Information for UF Instructors and Advisors about the Retention Predictor
Let me give you some information about the Predictor and the modeling. Josh Norman sent Noel Levitz as much data as we could garner from Banner as possible, and we added data that Kimberlie Moock had on EIU Reads, Prowl, and JumpStart to see what factors contribute to student retention. Josh sent NL a data file containing the FA10 new freshman cohort and the FA11 new freshman cohort. The file was sent in early summer, so we did not use FA12 to model because we did not know who was retained from that year until last week!
The wizards (otherwise known as statisticians) at Noel Levitz built our model based on all these data points, and 8 variables came back. Eight is a higher number than they often see in models; 5 or 6 is more common. I want to explain each factor, and I’m going to do that in the order you will find them on the spreadsheet I will send you. Each of you will receive an excel sheet from me with your students and their information from the Predictor. The AAC and Gateway advisors and the SSC have received a similar list from me concerning the students they advise.
The spreadsheets will include student names, email addresses, and E# as well as the predictor information. The list below is the order in the spreadsheet, not by order of “importance” or risk factor.
Variable 1: Athlete
This variable means that being an athlete gives a student a boost for retention, but because we didn’t want to give all non-athletes a strike for that, it is not really included in the total number of variables. All you really need to know about this is that being an athlete can really only help a student’s persistence.
Variable 2: EIU Reads Attendance
This variable indicates that students who attended the EIU Reads reading circle on Friday were more likely to persist than those who do not. This goes to motivation and willingness to do what they are told to do. There was an 8% retention difference. So, students who did not attend this fall are at-risk of not being here next fall. This attitude probably translates to their coursework and course attendance as well.
Variable 3: Ethnic Flag
Eastern distinguishes in Banner between race and ethnicity, so this flag indicates that students who indicate Hispanic origin are at-risk of persistence. This column will say “Hispanic” or “Not Hispanic” and only students that have the “Hispanic” flag have that variable as a risk factor.
Variable 4: Expected Family Contribution >0
This is the first of our 3 financial variables. If the answer is “true” here than the EFC is not zero and the student does not have this as a risk factor. If the answer is “false” the student does have this as a risk factor. That means that the student’s expected family contribution is zero, which is a risk factor.
Variable 5: Financial Aid Gap
This is the second of our 3 financial variables. This column tells us what the gap is between the financial aid offered by Eastern and the expected family contribution. The higher the number, the more likely it is that the student is worried about paying for college. The retention gap is $9006.00 or greater for the most at-risk group (indicated by a 3 in this column). That group’s retention was 67% in the model. The “2” indicates students in the middle gap in financial aid; this category of student had a 79% retention rate in the model. The “1” indicates students with the smallest gap, and the model retention for this group was 81%, so it is above our average for those 2 years.
Variable 6: High School GPA
Students with a high school gpa of 2.84 or lower are at risk, and those with 2.85 or higher are not.
Variable 7: Percentage of Need Met
This is the third of our 3 financial variables. The number represented here is a percentage. Again, Noel Levitz have found that students for whom 43.5% or higher of their financial need is met to pay for college are retained above our retention average at 82%. Students who have 22.5-43.5% of their need met retain at 6% lower than our average retention rate, and students who have 22.5% or less of their met have a retention rate of 56.9%, which is very scary. So, for this column, the higher the number, the better off a student is in terms of their need met.
Variable 8: Department or Program Area
Because we have so many majors and concentrations, major did not show up as a variable, but department or program area did. We are trying to keep this to ourselves until we know how to present it to the departments, but this variable may be the one that is most important for you to know and understand given your work with the students. Departments that are risk factors for students include (order of risk):
Africana Studies
Psychology
Sociology/Anthropology
Health Studies
FCS
Undeclared
Physics
Biological Sciences
Mathematics/Computer Sciences
Art
Communication Studies
All other programs contribute to persistence for students in their programs. You can probably speculate as well as I can on why this is, but it becomes even more important that students understand what it means to major in these programs and what it takes to succeed. Some of these programs are extremely challenging and students don’t fully understand that, and some of them attract our least prepared students. So, this makes Academic Foundation Day especially important this year. Please discuss AFD with your students, explain how they can use the presentations and fair to their advantage, talk to them individually about what programs and tables they might want to visit. Make this event as useful as possible though your discussions and assignment.
The last two columns on your spreadsheets will be the number of risk factors and the retention predictor percentage.
Number of Risk Factors
The number of risk factors indicates the total number of risk factors each student has. We are most concerned about students with 3 or more. Students with 0-2 risk factors are more likely to persist. Remember that no one will have 8 because we did not count not being an athlete against all the non-athletes.
Retention Predictor
The model is broken down into 4 quartiles of risk: the most at-risk are students with a score of .25 and lower, the next quartile of risk are those .26-.50; the third quartile is the .51-.75, and the lowest risk are those with a score of .76 to .99.