期刊名称：Journal of Educational, Cultural and Psychological Studies (ECPS Journal)
印刷版ISSN：2037-7924
出版年度：2014
期号：10
页码：463-483
DOI：10.7358/ecps-2014-010-asqu
语种：Italian
出版社：LED Edizioni Universitarie
摘要：Definition of a Coding Procedure of Open-ended Questions Based on the Models of International Studies The paper presents an in-depth study within the «Problem-solving and geographical skills» project funded by Sapienza University of Rome in 2011. The main research instrument consists of open-ended items which was used as a basis for designing a coding procedure of student responses in order to maximise coding and coder control and reliability. Bearing in mind that when analyzing open answers, the margins of the assessor’s subjective interpretation are certainly greater than with classic structured items, the adoption of strict encoding and coding control procedures allows us to steer this type of semi-structured questions toward objectivity, thereby significantly reducing errors of interpretation. The main reference is the procedure used in the OECD-PISA for openended questions, but also the assessment of written tests established in the IEA study on «Written composition». The aim is to improve the quality of the dataset of the study, with the least possible burden on resources. The results confirmed the effectiveness of the procedure in terms of reliability of coding, with the estimation of a low-level error. It was also possible to provide timely feedback to each coder, thereby enabling an improvement in coding ability.
其他摘要：Definition of a Coding Procedure of Open-ended Questions Based on the Models of International Studies The paper presents an in-depth study within the «Problem-solving and geographical skills» project funded by Sapienza University of Rome in 2011. The main research instrument consists of open-ended items which was used as a basis for designing a coding procedure of student responses in order to maximise coding and coder control and reliability. Bearing in mind that when analyzing open answers, the margins of the assessor’s subjective interpretation are certainly greater than with classic structured items, the adoption of strict encoding and coding control procedures allows us to steer this type of semi-structured questions toward objectivity, thereby significantly reducing errors of interpretation. The main reference is the procedure used in the OECD-PISA for openended questions, but also the assessment of written tests established in the IEA study on «Written composition». The aim is to improve the quality of the dataset of the study, with the least possible burden on resources. The results confirmed the effectiveness of the procedure in terms of reliability of coding, with the estimation of a low-level error. It was also possible to provide timely feedback to each coder, thereby enabling an improvement in coding ability.
