Abstract
Background. The English Grammar, Punctuation and Spelling (EGPS) Test is a statutory
assessment of writing convention skills in Year 6 in England. The relationship between the EGPS Test and children’s writing outcomes has not been examined empirically. This thesis aimed to establish whether writing convention skills assessed by the EGPS Test demonstrate a relationship with writing quality, accuracy, productivity, and complexity, and whether the EGPS Test can be used to identify struggling writers.
Method. A sample of N = 210 was drawn from English primary schools. Participants completed the EGPS 2018 Test, alongside tasks that assessed writing quality at word, sentence, and paragraph levels. Writing at the microstructural level was assessed by measures of productivity, accuracy, and complexity. In addition, pupils completed a measure of spelling ability, handwriting fluency, and reading comprehension. Assessments were completed as whole class task and administered by the teacher to adhere with COVID regulations at the time.
Results. The EGPS Test correlated strongly with writing quality. Domains of the EGPS Paper 1 loaded on a single factor, theorised as EGPS Writing Conventions (EGPS-WC). EGPS-WC correlated strongly with the composite Written Expression Subtest (WES) and paragraph-level writing, and moderately with sentence-level writing. Sentence-level writing was predicted by EGPS-WC performance; paragraph-level and overall quality (WES) of writing were predicted by EGPS-WC and spelling ability. Path analyses demonstrated that EGPS-WC contributed both directly and indirectly to overall writing quality. Multiple regression analyses identified that spelling and EGPS-WC predicted writing accuracy. Finally, The EGPS 2018 Test offered excellent predictive validity for struggling writers.
Conclusions. EGPS is theorised as a distal (indirect) measure of writing. A theoretical rationale for the role of EGPS-WC as part of the text generation process is proposed, emphasising the importance of freeing cognitive resources through the development of writing convention knowledge.
assessment of writing convention skills in Year 6 in England. The relationship between the EGPS Test and children’s writing outcomes has not been examined empirically. This thesis aimed to establish whether writing convention skills assessed by the EGPS Test demonstrate a relationship with writing quality, accuracy, productivity, and complexity, and whether the EGPS Test can be used to identify struggling writers.
Method. A sample of N = 210 was drawn from English primary schools. Participants completed the EGPS 2018 Test, alongside tasks that assessed writing quality at word, sentence, and paragraph levels. Writing at the microstructural level was assessed by measures of productivity, accuracy, and complexity. In addition, pupils completed a measure of spelling ability, handwriting fluency, and reading comprehension. Assessments were completed as whole class task and administered by the teacher to adhere with COVID regulations at the time.
Results. The EGPS Test correlated strongly with writing quality. Domains of the EGPS Paper 1 loaded on a single factor, theorised as EGPS Writing Conventions (EGPS-WC). EGPS-WC correlated strongly with the composite Written Expression Subtest (WES) and paragraph-level writing, and moderately with sentence-level writing. Sentence-level writing was predicted by EGPS-WC performance; paragraph-level and overall quality (WES) of writing were predicted by EGPS-WC and spelling ability. Path analyses demonstrated that EGPS-WC contributed both directly and indirectly to overall writing quality. Multiple regression analyses identified that spelling and EGPS-WC predicted writing accuracy. Finally, The EGPS 2018 Test offered excellent predictive validity for struggling writers.
Conclusions. EGPS is theorised as a distal (indirect) measure of writing. A theoretical rationale for the role of EGPS-WC as part of the text generation process is proposed, emphasising the importance of freeing cognitive resources through the development of writing convention knowledge.
| Original language | English |
|---|---|
| Type | Approved Doctoral Thesis |
| Media of output | Doctoral Thesis |
| Publisher | UCL Institute of Education, University College London |
| Number of pages | 197 |
| Publication status | Published - 2022 |