Running the Numbers: Improving Your Position for Enrollment Planning and Forecasting - Jeancarlo Bonilla

Running the Numbers: Improving Your Position for Enrollment Planning and

Forecasting”

TxGAP – Summer Conference July 20th, 2012

JeanCarlo (J.C) Bonilla

Director of Graduate Enrollment Management New York University, Polytechnic Institute

The Plan for the Session: •  Overview of predictive modeling & optimization for enrollment

management •  Case 1: Do cycles have memory? The case of the 3-yr adjusted yield (4

examples) •  Case 2: Am I making my class? Modeling for scenarios forecasting (3

examples) •  Case 3: The magic ball, ranking and an enrollment predictor (2 examples) •  Case 4: Opps, I ran out of time, but this is a very cool model

Worksheets download at EnrollmentAnalytics.com

A little bit about me & where I work...

Where is the industry today with the idea of business analytics &

intelligence?

Standard  Reports  

“what  happened”  

Ad  hoc  Report  “how  many,  how  often,  where”  

Query  “what  is  exactly  the  problem”  

Alert  “what  actions  are  

required”  

Degree of intelligence

Descriptive Analytics

Statistical  Model  

“why  is  this  happening”  

Randomized  testing  

 “what  happens  if  we  try  this”  

Predictive  Model/Forecast    

“what  will  happened  next”  

Optimization    “what  is  the  best  

that  can  happened”  

Degree of intelligence

Predictive Analytics

Informatics/Analytics industry is moving

from small data to big data from data

analytics to data scientist

So, what do we know so far about

predictive modeling for Enrollment Management?

TACTIC

SUSPECTS> PROSPECTIVE> APPLICANTS> ADMITS> DEPOSITS> NEW

Standard  Reports  “what  happened”  

Ad  hoc  Report  “how  many,  how  often,  where”  

Query  “what  is  exactly  the  

problem”  

Alert  “what  actions  are  

required”  

TACTIC

Statistical  Model  “why  is  this  happening”  

Random  Testing    “what  happens  if  we  

try  this”  

Predict  &  Forecast    

“what  will  happened  next”  

Optimization    “what  is  the  best  that  

can  happened”  

Predictive Analytics

SUSPECTS> PROSPECTIVE> APPLICANTS> ADMITS> DEPOSITS> NEW

Examples of Enrollment Predictive Modeling •  Case 1: North Dakota University –  Type of Model: inquiry model using geo-demographic

–  Predictive Power: 36% of students who will enroll & 97% of student who will not enrolled

Examples of Enrollment Predictive Modeling •  Case 2: University of Minnesota –  Type of Model: application generation model using, ACT and geo-

demographic information

–  Predictive Power: 85% of applicants to a “large research university” are from within the same state or form a neighboring state

Examples of Enrollment Predictive Modeling: •  Case 3: State University of New York –  Type of Model: lead modeling using geo-demographic, academic data,

and financial aid data

–  Predictive Power: 45.67% of applicants predicted to enroll did in fact matriculate and 82.16% who where predicted not to enroll did not matriculate

Better predictive power with students who

do not matriculate than with model that forecast actual students enrollments

The “technique” is used in other consolidated markets... if it works for them, it should work for us!

It requires quantitative analysis of past student characteristics to predict probabilities of

future results

Your predictive modeling team should have people who are confortable doing:

The modeling guy: 1.  Regression Analysis (logistic regression)

2.  Business analytics

The computer guy: 1.  Database architecture & design

2.  Database querying 3.  Data aggregation & integration

4.  Data reporting

Access to historical data is required!

Modeling 101: Defining Model Attributes

Student Behavior (influences, emotions,

competition)

Student Characteristics (geo-demographic, academic,

financial aid)

...and off course, there is a problem with that!

SUSPECTS> PROSPECTIVE> APPLICANTS> ADMITS> DEPOSITS> NEW

Stealth Applications 30%-40% of adult students

Source: Aslanian Market Research

...an approach for predictive modeling in enrollment management

Applicants Prospective Students

New students Applicants

CASE#1:

Do cycles have memory? The case of the 3-yr adjusted yield

Predictions through the admissions funnel

Download worksheets at: www.EnrollmentAnalytics.com

Recommendations

1.  Rapid “back-of-the-envelop” modeling

2.  You can go “up” or “down” the funnel

3.  Need for historical data (static snapshots of cycles)

4.  Student characteristics add more resolution to the model

5.  Use of adjusted 3-year cycles are useful for historical modeling

6.  Historical validity: account for new initiatives

CASE#2: Am I making my class?

New Student forecaster

Download worksheets at: www.EnrollmentAnalytics.com

CASE#3: The magic ball

ranking and an enrollment predictor (2 examples)

Download worksheets at: www.EnrollmentAnalytics.com

5-Stage Admissions Funnel

Prospects

Applications

Admits

Deposits

New 2%

Example: say that you have 20k leads in your cycle and only 300 matriculate, then you have a

2% conversion rate

Now, you “observe” that 200 out of your 300 new students presents a subset of 5% of your

prospective students pool. This means that 1000 prospective students (5% of 20k) converted into 200 enrollments, which

means that your conversion rate for this subset is 20%

95%  

5%  

33%  

67%  

20,000 prospective students

300 new students

New conversion rate or predictability of 20%

20,000 prospective students

300 new students

FT-­‐Dom,  20%  

PT,  40%  

FT-­‐Int'l,  40%  

FT-­‐Dom,  5%  

PT,  40%  FT-­‐Int'l,  55%  

..and if you get really good at understanding your students...

Build a model that does the following:

Student  Name       Status       Predictor  

Hall,  Joy  

    Inquiry   11/16/10      

0.4       App.   N/A           Adm.   N/A           Conf.   N/A           Enr.   N/A      

Li,  Xiao  

    Inquiry   12/22/10      

0.6       App.   12/24/10           Adm.   3/23/11           Conf.   4/2/11           Enr.   N/A      

Lopez,  Jose  

    Inquiry   12/5/10      

0.2       App.   1/5/11           Adm.   1/29/11           Conf.   3/16/11           Enr.   N/A      

Mitchell,  Tamara  

    Inquiry   12/20/10      

0.2       App.   2/3/11           Adm.   N/A           Conf.   N/A           Enr.   N/A      

Smith,  John  

    Inquiry   1/26/11      

0.4       App.   1/28/11           Adm.   4/16/11           Conf.   5/5/11           Enr.   N/A      

Troy,  Bryan  

    Inquiry   12/13/10      

0.9       App.   N/A           Adm.   N/A           Conf.   N/A           Enr.   N/A      

So, how can I build a model like that predicts enrollments?

Student Uncertainty & Variance

Academic

Financial

Geographical

Demographical

Behavior & Personal Life

FT vs intl vs PT

Recommendations for Advanced Models

1.  It gets complicated

2.  Its “easy” to model for student characteristics, but complexity increases when accounting for student behavior

3.  Models are better at predicting student who do NOT register

4.  Every school is different, so every model is also different

5.  Trust your instincts! No one knows students better than you... Your job is then trying to articulate and generalized characteristics and behavior

CASE#4: Opps, I ran out of time, but this is a very

cool model

Download worksheets at: www.EnrollmentAnalytics.com

Final Recommendations

1.  Plan for good, bad, and what you think is going to realistic

2.  Avoid predictions but give options

3.  Its about resource allocation

4.  Work with other groups in your institution

5.  Trust your GEM instincts

6.  Its earsier to account for student characteristics, but modeling and forecasting behavior is very complex

JeanCarlo (J.C.) Bonilla jbonilla@poly.edu

www.EnrollmentAnalytics.com 718-260-3201

Thanks