7 Quantitative Targets and School Lists

Mr. Curtis Westbay

2023-05-19

Questions to Find College Fit

College Fit Quiz

Rugg's Recommendations


7.1 The Academic Index

Colleges will consolidate the academic data of applicants into a single quantitative metric: the academic index

  • This is an unpublished, institution-specific metric
  • Some colleges use a formula, others make estimates
  • Some colleges use the index to get started, others to make final decisions

7.2 Threshold v. Holistic Admission

  • When a college makes admission decisions only based on index: threshold
  • When a college uses index as first step in review: holistic
  • Selective colleges conduct holistic review

7.3 Example Academic Indices


These variables would be the place to input your data.

U: unweighted GPA

S: highest SAT score from a single administration

T: highest ToEFL score from a single administration

R: sum of all AP scores

7.3.1 Range-based Index


For the BIPH Classes of 2022 and 2023, these were the minima and maxima for each variable:

Variable Min. Max. \(\Delta\)
GPA (U) 2.08 4.00 1.92
SAT (S) 1100 1600 500
ToEFL (T) 74 120 46
AP Sum (R) 6 70 64

\(AI_{U} = (U - U_{min})/\Delta U\)

\(AI_{S} = (S - S_{min})/\Delta S\)

\(AI_{T} = (T - T_{min})/\Delta T\)

\(AI_{R} = (R - R_{min})/\Delta R\)

\(AI = ((\)\(AI_{U}\) + \(AI_{S}\) + \(AI_{T}\) + \(\frac{1}{2}\)\(AI_{R}\)\()/(.035))\)


7.3.2 Example Academic Index: Range-based

This would be how to calculate the range-based academic index score for a student with a 3.33 unweighted GPA, a 1380 SAT score, a 104 ToEFL score, and a sum of AP scores equal to 28.


\(U: AI(3.33) = (3.33 - 2.08)/1.92 = 0.65104\)

\(S: AI(1380) = (1380 - 1100)/500 = 0.56\)

\(T: (104) = (104-74)/46 = 0.65217\)

\(R: AI(28) = (28 - 6)/64 = 0.34375\)

\(AI = ((0.65104 + 0.56 + 0.65217 +\) \(\frac{1}{2}\)\(0.34375)\)\(/(.035))\)

\(AI = 58.1454\)

7.3.3 Mean-based Index


For the BIPH Classes of 2022 and 2023, these were the mean values and maxima for each variable:

Variable Mean Max. \(\Delta\)
GPA (U) 3.5666 4.00 0.4334
SAT (S) 1415.8594 1600 184.1406
ToEFL (T) 102.74219 120 17.25781
AP Sum (R) 38.8594 70 31.1406

\(AI_{U} = (U - \bar U )/\Delta \bar U\)

\(AI_{S} = (S - \bar S)/\Delta \bar S\)

\(AI_{T} = (T - \bar T)/\Delta \bar T\)

\(AI_{R} = (R - \bar R)/\Delta \bar R\)

\(AI = ((\)\(AI_{U}\) + \(AI_{S}\) + \(AI_{T}\) + \(\frac{1}{2}\)\(AI_{R}\)\()/(.035))\)


7.3.4 Example Academic Index: Mean-based

This would be how to calculate the mean-based academic index score for a student with a 3.33 unweighted GPA, a 1380 SAT score, a 104 ToEFL score, and a sum of AP scores equal to 28.


\(U: AI(3.33) = (3.33 - 3.5666)/0.4334 = -0.54592\)

\(S: AI(1380) = (1380 - 1415.8594)/184.1406 = -0.19474\)

\(T: AI(104) = (104-102.74219)/17.25781 = 0.07288\)

\(R: AI(28) = (28 - 38.8594)/31.1406 = -0.34872\)

\(AI \approx (-0.55 + -0.19 + 0.07 +\) \((\frac{1}{2})\)\(-0.35)\)\(/(.035)\)

\(AI \approx -34.02462\)


7.4 Predictive Efficacy

This is a table of the correlation coefficients between mean-based and range-based academic index scores and the “best” offer ranking for each student in the Classes of 2022 and 2023:

Correlation Coefficients Mean-based Range-based
GPA (U) -0.83204 -0.83204
SAT (S) -0.77124 -0.77701
ToEFL (T) -0.67843 -0.67594
AP Sum (R) -0.77058 -0.77094
Academic Index -0.85662 -0.84851

7.4.1 Limitations

  • Not every student submits (or even has) some of this data
  • Based on U.S. News and World Report “Top U.S. Universities” ranking list
  • Doesn’t account for qualitative components, whatsoever
  • Rank \(\neq\) selectivity
  • Not specific to programs within universities

7.4.2 Chancing Model for Range-based Index


\(\hat{y} = 82.63184 + (-.71984 * AI)\)


\(\hat{y}\): the projected college ranking

\(AI\): your range-based academic index value

Note: the coefficient of determination for this model (\(R^{2}\)) is 0.7246.


7.5 Holistic Review: Bucketing

Selective colleges will have an admission policy which utilizes holistic review. That is, the automatic calculation for an academic index value only provides an initial context for admission decisions. Typically, students are placed into “buckets” based on their academic index evaluation. From here, colleges will set internal targets for selectivity, yield, and resource usage for these buckets.

7.6 School Lists

  • Choice of an Early Decision school, if applicable
  • Balanced, as concerns likelihood of admission:
    • Reach
    • Match
    • Likely
  • Typically, 10-15 schools

7.6.1 Process

  • Determine needs and fit
  • Make a “longlist”
    • Consists of schools that fit priorities
  • Make a “shortlist”
    • Schools that appeal after deep research
  • Make a school list
    • 10+ reasons to apply to each school
    • ED, EA, RD, etc. determined

Step 0: Determine Fit Priorities

Be deeply introspective– what matters to you? If you don’t really know, then take the “fit quiz” and inventories on Cialfo.

  • Computer Science (Web Development)
  • Practical > Experimental
  • Industry experience and portfolio creation
  • Front-end \(\nless\) back-end
  • Teaching faculty > researching faculty
  • Rural setting, small classes
  • Cold weather, Midwest or mountains

STEP 1: Find Schools that Fit

Here, I started by finding schools with strong CS programs in web development using Rugg’s Recommendations, then looked for schools in the right geographical locations.

Allegheny College Bennington
Bellarmine University Bentley University
Brandeis Bucknell
Carleton College Carroll College
Centre College Clark University
Colorado College Davidson College
Dickinson College Elon University
Grinnell College Kalamazoo College
Kenyon College Macalester
Marist College Miami University (OH)
Middlebury College Milwaukee School of Engineering
Missouri Univ. of Science and Tech. Monmouth College
Montana Tech. University Univ. of Northern Iowa
Oberlin College Rhodes College
Univ. of Richmond Robert Morris Univ.
St. Olaf College Truman State Univ.
Willamette Univ. College of Wooster

STEP 2: Research

Here, I looked at the website of each school from Step 1 (my “Longlist”) and made a judgment about the suitability of their academic programs.

Allegheny College Bennington
Bellarmine University Bentley University
Brandeis Bucknell
Carleton College Carroll College
Centre College Clark University
Colorado College Davidson College
Dickinson College Elon University
Grinnell College Kalamazoo College
Kenyon College Macalester
Marist College Miami University (OH)
Middlebury College Milwaukee School of Engineering
Missouri Univ. of Science and Tech. Monmouth College
Montana Tech. University Univ. of Northern Iowa
Oberlin College Rhodes College
Univ. of Richmond Robert Morris Univ.
St. Olaf College Truman State Univ.
Willamette Univ. College of Wooster

STEP 3: Trim Down the List

Here, I removed the schools with CS programs that seemed more focused on theory than on application.

Allegheny College Carleton College
Carroll College Clark University
Colorado College Davidson College
Dickinson College Elon University
Grinnell College Marist College
Miami University (OH) Middlebury College
Monmouth College Montana Tech. University
Univ. of Northern Iowa Oberlin College
Univ. of Richmond Robert Morris Univ.
Truman State Univ. Willamette Univ.
College of Wooster

STEP 4: Personal Ranking and Heuristic Chancing

Here, I made an initial personal ranking of the schools from Step 3 (my “Shortlist”), estimated my chances of admission for each, and continued to conduct college research on them each.

Marist College Match, ED
Colorado College Reach, ED
Elon University Match, EA
Grinnell College Reach, RD
Allegheny College Match, EA
Truman State Likely, RD
Davidson Match, RD
Clark Match, RD
Oberlin Reach, RD
Robert Morris Likely, RD
Carroll College Likely, RD
Montana Tech. Likely, RD
University of Northern Iowa Likely, RD
Carleton Reach, RD
Dickinson Reach, RD

STEP 5: 10 Things Other Than Rank

As I continued to research (Step 4), I kept notes with links to the information I found and tried to find a minimum of 10 things other than rank that would lead me to apply to the schools on my school list.

Example: Marist

  1. Liberal arts curriculum
  2. “Academic Core” (first-year seminar, cognitive studies pathway)
  3. Major: CS with Software Dev. concentration, 5-year BS/MS, digital media, etc.
  4. IBM Enterprise Computing Research Lab
  5. Marist and IBM Joint Study
  6. 83% participated in 1+ internship (Class of 2017)
  7. Strong tendency toward industry experience in CS faculty
  8. “at least 35% of courses require projects prepared by a team of students”
  9. Poughkeepsie, NY (pop. 31,000); rural, near Hudson River
  10. Career Services Center (portfolio review, career search, interviews, etc.)