| A literary debate has been ongoing from the early 1900s regarding not only the place of grammar instruction in the classroom but even of the veracity of a grammatical standard, such as Standard American English, by which grammar skills may be measured. Very little empirical research has been attempted to compare the effectiveness of teaching methods because grammar assessment has been given even less attention. Therefore, to address this gap in the literature and to explore whether objective testing is suitable for assessing grammar skills, a 51-item measure was constructed to test the use of internal punctuation (commas, semicolons, and colons) and identification of syntactical structures (phrases and clauses). Rasch analysis found the measure as a whole possibly supporting a second dimension; therefore, the measure was analyzed as two scales: (a) a 34-item test of internal punctuation use and (b) a 16-item identification of syntactic structures. Both scales were found to be sufficiently unidimensional and reliable. In addition, scalar invariance of both was determined through DIF analysis by educational level. Validity evidence was obtained through a series of correlations with survey items assessing self-confidence and knowledge of the constructs tested in each scale. With the promising results of this endeavor in that objective testing can be effective, perhaps the debate may inspire researchers and educators alike to consider formal instruction of grammar in the context of a standard. |
