|
EPSA [1]-21This is one of the Main-Experiments dealing with Japanese. Unlike the other Japanese Main-Experiments discussed in the book (EPSA [33]-2, [33]-3, [33]-8, [33]-9, [33]-17, [33]-18), most of the informants in EPSA [1]-21 (and also in EPSA [1]-18, [1]-20, [1]33, and [1]-34) had background in linguistics. The number of the informants in the latter EPSA Experiments is also relatively small (6-9), significantly smaller than that in the former set (87-46). EPSA [1]-21 makes a pair with EPSA [1]-20. The only difference between these two EPSA Experiments is their Schema A. It is of the SOV form in EPSA [1]-20 while it is of the OSV form in EPSA [1]-21. Their LGs are:
EPSA [1]-20 and [1]-21 on the one hand and EPSA [1]-33 and [1]-34 on the other differ from each other only in the following respect: "N-cm" and "3-CL" (CL=classifier) are adjacent to each other in EPSA [1]-20 and [1]-21, but they are not in EPSA [1]-33 and [1]-34. In the result charts below, we only consider their SG1 of EPSA [1]-21 (and [1]-20, [1]-33 and [1]-34), just as in the CUP book. See Section 7.5.4. The "Raw Data" file, however, contains the informant judgments on the Examples of the excluded SGs. DesignExamplesResults[1]-21, SG1 only, Lexical-group-based Effects of informant classification on a *Schema-based predictionInformant list for [1]-21: The informant classification is based on "Familiar." We consider only the informants who are native speakers of Japanese and who are familiar with both "bound variables" and "wide-scope interpretation" as they are used in linguistic discussion. [1]-21, SG1 only, Lexical-based, based on the above informant list Informant list for [1]-21: The informant classification is based on the result of [10]-10 and [10]-11. We consider only those informants (i) who are native speakers of Japanese and familiar with both "bound variables" and "wide-scope interpretation" as they are used in linguistic discussion, (ii) whose %(Y) on B is 0% and whose reported %(Y) on A is 25% or higher both in [10]-10 and [10]-11. [1]-21, SG1 only, Lexical-based, based on the above informant list Raw data |