Korpus - gramatika - axiologie

Datum publikování: 6.2025

Autor: Místecký, Michal

Autor: Radková, Lucie

Autor: Stiborská, Žaneta

Autor: Hrubá, Darina

Abstrakt: The paper focuses on morphology-oriented quantitative linguistic analysis of the texts contained in didactic tests, which are used (1) as the Czech-language entrance exams, and (2) as part of the secondary school-leaving Czech-language exam (“maturita”). The studied corpus comprises 334 texts from 56 tests. The markers that are investigated are the grammatical cases, grammatical number, verb perfectivity, past and perfect participles, and prepositions. It has been found out that the test texts differ as to the frequencies of the genitive, accusative, past, and perfect participles. These differences always concern the maturita exams, which stand off as more complex than the entrance exam texts. Next, we perform correlation and cluster analyses. The correlation analysis has shown that there is some correlation between the markers, but it is not very strong. The two-step clustering procedure, making use of t-distributed stochastic neighbor embedding and the k-means analysis, has produced two groups of texts the distributions of which differ as to the type of exam, too.

Rozsah stran: s. 25–37

Status recenzování: recenzovaný článek

Licence: CC BY 4.0