Application of Statistical Classification Analysis to Engineering Student Recruitment

统计分类分析在工科生招生中的应用

• 批准号: 0836028
• 负责人: Edward Anderson
• 金额: --
• 依托单位: Texas Tech University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-01-01 至 2010-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0836028&HistoricalAwards=false
• 关键词: Application; Statistical; Classification; Analysis; Engineering

This exploratory project will apply statistical classification analysis to college admission data to identify those variables that impact a student?s choice of engineering as the first and perhaps only major field of study. Variables, such as entrance exam scores, high school GPA, socio-economic status, ethnicity, gender, pre-engineering courses completed, and other factors that are suspected to influence an adolescent?s decision to study engineering, will be examined to determine their role in this decision. Specifically, statistical logistic regression, classification tree, and recently developed random forest techniques will be applied to a sample of students admitted to Texas Tech University. Once accomplished, these techniques will classify the variables and yield a predictive model for student recruitment to study engineering. These findings can then be used by the engineering community to design courses, curricula and recruitment programs that both attract students and also introduce students to the study of engineering in a manner more consistent with their expectations and abilities. Improved recruitment and retention strategies will help increase the number of engineering graduates needed by the nation?s workforce, and the results of this project will be a resource for guidance about recruiting students to engineering programs. This project will apply recently developed statistical classification techniques to a large data sample and can filter through many variables simultaneously and identify the significance of each variable. General recruitment of engineering students can be improved by taking advantage of this additional insight into why students select engineering prior to entering college. Using these results efforts and initiatives directed towards increasing the number of students enrolling in engineering can be focused on to the significant decision variables. The findings will also be important for college recruiters as they design and implement engineering student recruitment programs.

该探索性项目将对大学录取数据进行统计分类分析，以确定影响学生选择工程学的变量，这是第一个，也许是唯一的主要研究领域。变量，例如入学考试成绩，高中GPA，社会经济地位，种族，性别，完成的工程学课程以及其他怀疑会影响青少年的决定研究工程的因素，以确定他们在这个决定中的作用。具体而言，统计逻辑回归，分类树以及最近开发的随机森林技术将应用于入学的学生样本中。一旦完成，这些技术将对变量进行分类，并为学生招聘研究工程提供预测模型。然后，工程社区可以将这些发现用于设计课程，课程和招聘计划，这些课程既吸引学生，又以与他们的期望和能力更一致的方式介绍学生学习工程学。改进的招聘和保留策略将有助于增加国家劳动力所需的工程毕业生的数量，该项目的结果将成为指导招募学生参加工程课程的资源。该项目将对最近开发的统计分类技术应用于大型数据样本，并可以同时透过许多变量过滤并确定每个变量的重要性。可以通过利用这种额外的见解来改善工程专业学生的一般招聘，以了解为什么学生在进入大学之前选择工程学。利用这些结果和针对增加工程学学生人数的努力和举措可以集中在重大的决策变量上。这些发现对于大学招聘人员设计和实施工程学学生招聘计划也将很重要。