This dataset includes Dutch secondary school students' responses to a spelling test. A total of 259 students (200 of them were used in the analyses for the paper), from first grade up and to sixth grade, from two educational levels (havo and vwo), were asked to spell Dutch homophonous verb forms to investigate how well these verb forms are spelled and what this tells us about the participants' (application of) grammatical knowledge. These verb forms sound identical, but are spelled differently because of their grammatical function. The students performed a self-paced spelling task. Students were presented with a sentence with the target form replaced by a gap. Behind this gap, the infinitive of the verb form was given between brackets. The students' task was to type in the correct form. They had to answer to proceed to the next sentence. No time pressure was used. The dataset includes 72 test verbs and 19 filler verbs. Half of the test verbs have third person singular present tenses that are homophonic with the past participles, but are spelled differently. The other half of the test verbs have past tense singulars that are homophonic with the adjectival past participles, but are spelled differently. Each verb occurred in both forms in the experiment. The dataset belongs to the following paper: Chamalaun, R.J.P.M., Bosman, A.M.T., & Ernestus, M.TC. (2021). The role of grammar in spelling homophonous regular verbs. Written Language & Literacy, 24(1), 38-80.