The document discusses the development of a predictive model by an Office of Institutional Research to help determine the makeup of an upcoming freshman class. It describes how the office created a data warehouse of 6 years of enrollment data and, in collaboration with an Office of Enrollment Management, developed a dataset called Inq2005 from demographic, academic, financial, and other variables to analyze inquiries for the Fall 2005 semester. Some nominal variables were replaced with interval variables to improve the model, and some variables were excluded due to missing values or because they could not be used in admission decisions.
The document discusses the development of a predictive model by an Office of Institutional Research to help determine the makeup of an upcoming freshman class. It describes how the office created a data warehouse of 6 years of enrollment data and, in collaboration with an Office of Enrollment Management, developed a dataset called Inq2005 from demographic, academic, financial, and other variables to analyze inquiries for the Fall 2005 semester. Some nominal variables were replaced with interval variables to improve the model, and some variables were excluded due to missing values or because they could not be used in admission decisions.
The document discusses the development of a predictive model by an Office of Institutional Research to help determine the makeup of an upcoming freshman class. It describes how the office created a data warehouse of 6 years of enrollment data and, in collaboration with an Office of Enrollment Management, developed a dataset called Inq2005 from demographic, academic, financial, and other variables to analyze inquiries for the Fall 2005 semester. Some nominal variables were replaced with interval variables to improve the model, and some variables were excluded due to missing values or because they could not be used in admission decisions.
Professor’s Name: Course Name: Date of Submission: April 16th, 2021
University Enrollment Prediction
The role of developing a predictive model was assigned to the Office of Institutional Research, with the Office of Enrollment Management serving as a consultant on the project. The Office of Institutional Research created and managed a data warehouse that held enrollment data for the previous six years. It was determined that the Fall 2004 inquiries would be used to create a blueprint that would help form the Fall 2005 freshman class. In collaboration with Enrollment Management, the data set Inq2005 was created over a period of several months. Demographic, financial, number of correspondences, student interests, and campus visits were among the variables in the data collection. Using historical data and patterns, several variables were developed. For example, the percentage of inquirers from that high school who enrolled over the previous five years replaced the high school code. Over 90,000 measurements and 50 variables were included in the final data collection. The number of variables in this case study was decreased. The Inq2005 data set is available in the AAEM library, and the variables are mentioned in the table below. The number of missing values in some of the variables caused them to be automatically rejected. Since the nominal variables ACADEMIC INTEREST 1, ACADEMIC INTEREST 2, and IRSCHOOL were replaced by the interval variables INT1RAT, INT2RAT, and HSCRAT, respectively, they were rejected. Academic interest codes 1 and 2 were replaced, for example, by the percentage of inquirers who reported certain interest codes and then enrolled over the previous five years. The vector IRSCHOOL used to be the student's high school code, but it was replaced by the percentage of inquirers from that high school who enrolled in the last five years. Since they cannot be included in admission decisions, the variables ETHNICITY and SEX were eliminated. Several variables monitor the university's different forms of interactions with students.
7.9 Consider The Single-Sampling Plan Found in Exercise 15.8. Suppose That Lots of N 2,000 Are Submitted. Draw The ATI Curve For This Plan. Draw The AOQ Curve and Find The AOQL