The primary aim of this study is to determine whether it is possible to standardize written feedback on L2 student writing for use in a computerised marking environment. It forms part of a bigger project aimed at enhancing the feedback process as a whole The study attempts to establish "best practice" with regards to feedback on writing, by establishing from the literature what works and what should be avoided. Also, an empirical study was launched to establish what lecturers focus on and what marking techniques they use. A set of randomly selected essays from the Tswana Learner English Corpus and the Afrikaans Learner English Corpus were sent to the English departments of different tertiary institutions across the country. The essays were marked by the English lecturers at the relevant institutions. The conclusion was that lecturers typically focus on surface structures, and use ineffective marking techniques. The best practice (and data from the empirical study) was then used to create a set of standardised feedback comments (tag set) that can be used in a specially programmed software package in which students submit their texts electronically. Lecturers can then mark the student essays on the computer, hopefully speeding up the process, while at the same time giving much more detailed feedback. In later stages of the bigger project, students will get individualized exercises based on the feedback, and there are experiments currently being run to try and automate certain pans of the marking process in order to take some strain off the lecturers when marking. The immense archiving abilities of the computer will also be utilized in order to create opportunities for longitudinal studies. The effectiveness of the feedback tag set was tested in comparison to the marking techniques used by the lecturers in the empirical study and a self-correcting exercise. The conclusion was that the feedback tag set is more effective than the other two techniques. but students seem to perform weak overall when it gets to the revision of cohesive devices and supporting arguments. I argue that students are not used to revising these features, since lecturers seldom (if ever) comment on the structural elements of texts. However, the experiment proves that standardization of written feedback is possible to an extent. The implications of the findings are discussed, and recommendations for further research are made. / Thesis (M.A. (English))--North-West University, Potchefstroom Campus, 2006

Thesis (M.A. (English))--North-West University, Potchefstroom Campus, 2006.

The primary aim of this study is to determine whether it is possible to standardize written feedback on L2 student writing for use in a computerised marking environment. It forms part of a bigger project aimed at enhancing the feedback process as a whole The study attempts to establish "best practice" with regards to feedback on writing, by establishing from the literature what works and what should be avoided. Also, an empirical study was launched to establish what lecturers focus on and what marking techniques they use. A set of randomly selected essays from the Tswana Learner English Corpus and the Afrikaans Learner English Corpus were sent to the English departments of different tertiary institutions across the country. The essays were marked by the English lecturers at the relevant institutions. The conclusion was that lecturers typically focus on surface structures, and use ineffective marking techniques. The best practice (and data from the empirical study) was then used to create a set of standardised feedback comments (tag set) that can be used in a specially programmed software package in which students submit their texts electronically. Lecturers can then mark the student essays on the computer, hopefully speeding up the process, while at the same time giving much more detailed feedback. In later stages of the bigger project, students will get individualized exercises based on the feedback, and there are experiments currently being run to try and automate certain pans of the marking process in order to take some strain off the lecturers when marking. The immense archiving abilities of the computer will also be utilized in order to create opportunities for longitudinal studies. The effectiveness of the feedback tag set was tested in comparison to the marking techniques used by the lecturers in the empirical study and a self-correcting exercise. The conclusion was that the feedback tag set is more effective than the other two techniques. but students seem to perform weak overall when it gets to the revision of cohesive devices and supporting arguments. I argue that students are not used to revising these features, since lecturers seldom (if ever) comment on the structural elements of texts. However, the experiment proves that standardization of written feedback is possible to an extent. The implications of the findings are discussed, and recommendations for further research are made. / Thesis (M.A. (English))--North-West University, Potchefstroom Campus, 2006

Kristýna Komárková ABSTRACT The diploma thesis deals with dealing with errors while teaching French as a foreign language with regard to speaking. The first part of the thesis contains the definition of the term error, its conception within modern teaching methods, classification of errors and most common causes of their occurrence. Corrective strategies analysis and particularities of dealing with errors within spoken language can also be found within the theoretical part of the thesis. Those theoretical findings are then applied in the practical part of the thesis which contains two case studies that are subject of the analysis of the corrective strategies used in French language classes at primary school. The practical part continues with a survey conducted between both teachers and pupils. The questionnaire is designed in order that it is possible to compare the views of pupils and teachers on the issue of error correction in the process of teaching and learning foreign languages. In particular, teachers are asked about the procedure and techniques they apply while they are dealing with pupils' errors, while questions for pupils focus more on their feelings that the error and error correction arise in themselves. The aim of the thesis was to compare teachers' and pupils' opinions on dealing with...

The diploma thesis deals with the position of error in the teaching of mathematics in the awareness of pupils and teachers. It also focuses on the responses of pupils and teachers to errors. Main attention is paid to mathematical errors. The thesis consists of two parts, theoretical and practical. The theoretical part contains an overview of selected error views. It contains a look at the development and changes of views on the concept of "error", the different concepts of error classification and pedagogical and psychological views on the term "error". The practical part is based on the research carried out at the 2nd level of two primary schools in Prague and Neratovice. Pupils filled in a questionnaire and a guided interview was conducted with teachers. KEYWORDS work with error, error classification, causes of errors, learning without errors, learning with error

Transformers have shown remarkable results in the domains of Natural Language Processing and Computer Vision. This naturally raises the question whether the success could be replicated in other domains. However, due to Transformers being inherently data hungry and sensitive to weight initialization, applying the Transformer to new domains is quite a challenging task. Previously, the data demands have been met using large scale supervised or self-supervised pre-training on a similar task before supervised fine-tuning on a target down stream task. We show that Transformers are applicable for the task of multi-label error classification of trace data, and that masked data modelling based self-supervised learning methods can be used to leverage unlabelled data to increase performance compared to a baseline supervised learning approach. / Transformers har visat upp anmärkningsvärda resultat inom områdena Natural Language Processing och Computer Vision. Detta väcker naturligtvis frågan om dessa framgångar kan upprepas inom andra områden. På grund av att transformatorer i sig är datahungriga och känsliga för initialisering av vikt är det dock en utmaning att tillämpa transformatorn på nya områden. Tidigare har datakraven tillgodosetts med hjälp av storskalig övervakad eller självövervakad förträning på en liknande uppgift före övervakad finjustering på en måluppgift i efterföljande led. Vi visar att transformatorer kan användas för klassificering av spårdata med flera etiketter och att metoder för självövervakad inlärning som bygger på modellering av maskerade data kan användas för att utnyttja omärkta data för att öka prestandan jämfört med en grundläggande övervakad inlärningsmetod.