<br>
<br>
I am now in Jamaica, far from the paper that
I would like to remind you of. Haj Ross published a paper on a related
topic, focusing on grammaticality judgments, about a uniform list of
sentences he gave naive and linguist informants, native and non-native
speakers among the linguists, non-natives living in the USA and
non-natives living abroad. His agenda was not exactly what you have
conceived but comes close. I suppose that data from Jim McCawley's The
Linguistic Flea Circus would be useful in this particular case.<br>
<br>
I am also using this opportunity to test
whether I can post on this list without having to subscribe again. I
am having my email to the University of Chicago rerouted to
UWIMonaNet.<br>
<br>
Sali.<br>
<br>
----- Original Message -----<br>
From: "Arnold Zwicky" <zwicky@CSLI.STANFORD.EDU><br>
To: <ADS-L@LISTSERV.UGA.EDU><br>
Sent: Monday, August 06, 2001 3:38 PM<br>
Subject: research methodology<br>
<br>
<br>
| i taught a seminar this spring quarter on syntactic variation,<br>
| in which almost all the students used standard corpora or web<br>
| searches to find examples of particular usages. this makes
a<br>
| great first step, but after that the students began to piece out<br>
| pictures of clusters of idiolects, differing in many different<br>
| details, and for this purpose, corpora alone aren't particularly<br>
| useful, since you need to know what *isn't* natural for specific<br>
| speakers; at some point, you have to collect judgments from<br>
| individuals.<br>
|<br>
| some of the students are pursuing these projects further, hoping<br>
| to do further work with specific people from the web searches<br>
| (or from adventitious collection of examples on the fly).
has<br>
| anyone explored this two-step methodology (finding potential<br>
| informants in the first step of data collection, which nets
people<br>
| who have some feature, then exploring the limits of that feature<br>
| through further data collection, interviews, judgment tasks,
etc.)?<br>
|<br>
| (the amount of systematic and structured variation among
individuals<br>
| on the details of specific constructions is quite impressive, by<br>
| the way.)<br>
|<br>
| arnold (zwicky@csli.stanford.edu)<br>
|</blockquote>
<div><br></div>
<div>-- <br>
Dennis R. Preston<br>
Department of Linguistics and Languages<br>
Michigan State University<br>
East Lansing MI 48824-1027 USA<br>
preston@pilot.msu.edu<br>
Office: (517)353-0740<br>
Fax: (517)432-2736</div>
</body>
</html>