Purpose: Studies of reading comprehension and syntactic knowledge make use of syntactic knowledge tests that vary in both format and items, putting into question whether the different syntax tests measure the same skill. This paper compares three tests of syntactic knowledge in terms of 1) how highly the tests correlate with each other, 2) how dependent the tests are on vocabulary and problem-solving skill, and 3) to which extent the tests explain variance in reading comprehension after controls.
Method: 69 4th grade students were given tests of letter decoding, vocabulary, problem-solving skill, reading comprehension, and three tests of syntactic knowledge: the TROG-2, a sentence repetition test, and a multiple-choice sentence comprehension test.
Results: The three syntax tests differed in several ways. First, they correlated less than expected for measures of the same skill. Second, the TROG-2 correlated significantly higher with problem-solving skill than the two other tests did. Third, while the TROG-2 and the sentence comprehension test explained unique variation in reading comprehension after controls, the sentence repetition test did not.
Conclusion: The results of the study indicate that different syntax tests may not always be measuring the same skill(s). This has consequences for the interpretation of studies of syntactic knowledge and reading comprehension as studies that make use of different syntax measures may not be directly comparable. Items were not completely parallel across tests, so further research is needed to pinpoint if the differences found in the present study are due to differences in format or items.