Child heritage speakers’ acquisition of the Spanish subjunctive in volitional and adverbial clauses

(In)frequently asked questions: On types of frequency and their role(s) in heritage language variability

“…A second (and related) concern with quantifying the effects of HL use, as mentioned in the preceding paragraphs, is that HL use/ exposure is often strongly correlated with-and therefore hard to disentangle from-other potentially influential factors such as HL proficiency, age of acquisition of the majority language, and even formal education in the HL. As noted above, Dracos and Requena (2022)-as well as López-Beltrán-found that both HL use/experience and HL proficiency were statistically significant predictors of HSs' grammatical performance, making it impossible to isolate the specific influence of HL use itself. (In the first of these studies, recall that proficiency was actually a stronger predictor than reported HL experience).…”

“…What sets these studies apart from Cuza (2016) and Montrul and Sánchez-Walker (2013) is that in each case, HL experience is seamlessly integrated into the inferential statistical modeling, thereby allowing for more reliable insight into the effects of this potentially critical explanatory variable. Dracos and Requena (2022) investigated child HSs’ production of a few different types of subjunctive mood morphology in Spanish, including, most relevantly for the present study, and volitional subjunctive forms (e.g., quiero que bailes SUBJ ‘I want you to dance’). Critically—and in contrast with previous studies of HSs and subjunctive mood— Dracos and Requena’s (2022) analyses incorporated information about HL usage/exposure, which they combined into a single variable that was included as a fixed factor in their mixed-effects statistical models.…”

“… Dracos and Requena (2022) investigated child HSs’ production of a few different types of subjunctive mood morphology in Spanish, including, most relevantly for the present study, and volitional subjunctive forms (e.g., quiero que bailes SUBJ ‘I want you to dance’). Critically—and in contrast with previous studies of HSs and subjunctive mood— Dracos and Requena’s (2022) analyses incorporated information about HL usage/exposure, which they combined into a single variable that was included as a fixed factor in their mixed-effects statistical models. Results indicated that HSs with higher use of the HL (as reported by their caretakers) were significantly more likely to produce subjunctive mood morphology, pointing to HL experience as a factor in explaining certain between-speaker differences 1 .…”

“… Dracos and Requena (2022) calculated (child) HSs’ experience with the HL by asking the HSs’ parents to provide approximate percentages of the time that their children hear/use the HL during the week, as well as on weekends. After receiving the responses, the researchers recoded the data into five different categories (0%–19%, 20%–39%, 40%–59%, 60%–79% and 80%–100%), which became levels of a fixed factor (HL usage/exposure) in their subsequent statistical modeling.…”

“…After receiving the responses, the researchers recoded the data into five different categories (0%–19%, 20%–39%, 40%–59%, 60%–79% and 80%–100%), which became levels of a fixed factor (HL usage/exposure) in their subsequent statistical modeling. While we applaud Dracos and Requena’s (2022) utilization of this variable in their analyses, we must acknowledge, too, the potential unreliability of the percentage estimates that they received. When a parent estimates the percentage of the time that their child uses the HL, what factors might they be considering (or not)?…”