We empirically evaluated three theoretical models—the threshold hypothesis, transfer theory, and time-on-task theory—for educating English language learners (ELLs), with a focus on the role of language factors in explaining achievement differences among ELLs. Participants were 196 sixth graders with Spanish language backgrounds who started learning English in kindergarten and then were continuously enrolled in a U.S. school. Structural equation modeling was used to estimate the extent to which Spanish and English language and literacy—skills that are emphasized differentially in competing theories for educating ELLs—predict academic achievement assessed in English. Results indicated that Spanish literacy, over and above English language proficiency, was substantially predictive of academic achievement, consistent with the transfer theory. This model was a more focused version of the threshold hypothesis, in that the weaker predictor of Spanish oral language proficiency was excluded. Time-on task theory was not supported.

中文翻译：

---

语言教育计划效果的三种理论：双语和纯英语政策的实证评估

我们经验性地评估了三个理论模型——阈值假设、迁移理论和任务时间理论——用于教育英语学习者 (ELL)，重点是语言因素在解释 ELL 之间的成就差异中的作用。参与者是 196 名具有西班牙语背景的六年级学生，他们在幼儿园开始学习英语，然后继续在美国学校就读。结构方程模型用于估计西班牙语和英语语言和读写能力（在教育 ELL 的竞争理论中差异化强调的技能）预测用英语评估的学术成就的程度。结果表明，除了英语语言能力之外，西班牙语读写能力在很大程度上可以预测学业成绩，这与迁移理论一致。该模型是阈值假设的更集中版本，因为排除了西班牙语口语能力的较弱预测因子。不支持按时任务理论。