SBIR Phase I: Within-test measurement of learning progress: unbiased assessment towards bridging the educational opportunity gap

小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力是帮助贫困学生接受高等教育，到2035年培养超过15万名高素质的专家。在一个快速变化的文明中，人们无法预测哪些职业和技能将在未来几十年内消失，哪些将出现新的职业和技能。这使得一个人现有的技能水平比他们获得新技能的潜力提供的信息要少。与现有的评估方法不同，该项目侧重于寻找已经具备技能的个人，该项目将帮助识别那些更善于获得新技能的人。这个小型企业创新研究(SBIR)第一阶段项目将创建一个评估算法和一个在线测试平台，以解决目前高风险认知和教育评估中缺乏公平和公平的问题。这将通过结合两项创新来实现：(I)在教育干预期间直接衡量学习能力的增长，而不分成测试前、测试后和培训阶段；以及(Ii)通过应用适应性项目选择和个性化反馈，最大限度地提高学生认知能力的变化率。这两项创新将最大限度地提高干预的效果及其测量的准确性。因此，学习能力可以在一个相对较短的时间内进行衡量，与通常分配给标准化测试的时间相当。这项建议的第一阶段目标是(I)展示提取学习能力的可行性，(Ii)提高信噪比，以及(Iii)消除初始熟练程度的不平衡。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to help underprivileged students obtain higher education, leading to more than 150,000 highly-qualified specialists by 2035. In a rapidly changing civilization, one cannot predict which professions and skills will disappear in the next decades and which new ones will emerge. This makes an individual’s existing skill level less informative than their potential to acquire new skills. Unlike existing assessment methods, which focus on finding already skilled individuals, this project will help identify those better at obtaining new skills.This Small Business Innovation Research (SBIR) Phase I project will create an assessment algorithm as well as an online testing platform to address the current lack of fairness and equity in high-stakes cognitive and educational assessments. This will be accomplished by combining two innovations: (i) measuring the growth of learning capabilities directly during an educational intervention without separation into pre-test, post-test, and training phases; and (ii) maximizing the rate of change of a student’s cognitive capabilities by applying adaptive item selection and personalized feedback. These two innovations will maximize both the effect of intervention and the accuracy of its measurement. As a result, learning capability can be measured in a single relatively short session comparable to the time typically allocated for standardized testing. The Phase I goals of this proposal are (i) demonstration of feasibility of extracting learning capabilities, (ii) increased signal-to-noise ratio, and (iii) elimination of the imbalance in initial proficiency levels.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.