using mcat® data in 2016 medical student selection

14-282 MCAT Student Selection Cover_Flat_Final2.indd 1 6/11/15 4:24 PM

www.aamc.org/mcat

© 2015 Association of American Medical Colleges. May not be reproduced or distributed without prior written permission.

This is a publication of the Association of American Medical Colleges. The AAMC serves and leads the academic medicine community to improve the health of all. www.aamc.org.

June 2015


(ver. 2)

www.aamc.org

i

Using MCAT Data in Medical Student SelectionUsing MCAT Data in Medical Student Selection

Dear Admissions Officers:

The new Medical College Admission Test® (MCAT®) was introduced in April 2015. The new MCAT exam is designed for today’s medical students and tomorrow’s doctors. It tests the knowledge and skills that students need when they enter medical school.

The transition to the new MCAT exam coincides with your admissions cycle for the 2016 entering class. Transitioning to the new MCAT exam will present new opportunities and challenges to your admissions committee. When the 2016 application cycle opens on July 1, you will receive scores from applicants who took the new exam, as well as scores from applicants who took the old exam. We expect about half of your applicants will report scores from the new MCAT exam and half will report scores from the old MCAT exam. As a result, this guide provides the information and data you need to work with both new and old MCAT scores.

• The first part of this guide focuses on the interpretation and use of scores from the new MCAT exam.

• The second part of this guide focuses on the interpretation and use of scores from the old MCAT exam.

The AAMC will be available every step of the way to help you make the transition from using scores from the old MCAT exam to using scores from the new MCAT exam. Please don’t hesitate to contact us if you have questions. You can reach us by email at [email protected].

Sincerely,

Cynthia A. Searcy Senior Director, MCAT Research and Development

ii


Table of Contents

Part 1: Using Scores from the New MCAT Exam in Student Selection .............................................. 1

The New Medical College Admission Test (MCAT) .................................................................................... 2

Benefits of the New MCAT Exam ............................................................................................................. 5

Content of the New MCAT Exam ............................................................................................................. 7Biological and Biochemical Foundations of Living Systems.....................................................................9Chemical and Physical Foundations of Biological Systems ....................................................................10Psychological, Social, and Biological Foundations of Behavior .............................................................11Critical Analysis and Reasoning Skills ..................................................................................................13

Scores from the New MCAT Exam ........................................................................................................... 14Using Percentile Ranks in 2016 Admissions ..........................................................................................16Using Confidence Bands in 2016 Admissions ......................................................................................19Using Score Profiles in 2016 Admissions .............................................................................................20Making Decisions about Applicants in the 2016 Admissions Cycle .......................................................21Designing Thresholds for Screening and Interviewing .........................................................................21Making Decisions about Individual Applicants ....................................................................................22

Data about the New MCAT Exam .......................................................................................................... 23Timeline for Reporting Data about the New Exam ..............................................................................23MCAT Validity Research Agenda .........................................................................................................23The Psychological, Social, and Biological Foundations of Behavior Validity Study ..................................24The Longitudinal MCAT Validity Study ................................................................................................25National MCAT Validity Data .............................................................................................................25

References ............................................................................................................................................. 25

Appendix A: The MCAT Psychological, Social, and Biological Foundations of Behavior Validity Study Results ......................................................................................................................................... 26

Appendix B: Sample Questions for the New MCAT Exam ....................................................................... 29

Part 2: Using Scores from the Old MCAT Exam in Student Selection .............................................. 39

Content of the Old MCAT Exam ............................................................................................................. 40

Scores from the Old MCAT Exam ........................................................................................................... 40

Validity Data on the Old MCAT Exam ..................................................................................................... 44

References ............................................................................................................................................. 51

1


Part 1: Using Scores from the New MCAT Exam in Student Selection

2

Using MCAT Data in Medical Student Selection

The New Medical College Admission Test (MCAT)

The new MCAT exam is designed for today’s medical students . . . and tomorrow’s doctors.

There is no question that the nation’s health care system is changing and future physicians will need more and better tools to manage the delivery system and meet the needs of the nation’s changing, growing, aging population. Medical schools are changing their curricula to meet these challenges by targeting the competencies that tomorrow’s physicians will need. The new MCAT exam, introduced in April 2015, was designed to match the adoption of competency-based curricula in medical schools by testing foundational knowledge and skills that students need when they enter medical school.

The transition to the new MCAT exam coincides with the 2016 application cycle. Transitioning to a new MCAT exam will present challenges to you and your admissions committee, but it also will present opportunities to consider a wider range of applicants with a broader range of academic competencies.

This guide is designed to help you use MCAT scores during this transition.

• It places the new exam in the broader context of your individualized, holistic review of applicants’ experiences, attributes, and academic preparation to admit students who best fit your school’s mission and goals.

• It describes the foundational concepts and skills in the natural, behavioral, and social sciences that are tested by the new exam, making connections between pre-medical preparation and the medical school curriculum.

• It describes ways to attach meaning to scores from the new exam, highlighting important changes in the score scales and score reports intended to help you make better decisions about the applicants who will best fit your medical school curriculum.

• It acknowledges the complexities of considering applicants in this admissions cycle—during which half of your applicants will report scores from the old exam and half will report scores from the new exam—providing guidance on ways to think about MCAT scores in this context.

• It concludes with a timeline and sources of data to help you learn more about the ways the new exam will support your admissions decision making.

Holistic review practices can help you transition from the old MCAT exam to the new exam.

Holistic review practices provide a firm foundation for selecting applicants with the academic and personal competencies that future physicians need. Your institutional mission, goals, and priorities provide a framework for evaluating applicants. They lay the foundation for admitting a class of capable, caring students who bring diverse interests, talents, and experiences to your institution.

Applicants provide admissions committees with rich information about their experiences, attributes, and academic backgrounds through their applications, personal statements, and interviews. Letter writers also provide information about applicants’ academic and personal competencies. Holistic review practices provide an important foundation for transitioning to the new exam. With the introduction of the new MCAT exam, it is important to remember the following best practices for considering data about academic preparation in the context of the varied information collected during the admissions process.

As you did for the old MCAT exam, you should triangulate score information from the new MCAT exam with information about applicants’ course completion, grades, grade trends, institutional selectivity, research experience, and other academic indicators. You should look for consistencies and inconsistencies in the stories these data tell.

3


Your admissions committee should carefully consider the rich and varied data that applicants provide. Table 1 describes the results of a 2013 survey about the importance of academic metrics, experiences, demographics, and other sources of applicant information for deciding which applicants to interview and which to admit. The table highlights the importance of different types of data in admissions decision making. Consideration of these data will help your admissions committee construct a class that meets the academic, clinical, service, and research missions of your medical school.

Using MCAT Data in Medical Student Selection7 MCAT® is a program of theAssociation of American Medical Colleges

Table1.MeanImportanceRatingsofAcademic,Experiential,andDemographicApplicationDataUsedbyAdmissionsCommitteesforMakingDecisionsaboutWhichApplicantstoReceiveanInterviewInvitationandOfferAcceptance(N=127)1

MeanImportanceRatings

AcademicMetrics Experiences Demographics2 SourcesofApplicantInformation

Highest Importance Ratings(> 3.0)

• UGPA:Cumulativescience/math

• MCATtotalscore• Upwardordownward

grade trend• UGPA:Cumulativetotal• Performanceinapost-

baccalaureate program• Selectivity of

undergraduateinstitution (Private)5

• Healthcareexperience• Communityservice/

volunteer experience• Experiencewith

underservedpopulations

• Navigatedthroughcultural barriers orchallenges

• Leadershipexperience

• U.S. citizenship/permanent residency(Public)5

• State residency (Public)5

• Interviewassessments3

• Lettersofevaluationor recommendation

• Personalstatements

Medium Importance Ratings(>2.5 and <3.0)

• Onscheduletomeetpre-medical coursework

• UGPA:Cumulativenon-science/math

• Research experience (Private)5

• Experiencewithpopulations unlike theapplicant

• Lackofaccesstooptimal educationalresources

• Specialfamilyobligations or othercircumstances

• Workorathleticscholarship obligationswhile in school

• Rural or urbanbackground (Public)5

• First-generationcollegestudent

• U.S. citizenship/permanent residency(Private)5

• Race/ethnicity• Socioeconomicstatus

(SES)

• Secondaryapplicationresponses4

Lowest Importance Ratings(<2.5)

• Completionofchallenging non-sciencecourses

• Selectivity ofundergraduateinstitution (Public)5

• Experiencewithprejudice

• Research experience(Public)5

• Rural or urbanbackground (Private)5

• Multilingual• Legacy• Gender• State residency (Private)5

1 Importance was rated on a four-point scale ranging from 1 to 4 ("Not Important," "Somewhat Important," "Important," and "Very Important," respectively). For each variable, we computed an overall mean importance rating based on admissions officers’ ratings of importance for making decisions about whom to interview and whom to accept (the mean importance rating for the interview variable is the exception to this rule because interview data were not available until applicants were invited to interview). We chose to classifyvariables using overall mean importance ratings because their mean importance ratings were similar for the interview and the acceptance phases. Variables are ordered by overall mean importance rating.2 The demographic variables listed here were culled from a larger list of variables that included personal attributes such as "Service Orientation," "Integrity," "Communication Skills," and "Cultural Competence," for example.3 Only available at the admissions stage where admissions committees make a decision to offer an acceptance.4 Only available for institutions that use application data to select the applicants who will receive secondary applications.5 Overall mean importance ratings for public and private institutions were significantly different from one another.

4


As shown in Table 2, national-level data on acceptance rates for applicants who took the old MCAT exam between 2012 and 2014 reinforce the messages these 2013 survey data provide. Each year, some applicants with high MCAT scores and undergraduate GPAs are rejected by all the medical schools to which they applied. In contrast, other applicants with more modest MCAT scores and undergraduate GPAs are accepted by at least one medical school. While undergraduate GPAs and MCAT total scores are important factors in admissions, they are not the sole determinants of admissions decisions.

In next year’s version of this guide, you are likely to see a similar pattern of results for applicants who took the new MCAT exam.

Scores from the new exam should not outweigh other application data in deciding which applicants will get secondary invitations, interview invitations, or acceptances.

As shown in Table 2, national-‐level data on acceptance rates for applicants who took the old MCAT exam between 2012 and 2014 reinforce the messages these 2013 survey data provide. Each year, some applicants with high MCAT scores and undergraduate GPAs are rejected by all the medical schools to which they applied. In contrast, other applicants with more modest MCAT scores and undergraduate GPAs are accepted by at least one medical school. While undergraduate GPAs and MCAT total scores are important factors in admissions, they are not the sole determinants of admissions decisions.

In next year’s version of this guide, you are likely to see a similar pattern of results for applicants who took the new MCAT exam.

Scores from the new exam should not outweigh other application data in deciding which applicants will get secondary invitations, interview invitations, or acceptances.

Table 2. Percentage of 2012−2014 Applicants Accepted into at Least One Medical School, by Old MCAT Total Score and Undergraduate GPA Range

GPA Total Old MCAT Total

5−14 15−17 18−20 21−23 24−26 27−29 30−32 33−35 36−38 39−45 All 3.80−4.00 4% 2% 11% 21% 38% 61% 76% 84% 88% 91% 69%

3/80 4/162 57/524 317/1,526 1,363/3,554 4,232/6,978 7,110/9,361 6,298/7,504 3,688/4,176 1,385/1,519 24,457/35,384 3.60−3.79 0% 2% 8% 16% 27% 45% 64% 76% 80% 85% 52%

0/177 8/367 83/1,024 371/2,332 1,332/4,866 3,725/8,284 5,997/9,359 4,513/5,973 1,782/2,235 435/514 18,246/35,131 3.40−3.59 0% 2% 5% 12% 22% 32% 48% 62% 70% 75% 36%

1/336 13/553 67/1,278 314/2,607 1,010/4,691 2,307/7,151 3,600/7,455 2,382/3,854 819/1,176 176/234 10,689/29,335 3.20−3.39 0% 1% 4% 11% 18% 23% 35% 47% 58% 65% 25%

0/370 5/561 41/1,168 249/2,262 604/3,344 1,012/4,369 1,453/4,106 889/1,902 316/547 74/113 4,643/18,742 3.00−3.19 0% 0% 3% 8% 17% 19% 29% 39% 48% 53% 18%

0/388 2/553 25/928 123/1,578 373/2,218 455/2,361 530/1,851 313/808 112/233 21/40 1,954/10,958 2.80−2.99 0% 1% 3% 6% 12% 16% 24% 27% 26% 29% 12%

0/368 4/386 19/626 54/908 132/1,069 158/998 179/746 85/310 22/86 7/24 660/5,521 2.60−2.79 0% 0% 3% 5% 9% 15% 21% 28% 32% 30% 9%

0/274 1/284 12/355 24/486 47/512 57/388 59/276 33/117 15/47 3/10 251/2,749 2.40−2.59 0% 0% 1% 3% 9% 14% 17% 16% 18% -- 6%

0/196 0/151 2/179 8/240 19/221 22/152 18/109 6/37 3/17 79/1,304 2.20−2.39 0% 0% 0% 0% 8% 12% 15% 7% -- -- 4%

0/132 0/77 0/94 0/91 7/88 8/68 6/39 1/14 22/611 2.00−2.19 0% 0% 0% 0% 0% 14% 18% -- 2%

0/53 0/40 0/42 0/28 0/30 2/14 2/11 4/220 1.47−1.99 0% -- 0% 0% -- -- -- 0%

0/42 0/10 0/12 0/90 All 0% 1% 5% 12% 24% 39% 57% 71% 79% 85% 44%

4/2,419 37/3,143 306/6,229 1,460/12,070 4,887/20,601 11,978/30,769 18,954/33,316 14,520/20,521 6,757/8,522 2,102/2,459 61,005/140,049 Notes: 1. Dark green shading = acceptance rates ≥ 75%; Light green shading = acceptance rates of 50−74%; grey shading = acceptance rates of 25−49%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score in each application year was used in this analysis.

5


Benefits of the New MCAT Exam

The new exam tests the academic preparation students need for today’s medical school.

The new MCAT exam is designed to help medical school admissions committees select students who are academically prepared for the curriculum they will begin in 2016. The blueprints for the new exam are evidence-based and target concepts that medical school faculty, residents, and medical students recently rated as important to entering students’ success. They are organized around the academic competencies described by seminal reports like the Scientific Foundations for Future Physicians (www.aamc.org/scientificfoundations) and the Behavioral and Social Science Foundations for Future Physicians (www.aamc.org/socialsciencefoundations). The quantitative and qualitative research that supports the definition and development of the new MCAT exam is strong (Schwartzstein et al. 2013).

The new exam shifts the focus from testing what applicants know to testing how well they use what they know. It focuses on the outcomes of learning by asking applicants to apply what they have learned. The new exam asks applicants to use their knowledge of the natural, behavioral, and social sciences to solve problems that call for scientific reasoning.

The new exam asks applicants to show that they can think and learn like physicians. In today’s information age, it is no longer possible for physicians to memorize all the information they need to make informed clinical decisions. Future physicians will need to know where to find credible information, how to evaluate it, and how to translate it into high-quality care. The new MCAT blueprints recognize this by emphasizing scientific reasoning and analysis skills, asking applicants to reason about research designs and results, and asking applicants to interpret data and draw conclusions from them.

The new exam underscores the important roles of behavioral and sociocultural factors in health and illness by including a section that tests concepts from the behavioral and social sciences. This foundational knowledge will prepare medical students to learn how behavior interacts with biological factors to influence health outcomes and how social inequities affect patients’ health. The new exam communicates the value that medical school admissions committees place on broad preparation for medical school, and it reinforces the diversity of interests and preparation that committee members look for in their applicants.

Finally, the new exam recognizes the heavy information-processing requirements of medical school curricula and asks students to demonstrate that they understand and can critically analyze text.

The new exam provides more reliable scores, supporting better decisions.

The architects of the new MCAT exam understood that medical schools need reliable information about applicants’ academic preparation. In response, they designed each section of the new exam with more test questions than were included on the old exam. More questions per section provide more information about applicants. The new scores are estimated with greater precision. Improved precision means improved decision making, allowing you to consider carefully applicants’ strengths and weaknesses across the four sections of the new exam as you judge your applicants’ fit with your curriculum.

The new exam was designed with fairness in mind.

Concerns about fairness played a critical role in developing the new exam blueprints, and particularly in ensuring that the concepts and skills tested on the new exam are taught widely. The architects of the new exam checked course offerings at minority-serving and under-resourced colleges and universities to ensure applicants have access to the needed courses. The designers also surveyed faculty at these institutions to learn about the content of their courses, and they examined data on applicants’ course-taking behaviors to learn whether applicants at these institutions complete the needed classes. These

www.aamc.org/scientificfoundations

www.aamc.org/socialsciencefoundations

www.aamc.org/socialsciencefoundations

6


analyses of course offerings, course content, and course taking at minority-serving and under-resourced institutions guided the blueprints for the new exam.

Another prominent fairness issue in designing the new exam was testing time. Although each section of the new exam includes more questions than sections of the old exam, the design of the new test gives examinees more working time per question than the old exam. Applicants have more time to review passages, read questions, and decide on their answers.

Finally, as previously mentioned, the new MCAT exam balances testing in the natural sciences with testing in the behavioral and social sciences and critical analysis and reasoning. The hope is that this balance will help diversify the physician workforce by making the exam and medical school application more attractive to individuals from more varied academic and demographic backgrounds.

7


Content of the New MCAT Exam

This section of the guide provides information about the design, interpretation, and use of the new MCAT exam. For ease of reading, this guide describes MCAT sections that have similar content and format together.

The new MCAT exam has four sections:

1. Biological and Biochemical Foundations of Living Systems

2. Chemical and Physical Foundations of Biological Systems

3. Psychological, Social, and Biological Foundations of Behavior

4. Critical Analysis and Reasoning Skills

The new and old MCAT exams are different.

To be ready for medical school in 2016, applicants need preparation in biology, chemistry, physics, and verbal reasoning, just as they did in the past. The new MCAT exam tests these concepts and also tests preparation in first-semester biochemistry, psychology, and sociology, with a greater emphasis on demonstrating scientific reasoning skills and by capitalizing on recent research about information-processing skills.

Content New MCAT Exam Old MCAT Exam The new exam:

• Requires broader preparation

• Asks students to apply what they’ve learned

• Focuses on behavioral and sociocultural aspects of health

• Incorporates the latest science on information processing

Biology ✔ ✔

Chemistry ✔ ✔

Physics ✔ ✔

Verbal Reasoning ✔ ✔

Biochemistry ✔

Psychology ✔

Sociology ✔

The two natural sciences and the behavioral and social sciences sections of the new MCAT exam test 10 foundational competencies and four scientific inquiry and reasoning skills. These sections ask students to combine their knowledge of concepts that are foundational for learning in medical school with their inquiry and reasoning skills. As shown in the figure on the following page, MCAT questions test foundational concepts and scientific reasoning skills together. The resulting scores provide information about applicants’ readiness to learn in medical school.

8


Fou

nd

atio

nal

Co

nce

pts

Tes

ted

on

th

e N

ew M

CA

T Ex

am

Bio

log

ical

an

d B

ioch

emic

al

Fou

nd

atio

ns

of

Livi

ng

Sys

tem

s

Ch

emic

al a

nd

Ph

ysic

al

Fou

nd

atio

ns

of

B

iolo

gic

al S

yste

ms

Psyc

ho

log

ical

, So

cial

, an

d B

iolo

gic

al

Fou

nd

atio

ns

of

Beh

avio

r

Fou

nd

atio

nal

C

on

cep

t 1

Fou

nd

atio

nal

C

on

cep

t 2

Fou

nd

atio

nal

C

on

cep

t 3

Fou

nd

atio

nal

C

on

cep

t 4

Fou

nd

atio

nal

C

on

cep

t 5

Fou

nd

atio

nal

C

on

cep

t 6

Fou

nd

atio

nal

C

on

cep

t 7

Fou

nd

atio

nal

C

on

cep

t 8

Fou

nd

atio

nal

C

on

cep

t 9

Fou

nd

atio

nal

C

on

cep

t 10

Biom

olec

ules

ha

ve u

niqu

e pr

oper

ties

that

det

er-

min

e ho

w

they

con

-tr

ibut

e to

th

e st

ruct

ure

and

func

tion

of c

ells

, and

ho

w t

hey

part

icip

ate

in

the

proc

esse

s ne

cess

ary

to

sust

ain

life.

Hig

hly

orga

nize

d as

sem

blie

s of

m

olec

ules

, ce

lls, a

nd o

r-ga

ns in

tera

ct

to c

arry

out

th

e fu

nctio

n of

livi

ng

orga

nism

s.

Com

plex

sy

stem

s of

tis

sues

and

or

gans

sen

se

the

inte

rnal

an

d ex

tern

al

envi

ronm

ents

of

mul

ticel

lu-

lar

orga

nism

s,

and

thro

ugh

inte

grat

ed

func

tioni

ng,

mai

ntai

n a

stab

le in

tern

al

envi

ronm

ent

with

in a

n ev

er-c

hang

-in

g ex

tern

al

envi

ronm

ent.

Com

plex

liv-

ing

orga

nism

s tr

ansp

ort

ma-

teria

ls, s

ense

th

eir

envi

ron-

men

t, p

roce

ss

sign

als,

and

re

spon

d to

ch

ange

s us

ing

proc

esse

s th

at

can

be u

nder

-st

ood

in t

erm

s of

phy

sica

l pr

inci

ples

.

The

prin

cipl

es

that

gov

ern

chem

ical

in

tera

ctio

ns

and

reac

tions

fo

rm t

he b

asis

fo

r a

broa

der

unde

rsta

nd-

ing

of t

he

mol

ecul

ar

dyna

mic

s of

liv

ing

syst

ems.

Psyc

holo

gica

l, so

cioc

ultu

ral,

and

biol

ogic

al

fact

ors

influ

-en

ce t

he w

ays

that

indi

vidu

-al

s pe

rcei

ve,

thin

k ab

out,

an

d re

act

to

the

wor

ld.

Psyc

holo

gica

l, so

cioc

ultu

ral,

and

biol

ogic

al

fact

ors

influ

-en

ce b

ehav

ior

and

beha

vior

ch

ange

.

Psyc

holo

gica

l, so

cioc

ultu

ral,

and

biol

og-

ical

fac

tors

in

fluen

ce

the

way

we

thin

k ab

out

ours

elve

s an

d ot

hers

.

Cul

tura

l an

d so

cial

di

ffer

ence

s in

fluen

ce

wel

l-bei

ng.

Soci

al

stra

tifica

tion

and

acce

ss

to r

esou

rces

in

fluen

ce

wel

l-bei

ng.

Way

s th

at s

tud

ents

use

th

eir

scie

nti

fic

inq

uir

y an

d r

easo

nin

g s

kills

to

so

lve

pro

ble

ms

on

th

e ex

am:

• D

emon

stra

ting

unde

rsta

ndin

g of

sci

entifi

c co

ncep

ts a

nd p

rinci

ples

• Id

entif

ying

the

rel

atio

nshi

ps b

etw

een

clos

ely

rela

ted

conc

epts

• Re

ason

ing

abou

t sc

ient

ific

prin

cipl

es, t

heor

ies,

and

mod

els

• A

naly

zing

and

eva

luat

ing

scie

ntifi

c ex

plan

atio

ns a

nd p

redi

ctio

ns

• D

emon

stra

ting

unde

rsta

ndin

g of

impo

rtan

t co

mpo

nent

s of

sci

entifi

c re

sear

ch

• Re

ason

ing

abou

t et

hica

l iss

ues

in r

esea

rch

• In

terp

retin

g pa

tter

ns in

dat

a pr

esen

ted

in t

able

s, fi

gure

s, a

nd g

raph

s

• Re

ason

ing

abou

t da

ta a

nd d

raw

ing

conc

lusi

ons

from

the

m

9


Biological and Biochemical Foundations of Living Systems

Medical school applicants must be prepared to learn about the biological and biochemical concepts that contribute to health and disease. This section tests applicants’ academic preparation in three foundational concepts that are building blocks for learning in medical school about how:

• The major biochemical, genetic, and molecular functions of the cell support health and lead to disease

• Cells grow and integrate to form tissues and organs that carry out essential biochemical and physiological functions

• The body responds to internal and external stimuli to support homeostasis and the ability to reproduce

What This Section TestsThe Biological and Biochemical Foundations of Living Systems section asks examinees to solve problems by combining their knowledge of foundational concepts from biology, biochemistry, general chemistry, and organic chemistry with their scientific inquiry and reasoning skills. This new section aligns most closely with the Biological Sciences section of the old MCAT exam.

The table below lists the foundational concepts and content categories tested in this section. It also provides examples of the ways examinees are asked to combine their knowledge of foundational concepts with their scientific reasoning skills to answer test questions on the Biological and Biochemical Foundations of Living Systems section.


Foundational Concept 1Biomolecules have unique properties that determine how they contribute to the structure and function of cells, and how they participate in the processes necessary to sustain life.

Foundational Concept 2Highly organized assemblies of molecules, cells, and organs interact to carry out the function of living organisms.

Foundational Concept 3Complex systems of tissues and organs sense the internal and external environ-ments of multicellular organisms, and through integrated functioning, maintain a stable internal environment within an ever-changing external environment.

Content Categories• Structure and functions of protein and

their constituent amino acids• Transmission of genetic information from

the gene to the protein• Transmission of heritable information

from generation to generation and the processes that increase genetic diversity

• Principles of bioenergetics and fuel mole-cule metabolism

Content Categories• Assemblies of molecules, cells, and

groups of cells within singular cellular and multicellular organisms

• The structure, growth, physiology, and genetics of prokaryotes and viruses

• Processes of cell division, differentiation, and specialization

Content Categories• Structure and functions of the nervous

and endocrine systems and ways in which the systems coordinate the organ systems

• Structure and integrative functions of the main organ systems

Questions in this section of the test ask examinees to combine their knowledge of the foundational concepts listed above with their scientific inquiry and reasoning skills. Below are examples:• Recalling the structural characteristics of two tissues and relating them to one another• Applying their understanding of Le Châtelier’s Principle to explain differences in deprotonation of organic acids when added to blood

versus pure water• Using knowledge of adaptive immune response to evaluate the acceptability of a treatment for use in a clinical context• Forming a hypothesis about the effect of the pineal gland on thermogenesis based on the data from an experiment investigating the

interaction of temperature and pineal gland activity on body and organ weights for hamsters under different experimental conditions• Using data about wavelength and light absorption to determine the color perception of an individual with a given phenotype

10


Chemical and Physical Foundations of Biological Systems

Medical school applicants must be prepared to learn about the mechanical, physical, and biochemical functions of human tissues, organs, and organ systems and how these contribute to health and disease. This section tests applicants’ preparation in two foundational concepts that are building blocks for learning in medical school about:

• The physiological functions of the respiratory, cardiovascular, and neurological systems in health and disease

• Molecular and cellular functions in health and disease

What This Section TestsThe Chemical and Physical Foundations of Biological Systems section asks test takers to solve problems by combining their knowledge of foundational concepts from biology, biochemistry, physics, and general and organic chemistry with their scientific inquiry and reasoning skills. This section aligns most closely with the Physical Sciences section of the old MCAT exam.

The table below lists the foundational concepts and content categories tested in this section. It also provides examples of the ways examinees are asked to combine their knowledge of foundational concepts with their scientific inquiry and reasoning skills to answer test questions on the Chemical and Physical Foundations of Biological Systems section.


Foundational Concept 4Complex living organisms transport materials, sense their environ-ment, process signals, and respond to changes using processes that can be understood in terms of physical principles.

Foundational Concept 5The principles that govern chemical interactions and reactions form the basis for a broader understanding of the molecular dynamics of living systems.

Content Categories• Translational motion, forces, work, energy, and equilibrium in

living systems• Importance of fluids for the circulation of blood, gas move-

ment, and gas exchange• Electrochemistry and electrical circuits and their elements• How light and sound interact with matter• Atoms, nuclear decay, electronic structure, and atomic chemi-

cal behavior

Content Categories• Unique nature of water and its solutions• Nature of molecules and intermolecular interactions• Separation and purification methods• Structure, function, and reactivity of biologically relevant

molecules• Atoms, nuclear decay, electronic structure, and atomic chemi-

cal behavior

Questions in this section of the test ask examinees to combine their knowledge of the foundational concepts listed above with their scientific inquiry and reasoning skills. Below are examples:• Identifying the relationship between the distribution of electric charges in the axon and the electric field lines they produce• Recognizing the principles of flow characteristics of blood in the human body and applying the appropriate mathematical model

to an unfamiliar scenario• Changing the experimental conditions of a test for proteins in a solution to prevent the formation of precipitates• Selecting between the standard and Doppler ultrasound techniques for a given context, considering the appropriateness, preci-

sion, and accuracy of each technique• Using, analyzing, and interpreting data in a graph to determine the half-life of a radioactive substance used to measure

cardiac function

11


Psychological, Social, and Biological Foundations of Behavior

Medical school applicants must be prepared to learn about the impact of behavioral and sociocultural factors on illness and health outcomes. This section tests examinees’ academic preparation in five foundational concepts in psychology, sociology, and biology that provide the building blocks for learning in medical school about the ways:

• Cognitive and perceptual processes influence the understanding of health and illness

• Behavior can either support health or increase risk for disease

• Perception, attitudes, and beliefs influence interactions with patients and other members of the health care team

• Patients’ social and demographic backgrounds influence their perceptions of health and disease, the health care team, and therapeutic interventions

• Social and economic factors can affect access to care and the probability of maintaining health and recovering from disease

What This Section TestsThe Psychological, Social, and Biological Foundations of Behavior section tests academic preparation in the behavioral and social sciences. Prior versions of the MCAT exam have not tested this content. This section tests the foundational concepts in psychology, sociology, and biology that tomorrow’s doctors need to serve an increasingly diverse population and have a clear understanding of the impact of behavior and sociocultural differences on health. Like the natural sciences sections, this section asks test takers to solve problems by combining their knowledge of foundational concepts with their scientific inquiry and reasoning skills. It does not measure applicants’ interpersonal skills, the way they will behave, or their attitudes and beliefs about social issues.

The table below lists the foundational concepts tested on this section. It also provides examples of the ways examinees are asked to combine their knowledge of foundational concepts with their scientific inquiry and reasoning skills to answer test questions on the Psychological, Social, and Biological Foundations of Behavior section.


Foundational Concept 6 Psychological, sociocul-tural, and biological fac-tors influence the ways that individuals perceive, think about, and react to the world.

Foundational Concept 7Psychological, socio-cultural, and biological factors influence behavior and behavior change.

Foundational Concept 8Psychological, socio-cultural, and biological factors influence the way we think about ourselves and others.

Foundational Concept 9Cultural and social differences influence well-being.

Foundational Concept 10 Social stratification and access to resources influ-ence well-being.

Content Categories• Sensing the

environment• Making sense of the

environment• Responding to the

world

Content Categories• Individual influences

on behavior• Social processes that

influence human behavior

• Attitude and behavior change

Content Categories• Self-identity• Social thinking• Social interactions

Content Categories• Understanding social

structure• Demographic charac-

teristics and processes

Content Categories• Social inequity

Questions in this section of the test ask examinees to combine their knowledge of foundational concepts listed above with their scientific inquiry and reasoning skills. Below are examples:• Drawing conclusions about the type of memory affected by an experimental manipulation when shown a graph of findings from a

memory experiment• Reasoning about whether a causal explanation is possible when given an example of how personality predicts individual behavior• Distinguishing the kinds of claims that can be made when using longitudinal data, cross-sectional data, or experimental data in studies

of social interaction• Identifying the relationship between demographic variables and health variables reported in a table or figure• Identifying the relationship between social institutions that is suggested by an illustration used in a public health campaign

12


Do scores from the Psychological, Social, and Biological Foundations of Behavior section add value to admissions decisions?

Eleven medical schools are partnering to study the Psychological, Social, and Biological Foundations of Behavior section. More than 2,000 medical students from these schools took a prototype version of this section before the start of the fall 2013 semester. Schools then collected participants’ grades in medical school courses related to the behavioral and social sciences.

Scores were correlated with participants’ grades in these courses.

These early results show that scores from the Psychological, Social, and Biological Foundations of Behavior section predict medical students’ grades in courses such as Foundations of Psychiatric Medicine, Neuroscience, and Epidemiology/Public Health and predict grades in these courses better than scores from the old MCAT exam. These results suggest that these scores add value to student selection decisions. Refer to Appendix A for more information about the study and results.

13


Critical Analysis and Reasoning Skills

The Critical Analysis and Reasoning Skills section has a different organizing structure than the other sections of the exam. It tests how well test takers comprehend, analyze, and evaluate what they read, draw inferences from text, and apply arguments to new ideas and situations. It asks applicants to process information, draw conclusions, and solve problems from information that is presented in passages. This section aligns most closely with the Verbal Reasoning section of the old MCAT exam.

This section tests examinees’ ability to process information by reading passages from a diverse set of disciplines in the humanities and social sciences. However, the questions on the Critical Analysis and Reasoning Skills test do not rely on specific background knowledge in the humanities and social sciences. Applicants get all the information they need to answer the questions in the accompanying passages or in the questions themselves.

The new exam gives increased attention to passages about population health, studies of diverse cultures, ethics, and philosophy to encourage applicants to read broadly and familiarize themselves with the issues and arguments these disciplines raise.

Questions in this section of the test ask examinees to demonstrate their information-processing skills by:

• Understanding the basic components of the text

• Inferring meaning from rhetorical devices, word choice, and text structure

• Integrating different components of the text to increase comprehension

• Applying or extrapolating ideas from the passage to new contexts

• Assessing the impact of introducing new factors, information, or conditions to ideas from the passage

14


Scores from the New MCAT Exam

This section of the guide provides information about interpreting scores from the new exam. It contains information about the new score scale and use of percentile ranks, confidence bands, and score profiles to make decisions about applicants in the 2016 application cycle.

Scores are reported on different scales because the new and old exams test different things.

The new score scales were developed to have nice round, memorable numbers in the center and scales that can’t be confused with the scales from the old exam, percent-correct scales, or other commonly used score scales. The four section scores are centered at 125 and range from 118 to 132. Scores from the four sections are summed to produce a total score that is centered at 500 and ranges from 472 to 528.

The new scores draw attention to the center of the scales and the top half of the distribution to encourage admissions committees to consider applicants with a wider range of scores than they have in the past. On the old exam, the average total score for accepted applicants was 31 and ranged from 3 to 45. Many selection decisions focused on the top third of the old score scale, but research on the old exam suggests that the students who enter medical school with scores at the center of the scale succeed; they graduate in four or five years and pass their licensing exams on the first try (Dunleavy et al. 2013).

472 528

500Total Score

+ + + =

Section Scores Sum to the Total Score

118 132

125


118 132

125


118 132

125


118 132

125


15


The new scores are more reliable.

As we have already mentioned, the new section scores are more reliable than the section scores from the old exam because the new exam has more questions per section. They provide more precise information about test takers’ academic preparation and provide better pictures of examinees’ strengths and weaknesses across the four test sections.

The new score report provides more information.

You will receive a lot of information about applicants’ preparation from the new score reports, which describe applicants’ scores using percentile ranks, confidence bands, and score profiles.

This is what the new score report looks like:

16


Using Percentile Ranks in 2016 Admissions

Because the new exam uses new score scales, admissions committees will essentially have to “start over,” attaching meaning to the scores and developing new conventions for selecting students. To help with this process, the new score reports will provide percentile ranks—one for each section score and one for the total score.

In the beginning, the percentile ranks will provide the clearest information about applicants’ work on the new exam. The percentiles show how individual test takers compare with the population of test takers who prepared for and took the new exam. They represent the percentage of examinees at or below a given score on the exam. Scores with higher percentile ranks represent higher levels of performance on the exam.

As you gain experience working with new MCAT scores, you will begin to attach meaning to new scores (just as scores from the old test have meaning for you) by observing the academic performance of students admitted with different MCAT scores. You will begin to see which scores spell success in your program. As you gain this experience, you are likely to start using the scores more than the percentile ranks. Eventually, you will find that confidence bands and score profiles provide helpful information about the scores themselves. We will talk about that next.

Developing Percentile Rank Tables for the 2015 Testing Year

The percentile ranks reported for the 2015 testing year are based on the scores of examinees who took the exam on the two April and May test dates (Figure 1). After careful examination of the April and May 2015 scores and based on historic MCAT data, the AAMC made adjustments so that the distribution of scores reported in these percentile rank tables reflects the best estimate of the score distribution that would result in a typical testing year. The AAMC carefully developed these adjustments using a wide range of data about past examinees.

These percentile rank tables will allow you to compare the performance of April and May examinees with the performance of other 2015 examinees.

The AAMC will use these same percentile ranks for all the scores reported in the 2015 testing year—that is, all scores from the 2015 testing year will be attached to the same set of percentile ranks.

17


Figure 1. Summary of total and section scores from the MCAT exam based on results for tests administered in April and May 2015.

MCAT Total Score

MCAT Total PercentileScore Rank Mean = 500.0528 100 Std. Deviation = 10.6527 100526 100525 100524 100523 100522 99521 99520 98519 98518 97517 96516 95515 93514 91513 89512 87511 85510 83509 80508 77507 74506 71505 68504 64503 61502 57501 54500 51499 47498 43497 40496 37495 34494 31493 28492 25491 22490 20489 18488 15487 13486 11485 10484 8483 7482 6481 5480 4479 3478 2477 1476 1475 1474 0473 0472 0

MCAT Total Score

0

1

2

3

4

5

472 474 476 478 480 482 484 486 488 490 492 494 496 498 500 502 504 506 508 510 512 514 516 518 520 522 524 526 528

Percent

MCAT Total Score

NotesThe percentile ranks in these tables are based on MCAT results from examinees who tested in April and May 2015. After careful analysis of the April and May 2015 scores and based on historic MCAT data,

made adjustments so that the distribution of scores reported in these tables reflects the best estimate of the score distribution that would result in a typical testing year. carefully developed these

adjustments using a wide range of data about past examinees.

The column labeled “Percentile Rank” provides the percentage of results in a typical testing year equal to or less than each score point.1

For example, 77 percent of MCAT scores are likely to be equal to or less than 508 across all exams administered in a typical year.

1Prior versions of these tables (pre‐2012), as well as current score reporting systems maintained by AMCAS report the "Percentile Rank Range" in these columns rather than a single percentile rank. The percentile rank columns in the current tables correspond to the upper bound of the percentile rank ranges provided in these other versions. The lower bounds in these other versions essentially serve to provide continuity across score points.

The AAMC

total

the

,

AAMC

(continued on next page)

18



Section PercentileScore Rank132 100131 99 Mean = 125.0130 97 Std. Deviation = 3.0129 93128 87127 79126 67125 55124 45123 32122 21121 12120 7119 3118 1


Section PercentileScore Rank132 100 Mean = 124.9131 99 Std. Deviation = 3.0130 97129 93128 87127 81126 70125 58124 44123 33122 22121 14120 8119 3118 1





0

2

4

6

8

10

12

14

Percent

Chemical and PhysicalFoundations of Biological Systems

Section Score

0

2

4

6

8

10

12

14

Percent

Critical Analysis and Reasoning SkillsSection Score

0

2

4

6

8

10

12

14

Percent

Biological and BiochemicalFoundations of Living Systems

Section Score

0

2

4

6

8

10

12

14

Percent

Psychological, Social, and BiologicalFoundations of Behavior

Section Score

Psychological, Social, and BiologicalFoundations of BehaviorSection Score

Biological and BiochemicalFoundations of Living SystemsSection Score

Critical Analysis and Reasoning SkillsSection Score

Chemical and PhysicalFoundations of Biological SystemsSection Score

Figure 1. Summary of total and section scores from the MCAT exam based on results for tests administered in April and May 2015. (Continued)

19


Using Confidence Bands in 2016 Admissions

Like scores from other standardized exams, MCAT scores are not perfectly precise. The new MCAT score report includes confidence bands that describe the precision of applicants’ total and section scores and show the ranges in which examinees’ true scores probably lie. Score reports show confidence bands both numerically and graphically.

MCAT total scores are reported with a confidence band of plus or minus two points, and MCAT section scores are reported with confidence bands of plus or minus one point. The example below shows a total score of 504 and its confidence band, which ranges from 502 to 506. The diamond shape shows the confidence band graphically and indicates that the best estimate of an applicant’s true score is in the center of the diamond, where the diamond is the tallest and the shading is the darkest.

Confidence bands remind you not to overemphasize small differences in scores.

When comparing two applicants’ scores, it is important to consider how much their confidence bands overlap. The greater the overlap in confidence bands, the less meaningful are the differences between the scores.

This chart below displays four pairs of MCAT total scores with confidence bands that differ in their degree of overlap. The top pair of scores, 504 vs. 503, is only one point apart on the total score scale. The chart shows that the confidence bands for these scores overlap on four score points—502, 503, 504, and 505.

In comparison, the bottom pair of scores, 504 vs. 500, is four points apart on the MCAT total score scale and has confidence bands that only overlap on one point on the total score scale—502—indicating that these scores are less comparable to each other than the first pair. Scores that are closer together have confidence bands that overlap.

Score Confidence Band

Percentile Rank of Score

MCAT Total Score

504 502 506 64%

504 v 503

504 v 502

504 v 501

504 v 500

498 506505504503502501500499

Applicants' Scores

20


Using Score Profiles in 2016 Admissions

Score profiles highlight applicants’ strengths and weaknesses across the four sections of the exam through reported scores for each section. The strengths and weaknesses on the exam can be considered with other information about applicants’ academic preparation and in relation to your institution’s missions and goals.

Remember that each of the four sections of the new exam includes more test questions and reports more reliable scores than the old exam. More reliable test scores provide more precise information at the section level and support stronger comparisons across sections.

MCAT® is a program of the Association of American Medical Colleges

Medical College Admission Test

What will a score profile tell me?

Score profiles highlight applicants’ strengths and weaknesses across the four sections of the exam through reported scores for each section. The strengths and weaknesses on the exam can be considered with other information about applicants’ academic preparation and in relation to your institutions’ missions and goals.

Remember that each of the four sections of the new exam includes more test questions than the sections of the old exam. More questions means more reliable scores. More reliable test scores provide more precise information at the section level and support stronger comparisons across sections. To learn more about what academic competencies each section tests, download the Testing Academic Competencies with the 2015 MCAT Exam.

s

MCAT Score Profile

Section Score Profile1





125118 132

125118 132

125118 132

125118 132

Looking at an Applicant’s MCAT Score Profile

1 For the four sections, non-overlapping confidence bands show a test taker’s likely strengths and weaknesses. Overlapping confidence bands suggest that there are not meaningful differences in performance between sections.

21


Making Decisions about Applicants in the 2016 Admissions Cycle

In the 2016 admissions cycle, we expect that about half of the applicants will submit scores from the new exam, and about half will report scores from the old exam. Your applicants likely were encouraged to take the MCAT exam when they were best prepared to test. So, for some, that meant taking the old exam, and for others, it meant taking the new exam.

Designing Thresholds for Screening and Interviewing

If you have traditionally used a threshold for deciding which applicants should receive secondary applications or which should be invited to interview, you may want to continue that practice. If MCAT scores are among the criteria you use to decide who should receive secondary applications or be invited to interview, the percentile ranks for the old exam provide a good starting point for setting thresholds on the new exam, but more thought is needed.

Applicants reporting scores from the new exam prepared in biochemistry, psychology, and sociology and practiced using the scientific inquiry and reasoning skills tested on the new exam. The old percentile ranks do not reflect that preparation. Especially in this first year, you are encouraged to use your previous thresholds as a starting point, and then lower them a little so you don’t miss applicants who might be good candidates for your school. For example, if in the past, you invited applicants to complete secondary applications (or to interview) if they scored at the 60th percentile on the old exam, you might want to look at applicants who score at the 50th percentile on the new MCAT exam.

If the course taking, grades, and other information about the academic preparation of these applicants look good to you, this lower percentile rank might be a good threshold for your 2016 screens using the new exam. Looking at applications below your historic thresholds may help you find candidates who meet your mission and goals.

2016 Applicant Pool

Evaluate applicants with scores from the old exam the way you always have.

Evaluate applicants with scores from the new exam using percentile ranks.

Evaluate applicants with scores from both exams using new scores and their percentile ranks.

examinee with scores from old exam

=

examinee withscores from newexam

=

examinee with scores from both exams

=

22


Making Decisions about Individual Applicants

New MCAT Scores Only As you do now for individual applicants, you are encouraged to consider applicants’ percentile ranks on the new exam in combination with information about their course completion, grades, grade trends, institutional selectivity, research experience, and other academic indicators. Look for consistencies and inconsistencies in the stories these data tell.

If an applicant with scores on the new MCAT exam seems otherwise comparable to an applicant with scores from the old MCAT exam, think about the importance to success in your curriculum of competence in biochemistry, psychology, sociology, and scientific problem-solving skills to decide how much weight to give the new scores in selection.

New and Old MCAT Scores For applicants with scores from the new and old exams, you are encouraged to use the scores from the new exam. These applicants likely were dissatisfied with their old scores and hoped to improve their applications by taking the new test. If the different sets of scores are discrepant, the differences may be explained by information about the courses and other preparation completed between the two versions of the test.

Old MCAT Scores Only You already know what these scores mean. You know which scores predict different levels of success in your program. So, you can think about the scores for these applicants in the way you have in the past.

Applicants who test with accommodations that were previously annotated will not have their scores annotated in applications processed for the 2016 admissions cycle. The change, which took effect in late March 2015, applies to scores from the new and old MCAT exams.

23


Data about the New MCAT Exam

Timeline for Reporting Data about the New Exam

The AAMC is committed to providing data about the new exam as soon as possible. The percentile tables in this book were estimated using data from examinees who tested in April and May 2015. Again, in creating these tables, the data for April and May examinees were weighted to represent the scores we would see in the typical testing year. These tables will be used for the entire 2016 application cycle. In fall 2015, these percentile tables will be updated with examinees’ performance for the entire 2015 testing year for the new exam, and will be available for use in the 2017 application cycle.

The following data will be available for use in selecting the entering class of 2017 and will be included in next year’s version of this guide:

• Updated percentile rank tables for the 2015 testing year

• MCAT total and section score means and standard deviations

• MCAT total scores by sex and race/ethnicity

• Changes in MCAT total scores for applicants who test more than once

• Acceptance rates for 2016 applicants by scores from the new MCAT exam and undergraduate GPAs

MCAT Validity Research Agenda

As the new exam matures, more and more data will be available to help you attach meaning to scores and develop conventions for selecting students. As the AAMC did with the old MCAT exam, the AAMC will use national data to study the validity of the new MCAT exam in predicting students’ academic performance throughout medical school. The AAMC has established two research projects to collect early evidence about the value of MCAT scores in predicting matriculants’ academic success in medical school: the Psychological, Social, and Biological Foundations of Behavior Validity Study and the Longitudinal MCAT Validity Study. The table that follows lists schools participating in this research.

24


Participating School

Psychological, Social, and Biological Foundations of Behavior Validity Study

Longitudinal MCAT Validity Study

Boston University School of Medicine ✔ ✔

Columbia University College of Physicians and Surgeons ✔ ✔

East Tennessee State University James H. Quillen College of Medicine ✔

Meharry Medical College ✔ ✔

Memorial University of Newfoundland Faculty of Medicine ✔ ✔

Morehouse School of Medicine ✔

Philadelphia College of Osteopathic Medicine ✔

Rutgers Robert Wood Johnson Medical School ✔

Saint Louis University School of Medicine ✔

Stanford University School of Medicine ✔

The Ohio State University College of Medicine ✔

The University of Texas School of Medicine at San Antonio ✔ ✔

Tulane University School of Medicine ✔

University of Arizona College of Medicine ✔

University of Calgary Faculty of Medicine ✔

University of California, San Francisco, School of Medicine ✔ ✔

University of Central Florida, College of Medicine ✔

University of Illinois College of Medicine ✔ ✔

University of Mississippi School of Medicine ✔ ✔

University of North Carolina at Chapel Hill School of Medicine ✔

Uniformed Services University of the Health Sciences F. Edward Herbert School of Medicine ✔

Prehealth advisers from Colgate University and The University of Hawaii at Manoa are members of the committee responsible for setting the MCAT validity research agenda.

The Psychological, Social, and Biological Foundations of Behavior Validity Study

Eleven medical schools are currently participating in a validity research project that is examining and documenting the value of the Psychological, Social, and Biological Foundations of Behavior section of the new MCAT exam. These schools are collecting validity data for 2012 and 2013 entrants into medical school. They are examining the relationships between scores from this section and grades in medical school courses and clerkships conceptually related to the behavioral and social sciences, as well as United States Medical Licensing Exam (USMLE) Step exam scores. See Appendix A for a summary of this study’s interim results.

To date, participating medical schools have correlated PSBB scores for their students with their students’ performance on more than 18 medical school courses, such as Foundations of Psychiatric Medicine, Epidemiology/Public Health, and Neuroscience. These preliminary results suggest that Psychological, Social, and Biological Foundations of Behavior scores predict students’ performance in medical school courses related to the behavioral and social sciences. Validity data will continue to be reported throughout this project, with final analyses to occur in winter 2016.

25


The Longitudinal MCAT Validity Study

Eighteen medical schools are participating in a second validity research project that will examine and document the value of the new MCAT exam. They will collect validity data for 2016 entrants as soon as they are available. Admissions officers from participating schools will examine the relationships between MCAT scores and grades in medical school courses and clerkships, USMLE Step exam scores, time to graduation, and graduation rates. Validity data will be reported on a regular basis beginning with data about the relationship between MCAT scores and academic performance in first-semester courses. The medical schools will also study the impact of the new exam on examinees and applicants, as well as issues related to test fairness.

National MCAT Validity Data

National MCAT validity data for the new MCAT exam will be published in future editions of this guide when they become available. These data will show the relationships between scores on the new MCAT exam, undergraduate GPAs, and national outcomes for students who started medical school in 2016. Outcome data will include four- and five-year graduation rates, withdrawal or dismissal from medical school for academic reasons, and USMLE Step exam pass rates.

Here is a timeline showing the MCAT validity research agenda:

Expected Year of Publication

2016 2017 2018 2019 2020 2021 2022

The Psychological, Social, and Biological Foundations of Behavior Validity Study

• Grades in behavioral and social science pre-clerkship courses

• Grades in behavioral and social science clerkships

• Step 1 behavioral and social science subscores

• Step 2 Clinical Knowledge behavioral and social science subscores

The Longitudinal MCAT Validity Study

• Grades in pre-clerkship courses

• Grades in clerkships

• National Board of Medical Examiners (NBME) subject exam scores

National MCAT Validity Data

• Academic withdraw or dismissal

• Step 1 pass/fail and scores

• Step 2 Clinical Knowledge pass/fail and scores

• Graduate in four years

• Step 2 Clinical Skills pass/fail and scores

• Graduate in five years

References

Admissions Initiative (2013). 2013 Study of Medical School Admissions Policies and Practices. Unpublished raw data.

Dunleavy, D.M., Kroopnick, M.H., Dowd, K.W., Searcy, C.A., and Zhao, X. (2013). The predictive validity of the MCAT exam in relation to academic performance through medical school: A national cohort study of 2001–2004 matriculants. Academic Medicine, 88(5), 666–671.

Schwartzstein, R.M., Rosenfeld, G.C., Hilborn, R., Oyewole, S.H., and Mitchell, K. (2013). Redesigning the MCAT Exam: Balancing multiple perspectives. Academic Medicine, 88(5), 560–567.

26


Appendix A: The MCAT Psychological, Social, and Biological Foundations of Behavior Validity Study Results

Eleven medical schools are partnering to study the Psychological, Social, and Biological Foundations of Behavior section of the new MCAT exam. These schools reflect a mix of educational missions and goals, have different applicant pool sizes and demographic compositions, include both public and private schools, and are located in diverse geographic regions.

At the start of the fall 2013 semester, 2,093 medical students (mostly first- and second-year students) from these 11 schools volunteered to take a prototype version of the Psychological, Social, and Biological Foundations of Behavior section. These schools are collecting students’ grades and test scores from courses related to the behavioral and social sciences taught in the first three years of medical school. This section reports interim results from the first two years of this study.

Psychological, Social, and Biological Foundations of Behavior scores correlate with academic performance in medical school courses such as

Foundations of Psychiatric Medicine, Epidemiology/Public Health, and Neuroscience.

Figure 2 shows median correlations between scores from the Psychological, Social, and Biological Foundations of Behavior section and students’ academic performance in Foundations of Psychiatric Medicine, Epidemiology/Public Health, and Neuroscience medical school courses. These correlations exceeded the standard for a medium effect in the social sciences (Cohen, 1988).

Figure 3 compares the median correlations between the Psychological, Social, and Biological Foundations of Behavior section and old MCAT sections and students’ academic performance in these courses. These data suggest that the new and old sections of the MCAT exam provide different information.

0.41

0.40

0.37

0.0 0.1 0.2 0.3 0.4 0.5

Neuroscience(6 courses)

Epidemiology/Public Health(5 courses)

Foundations of Psychiatric Medicine(7 courses)

Figure 2. Median correlations between the Psychological, Social, and Biological Foundations of Behavior section scores and students’ academic performance in three types of courses.

Medium Effect

Median Correlation

27


0.37

0.34

0.34

0.35

0.0 0.1 0.2 0.3 0.4 0.5

Verbal Reasoning

Physical Sciences

Biological Sciences


Neu

rosc

ien

ce(6

co

urs

es)

Medium Effect

Median Correlation

0.40

0.22

0.20

0.20

0.0 0.1 0.2 0.3 0.4 0.5

Verbal Reasoning

Physical Sciences

Biological Sciences


Epid

emio

log

y/Pu

blic

Hea

lth

(5 c

ou

rses

)

Medium Effect

Median Correlation

0.41

0.30

0.12

0.22

0.0 0.1 0.2 0.3 0.4 0.5

Verbal Reasoning

Physical Sciences

Biological Sciences


Fou

nd

atio

ns

of

Psyc

hia

tric

Med

icin

e(7

co

urs

es)

Medium Effect

Median Correlation

Figure 3. Comparison of median correlations between new and old MCAT section scores and students’ academic performance in three types of courses.

Medium Effect

28


Limitations

This study is not without limitations. It was conducted on a relatively small sample of schools, and data were collected for research purposes. Participants likely were less motivated and prepared than they would have been when taking the MCAT exam for admission into medical school. In addition, this research was conducted on a prototype version of the Psychological, Social, and Biological Foundations of Behavior section. In light of these limitations, the results are promising and suggest that this section may provide useful information for schools that include behavioral and social sciences content in their curricula.

Next Steps

Over the next two years, the Psychological, Social, and Biological Foundations of Behavior validity committee will continue to investigate the relationship between Psychological, Social, and Biological Foundations of Behavior scores and academic performance in Foundations of Psychiatric Medicine, Epidemiology/Public Health, Neuroscience, and other courses that include behavioral and social sciences content. Data also will be collected from clerkships and the Step exams.

Reference

Cohen, J. Statistical Power Analysis for the Behavioral Sciences. 2nd edition. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.

29


Appendix B: Sample Questions for the New MCAT Exam

This appendix provides examples of questions for each of the four test sections of the new MCAT exam. For the natural, behavioral, and social sciences, the examples show how applicants have to use their scientific reasoning skills to solve problems that rely on the 10 competencies and their supporting concepts. For each example, a passage provides the context for the questions. Some questions also rely on information presented in tables, charts, or graphs.

For each of the four sections, a question is shown that is easy for most MCAT test takers to answer, a question that is medium difficulty, and a question that is hard. Each question includes an explanation of the things that test takers need to know and do to get the question right.

Example: Biological and Biochemical Foundations of Living Systems

The myocellular transmembrane Na+ gradient is important for proper cellular function. During septic shock, disruption of Na+ homeostasis often occurs and leads to decreased membrane potential and increased intracellular Na+. It has been found that failure of cellular energy metabolism is a common symptom in septic patients who do not respond to therapeutics. Because normal intracellular levels of Na+ are maintained by the Na+K+ ATPase, it is important to understand how metabolic energy production is linked to cation transport.

Researchers are interested in whether the energy used for ion transport is derived from glycolysis or oxidative phosphorylation. This information would provide a better understanding of myocellular damage that occurs during critical illness. Experiments were conducted to evaluate the effects of glycolytic inhibition on cellular Na+ and K+ concentrations and lactate production in rat skeletal myocytes.

Rat skeletal muscle fibers were extracted and incubated in normal media (control), glucose-free media (G(–)), and glucose-free media with various concentrations of the glycolytic inhibitor iodoacetate (IAA). IAA directly prevents the formation of 1,3-bisphosphoglycerate. After one hour in the media, the muscle tissues were assayed for intracellular Na+ and K+ content and lactate production. Cellular viability was determined by measuring the amount of lactate dehydrogenase (LDH) released, as LDH release is an indicator of cell death. The results are displayed in Figure 1.

Figure 1. Effects of glycolytic inhibition on intracellular Na+ and K+ content and lactate production with cellular viability measured by LDH release. (Note: The * indicates p < 0.05 versus control.)

The researchers also examined the effect disruption of oxidative phosphorylation had on Na+ and K+ content. Inhibition of oxidative phosphorylation was caused by carbonyl-cyanide m-chlorophenylhydrazone (CCCP), an ionophore that allows protons to move freely through membranes. No correlation between Na+ and K+ content and oxidative phosphorylation was found.

Adapted from: Okamoto K, Wang W, Rounds J, Chambers EA, Jacobs DO. ATP from glycolysis is required for normal sodium homeostasis in resting fast-twitch rodent skeletal muscle. The American Journal of Physiology-Endocrinology and Metabolism. 2001 Sept;281(3):E479-88.

3

0

1

2

[Na+

]/[ K

+]

Control G(–) 0.1 0.3

*

1 2

IAA (mM)

10

0

5

Control G(–) 0.1 0.3 1 2

IAA (mM)

8

0

2

4

6

LD

H r

elea

se(μ

mol

• m

in–1

• g

–1)

Control G(–) 0.1 0.3 1 2

IAA (mM)

**

*

*

** *

**

Lac

tate

pro

duct

ion

(μm

ol •

h–1

• g

–1)

3

0

1

2

[Na+

]/[ K

+]


*

1 2

IAA (mM)

10

0

5

Control G(–) 0.1 0.3 1 2

IAA (mM)

8

0

2

4

6

LD

H r

elea

se(μ

mol

• m

in–1

• g

–1)

Control G(–) 0.1 0.3 1 2

IAA (mM)

**

*

*

** *

**

Lac

tate

pro

duct

ion

(μm

ol •

h–1

• g

–1)

3

0

1

2

[Na+

]/[ K

+]


*

1 2

IAA (mM)

10

0

5

Control G(–) 0.1 0.3 1 2

IAA (mM)

8

0

2

4

6

LD

H r

elea

se(μ

mol

• m

in–1

• g

–1)

Control G(–) 0.1 0.3 1 2

IAA (mM)

**

*

*

** *

**

Lac

tate

pro

duct

ion

(μm

ol •

h–1

• g

–1)

30


Question 1. The researchers chose a concentration of 0.3 mM IAA as the working concentration for any additional studies instead of 1 mM or 2 mM. What is the likely reason for this?

A) The lower concentration of IAA gave the largest Na+ response. B) Higher concentrations induced significant cytotoxicity. C) The solubility of IAA was not high enough. D) The researchers were trying to mimic control conditions as closely as possible.

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #2. It asks them to apply their understanding of the ways in which highly organized assemblies of molecules, cells, and organs interact to carry out the functions of living organisms. This question asks test takers to reason about the design and execution of research using their knowledge of cytotoxicity and cell lysis. To answer this question, test takers have to know that conducting an experiment where the level of IAA is cytotoxic to the cells (when compared to control conditions) will not help them understand the role of glycolysis in establishing ion concentration gradients, as these cells will lose membrane integrity and undergo lysis. Therefore, the experimental design should not use an IAA concentration that results in significantly increased cell lysis. B is the correct answer to this question.

Question 2. The information in the passage suggests that glycolysis:

A) is important for maintaining normal Na+ and K+ levels in skeletal muscle. B) facilitates membrane permeability in skeletal muscle. C) impedes the function of the Na+K+ ATPase in skeletal muscle. D) is regulated by the Na+K+ ATPase in skeletal muscle.

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #1. It calls on their understanding of the unique properties of biomolecules that determine how they contribute to the structure and function of cells and how they participate in the processes necessary to maintain life. This question asks test takers to reason about data using their knowledge of glycolysis. To get this question right, test takers have to recognize that the data trend in Figure 1, which shows increasing concentration of IAA results in a higher ratio of the concentration of Na+ to K+ than observed in the control sample, must be correlated with the role of IAA in the disruption of glycolysis. This is further supported by the drop in lactate production shown in Figure 1 at higher concentrations of IAA, because IAA prevents the formation of NADH, which is used when pyruvate is reduced to lactate. The combination of the proposed role of IAA and the results from Figure 1 should lead test takers to the conclusion that glycolysis is important to the Na+K+ ATPase and, therefore, important to the maintenance of the concentration ratio of Na+ to K+. A is the correct answer to this question.

This question is of medium difficulty for

most test takers

This question is easy for most test takers

31


Question 3. If the effects of IAA treatment in nerve cells are the same as those observed in myocytes, which feature of an action potential would be most affected by IAA treatment?

A) Initiation of depolarization B) Rising phase of depolarization C) Falling phase to undershoot D) Return to resting potential

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #3. It asks them to apply their understanding of the ways in which complex systems of tissues and organs sense the internal and external environments of multicellular organisms, and through integrated functioning, maintain a stable internal environment within an ever-changing external environment. This questions asks test takers to combine their scientific reasoning and problem-solving skills with their knowledge of the role of the Na+K+ ATPase in the recovery of the nerve cell resting potential after an action potential. To get this question right, test takers have to use information from the passage about the effect of IAA treatment to reason about the way in which inhibition of glycolysis by IAA will affect the cellular concentration of ATP. This question asks test takers to propose a hypothesis about which portion of an action potential would be affected by IAA treatment. D is the correct answer to this question.

Example: Chemical and Physical Foundations of Biological Systems

The heme enzyme indoleamine 2,3 dioxygenase (IDO) catalyzes Reaction 1, the first and rate-determining step of L-tryptophan (Compound 1) metabolism, and is an important enzyme of the human immune system.

Reaction 1

The IDO-catalyzed oxidation of Compound 1 by H2O2 does not occur. However, researchers have recently discovered that IDO-catalyzed oxidation of indole (Compound 3) by H2O2 (Reaction 2) does occur.

Reaction 2

CO2

NH3+NH

NH

O

O CO2

NH3

L-Trp N-FormylkynurenineCompound 1 Compound 2

IDO/O2 +

NH

IDO/H2O2

Compound 3

NH

O

NH

O

Compound 4 Compound 5

Compound 6

+

+NH

O

O

This question is hard for most test takers

32


Under the conditions employed, the number of catalytic turnovers appeared to stop at roughly 100, on average. A plot of the concentration of Compound 3 that was oxidized versus the concentration of H2O2 employed, at two different initial concentrations of IDO, gave the results shown in Figure 1.

Figure 1 Stoichiometry of IDO-catalyzed oxidation of Compound 3 by H2O2 at 1 µM (dashed line) and 10 µM (solid line) IDO

Aerobic oxidation of Compound 3 in the presence of 18O-labeled H218O2 resulted in the formation of 18O-labeled oxidation

products (Table 1).

Table 1 Isotopic Composition of Compound 3 Oxidation Products Using H218O2

The formation of Compound 6 does not appear to be the result of a sequential oxidation process. Isotopically labeled Compound 4 does not exchange 18O for 16O in water over 3 hours, but Compound 6 completely loses its 18O label in unlabeled water over the same time period.

Adapted from: Kuo HH, Mauk AG. Indole peroxygenase activity of indoleamine 2,3-dioxygenase. Proceedings of the National Academy of Sciences of

the United States of America. 2012;109(35):13966–71.

0.2

0.4

0.0[Com

poun

d 3]

oxi

dize

d (m

M)

0.0 0.2 0.4

[H2O2] (mM)

ProductPercentage of 18O incorporated (%)

Mono-18O Di-18O

Compound 4 100 —

Compound 5 100 —

Compound 6 60 40

33


Question 1. The progress of Reaction 2 can be monitored by observing what change to the IR spectrum of the product mixture?

A) Appearance of a broad peak at 3400 cm–1

B) Disappearance of a broad peak at 3400 cm–1 C) Appearance of a sharp peak at 1700–1750 cm–1 D) Disappearance of a sharp peak at 1700–1750 cm–1

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #4. It asks them to apply an understanding of the ways in which complex living organisms transport materials, sense their environment, process signals, and respond to changes using processes that can be understood in terms of physical principles. It asks test takers to combine their scientific reasoning and problem-solving skills with their knowledge of infrared spectroscopy. To get this question right, test takers have to understand the structural differences between the products and reactants of Reaction 2 and apply the scientific model of the differences in IR absorbance of various functional groups to the experiment described in the passage. Recognition of the presence of additional carbonyl groups in the products of the reaction should lead test takers to conclude that the appearance of a peak between 1700–1750 cm–1 in the IR spectrum will provide the most effective way to monitor product formation. C is the correct answer to this question.

Question 2. The following kinetic parameters were obtained for the IDO-catalyzed oxidation of Compound 3 by H2O2 in the presence of L-Trp.

Based on these data, what effect does L-Trp have on the reaction?

A) L-Trp oxidizes Compound 3 directly. B) L-Trp is oxidized instead of Compound 3. C) L-Trp does not interact with the enzyme. D) L-Trp inhibits the enzyme.

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #5. It asks them to use the principles that govern chemical interactions and reactions and demonstrate a broad understanding of the molecular dynamics of living systems. This question asks test takers to reason about data using their knowledge of enzyme kinetics. To get this question right, test takers have to understand what the decreasing values of kcat in the presence of higher concentrations of L-Trp mean with respect to the kinetics of IDO-catalyzed indole oxidation. The kcat

is representative of the rate of product turnover, which means that the enzyme produces less product in the presence of L-Trp. Combining this trend in the data with knowledge of enzyme kinetics should lead test takers to the conclusion that L-Trp is inhibiting the reaction. D is the correct answer to this question.

[L-Trp], µM kcat, s–1

0 1.3

2 0.34

5 ~0


most test takers


34


Question 3. Which experiment can be used to show that Compound 6 is not formed sequentially from either Compound 4 or Compound 5?

A) Conduct the reaction of Compound 4 with Compound 5, and identify the products. B) Oxidize Compound 4 and Compound 5 with IDO/H2O2, and identify the products. C) Reduce pure Compound 6 without added catalyst, and identify the products. D) Conduct the reaction of Compound 2 with H2O2 without added catalyst, and identify the products.

Like the last question, this question asks test takers to demonstrate their mastery of Competency #5. Drawing from principles that govern chemical interaction and reactions, this question asks test takers to reason about the design and execution of research using their knowledge of the way that enzymes catalyze reactions. To get this question right, test takers have to know how researchers can be sure that Compound 6 is not formed from either Compound 4 or Compound 5 in a sequential enzyme mechanism. Enzymes are not used up during catalysis, so any experiment that includes just Compound 4 or just Compound 5 would determine if either is also a substrate for IDO-catalyzed conversion to Compound 6. Having both compounds in solution with IDO adds unnecessary complexity to the interpretation of the experimental results. Examining the products of IDO-catalyzed reduction of Compound 6 does not give the necessary direct evidence, as Compound 6 could be sequentially reduced to Compound 3. B is the correct answer to this question.


35


Example: Psychological, Social, and Biological Foundations of Behavior

In Study 1, a study on stereotype threat, a group of men and women were randomly assigned to take a standardized math assessment under three different conditions. The conditions differed in terms of the information that was given to the participants before they started working on the task. The problem-solving group was told that the math assessment was a problem-solving task. The math assessment group was told that their scores were going to be used to study gender differences in mathematical ability. The teaching intervention group was given the same information as the math assessment group, but they also were instructed on what stereotype threat is and how it may affect performance. Figure 1 summarizes the results of this study.

Figure 1 Average percent correct on the math assessment by group

Anxiety has often been considered an explanation for stereotype threat. Study 2 was conducted to explore the role of anxiety on cognitive performance. Participants were required to memorize lists of target words. During the training phase, before each target word was presented to them, the participants were asked to read a sentence out loud. Half of the participants read sentences containing words that triggered self-doubt, and the other half read sentences containing words that triggered self-confidence. In the retrieval phase, the participants were asked to report as many of the target words as they could remember. The participants remembered more words under the self-confidence condition than the self-doubt condition. As part of the design of Study 2, the researchers also measured electrical skin conductance and salivary cortisol levels. They found that the participants in the self-doubt condition had higher levels of skin conductance and cortisol than those in the self-confidence condition.

Adapted from: Schmader T. Stereotype threat deconstructed. Current Directions in Psychological Science. 2010;19(1):14-18.

Question 1. The effect of stereotype threat was observed in the performance of:

A) the women in the problem-solving group. B) the men in the teaching intervention group. C) the women in the math assessment group. D) both the men and women in the teaching intervention group.

Along with other questions on the exam, this question asks test takers to demonstrate their mastery of Competency #8. It asks them to apply their understanding of the ways in which psychological and sociocultural factors influence the way we think about ourselves and others, as well as how we interact with others. This question asks test takers to reason about data and statistics using their knowledge of stereotype threat. To get this question right, test takers have to use information from the passage and graph to decide which observation allows researchers to draw conclusions about the effect of stereotype threat. The graph shows lower performance among women compared to men in the Math Assessment condition (shown as a significant difference, using the error bars). C is the correct answer to this question.

0

504030

1020

6070

Problemsolving

Mathassessment

Teachingintervention

Ave

rage

per

cent

cor

rect

menwomen


36


Question 2. Which statement best explains women’s performance in the teaching intervention group compared to the math assessment group?

A) Women in the teaching intervention group attributed their anxiety to situational factors, and thus their performance was better than women in the math assessment group.

B) Women in the teaching intervention group attributed their anxiety to dispositional factors, and thus their performance was better than women in the math assessment group.

C) Women in the teaching intervention group were informed about stereotype threat, and thus their performance was worse than women in the math assessment group.

D) Women in the teaching intervention group were reminded of gender stereotypes, and thus their performance was worse than women in the math assessment group.

Like the last question, this question asks test takers to demonstrate mastery of Competency #8. It asks test takers to combine their scientific reasoning and problem-solving skills with their knowledge of attributional processes. To get this question right, test takers have to evaluate which explanation best accounts for the findings in Study 1. Using the graph, they should conclude that women’s performance in the teaching intervention group is better than women’s performance in the math assessment group. Test takers have to recognize that in both conditions, gender stereotypes are activated. The question requires test takers to decide which statement provides the best explanation of the different findings across conditions. Test takers should conclude that, among the listed explanations, the most likely one is that women in the teaching intervention group attribute test-related anxiety to a situational factor. A is the correct answer to this question.

Question 3. Which statement presents the most likely explanation for why the researchers in Study 2 collected skin conductivity information? The researchers collected this information to determine whether:

A) the independent variable had the intended effect on the participants. B) the independent variable had an effect on the dependent variable. C) the self-doubt words caused stereotype threat whereas the self-confidence words did not. D) the self-doubt words were more effective than the self-confidence words.

Along with other questions on the exam, this question asks test takers to demonstrate mastery of Competency #7. It requires an understanding of the ways in which biological, psychological, and sociocultural factors influence behavior and behavior change. This question asks test takers to reason about the design and execution of research using their knowledge of the biological bases of behavior. To get this question right, test takers have to determine why an added measure was included as part of the research design in Study 2. They have to recognize that the independent variable in Study 2 (self-confidence versus self-doubt) is intended to manipulate participants’ anxiety levels and that the measures of electrical skin conductance and salivary cortisol levels can be used as a physiological check for the successful manipulation of the independent variable. Test takers should recognize that as anxiety levels increase, so do conductance and salivary cortisol levels. A is the correct answer to this question.


most test takers


37


Example: Critical Analysis and Reasoning Skills

A predetermined covenant of confidentiality characterizes the physician-patient relationship. Possession of contraband in prison is illegal. But suppose that during a routine medical examination, a prison physician notices that Prisoner A has drugs and paraphernalia. Should the physician report the crime, or should confidentiality prevail?

Professional communications between physicians and patients are statutorily protected as confidential. A routine physical examination is part of the confidential communication, like information obtained by taking a medical history and data entered in the patient’s health record. Health professionals have an interest in maintaining confidentiality so that patients will feel comfortable in revealing personal but necessary information. Prisoners do not possess full constitutional rights to privacy, but they generally retain rights to privacy when there is a special relationship between communicants, such as the physician-patient relationship. In fact, respect for confidentiality is particularly important in a prison hospital setting in which patients feel distrust because physicians often are employed by the incarcerating institution.

Clinical autonomy for health professionals in the prison setting is essential for good medical practice. Physicians working in prisons also retain the privilege of confidential interactions with patients, although the prison authorities may try to pressure doctors to supply information. Even if physicians are employed by the prison, their first responsibility is to their patients. The circumstances in which to give privileged information to prison authorities remains the physician’s decision.

The finding that contraband detected during an examination has the appearance of drugs and paraphernalia, like all results of the examination, is privileged information to be treated confidentially. The right to privacy supersedes a duty to report the discovery because there is no imminent threat to others. In contrast, a weapon harbored by a prisoner represents an imminent threat to other prisoners and prison staff. Thus, upon discovering a sequestered weapon during the course of a routine examination, the physician has a “duty to warn.” According to case law, when the physician believes that a significant threat of harm exists, the duty to warn takes precedence over the patient’s right to privacy.

The case of Prisoner A raises the issue of the point at which to draw the line between the duty to protect the public and the duty to protect patients’ privacy. Although legal guidelines can assist the physician in making the choice, the health professional must rely on a guiding principle of the medical profession: Where no danger to others exists, patients come first.

The possibility of discovering contraband during routine examinations of prisoner patients reinforces the need for informed consent at several stages. First, prisoner patients should be evaluated and treated only after they provide informed consent, unless they are incompetent. Before an X-ray is taken, they should be informed that it can demonstrate metal and other foreign bodies, and their agreement to the procedure should be obtained. Second, if a concealed weapon is discovered during a routine examination, the prisoner patient should be informed that the discovery will be reported and given the opportunity to surrender the weapon to authorities before more forcible means are taken to remove it. If Prisoner A is harboring drugs and a needle, drug use is quite possibly contributing to A’s health problem. It is the physician’s responsibility to educate A about the potential harm of drug use.

Adapted from: Levine C, ed. Cases in Bioethics: Selections from the Hastings Center Report. St. Martin’s Press; 1989.

38


Question 1. Which of the following conclusions about physician confidentiality can be inferred from the passage?

A) It is more likely to be assumed in a private setting than in a prison. B) It is especially important when patients are incompetent to give informed consent.C) It is threatened by the use of invasive diagnostic tools such as X-rays. D) It is an aspect of a constitutional right that is lost by prisoners.

Along with other questions on the exam, this question asks test takers to demonstrate that they can comprehend and draw inferences about what they read. Test takers do not need bioethics knowledge to get this question right. All of the information they need to answer the question is included in the passage. The passage says that patients may distrust physicians who work for incarcerating institutions. Test takers have to infer that patients would be more likely to trust that physicians in non-institutional or private settings would maintain confidentiality. A is the correct answer to this question.

Question 2. The author argues that a routine examination is part of the confidential communication between a patient and a physician, and that the clinical autonomy of the physician is essential for good medical practice in prisons. These beliefs imply that:

A) if the quality of medicine practiced in a prison declines, a physician has violated the confidentiality of a routine examination.

B) if all physicians in a prison refuse to reveal information about prisoners obtained during routine examinations, the physicians in that prison have clinical autonomy.

C) if all physicians who conduct routine examinations in a prison respect their patients’ confidentiality, the quality of medicine practiced in the prison is high.

D) if a physician is required to reveal information about a prisoner obtained during a routine examination, the quality of medicine practiced in the prison suffers.

Along with other questions on the exam, this question asks test takers to analyze and evaluate what they read. To get this question right, test takers have to synthesize the author’s claims in order to identify some logical conclusions. They have to conclude that the quality of medical care will suffer is clinical autonomy is absent. D is the correct answer to this question.

Question 3. Which of the following claims, if assumed to be true, would most weaken the argument made for the special importance of the physician-patient covenant within prisons?

A) Prisoners understand that X-rays will detect hidden weapons.B) Prisoners assume that physicians are independent of the institution.C) Prison officials often question physicians about prisoners.D) Prisoners often misunderstand their constitutional rights.

Along with other questions on the exam, this question asks test takers to make a judgment about the impact of new information on arguments made in the passage. This question makes a number of claims about prisoners’ understanding of their constitutional rights and the health care they will receive in prison. To get this question right, test takers have to identify which of the claims most weakens one of the central tenets of the passage. Test takers have to recognize that the physician-patient covenant is less critical to good care in prisons where physicians practice independently of the institution. B is the correct answer to this question.


most test takers



39

Part 2: Using Scores from the Old MCAT Exam in Student Selection


40


This section of the guide provides admissions officers and medical school faculty members who serve on admissions committees with information about the design, interpretation, use, and predictive value of scores from the old MCAT exam. It includes the following topics:

• Content of the old MCAT exam

• Scores from the old MCAT exam

• Validity data on the Old MCAT exam

Content of the Old MCAT Exam

The old MCAT exam was administered from April 1991 to January 2015. It was designed to measure applicants’ knowledge of undergraduate, introductory-level biology, organic chemistry, general chemistry, and physics concepts along with higher-order thinking skills like hypothesis testing, scientific problem solving, verbal reasoning, and quantitative reasoning. It included three multiple-choice sections.1

The Biological Sciences section examined introductory-level knowledge of biology and organic chemistry concepts along with applicants’ skill at identifying main ideas, testing hypotheses, evaluating information, reasoning with flexibility and adaptability, and reasoning using quantitative data.

The Physical Sciences section tested examinees’ introductory-level knowledge of physics and general (inorganic) chemistry concepts, along with the application of these concepts to scientific problems.

The Verbal Reasoning section tested examinees’ ability to understand, evaluate, and apply information and arguments presented in prose text, as well as examinees’ skill at incorporating information. Verbal Reasoning passages were drawn from the natural sciences, social sciences, and humanities. All the information that applicants needed to respond to Verbal Reasoning questions appeared in the passages.

Scores from the Old MCAT Exam

Scores on the multiple-choice sections of the old MCAT exam were reported on a numeric scale ranging from 1 to 15. Scores on the three multiple-choice sections were summed to create a total score, ranging from 3 to 45.

Figure 4 shows the distribution and percentile ranks of MCAT total and section scores for MCAT exams administered from January 2012 through September 2014. The mean MCAT total score for these exams was 25.2, and the standard deviation was 6.4.

1 Prior to January 2013, the old MCAT exam included a fourth section, the Writing Sample. This section tested examinees’ ability to develop a central idea, synthesize concepts and ideas, present ideas cohesively and logically, and write clearly following accepted writing conventions. Applicants who took the old MCAT exam prior to January 2013 will have MCAT results for this fourth section, reported on an alphabetic scale that ranges from J (low) to T (high).

As it was concluding the administration of the old MCAT exam, the AAMC offered 10 additional testing dates from October 2014 to January 2015. During this period, 36,842 exams were administered. The mean MCAT total score for these exams was 23.9 and the standard deviation was 6.4.

41


MCAT Total Score

MCAT Total PercentileScore Rank Mean = 25.245 100 Std. Deviation = 6.444 10043 10042 10041 10040 10039 10038 9937 9836 9735 9634 9433 9132 8831 8330 7929 7328 6727 6126 5525 4924 4323 3722 3221 2720 2319 1918 1517 1216 1015 814 613 512 311 210 2 9 1 8 1 7 0 6 0 5 0 4 0 3 0

MCAT Total Score

012345678910

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Percen

t

MCAT Total Score

NotesThese tables include exam results for all examinees—not just those who have applied (or are planning to apply) to medical school. They also include multiple exam results for examinees who took the MCAT exam more than once during the testing years, but exclude exam results for administrations in which examinees opted to void their exam scores.

The column labeled "Percentile Rank" provides, for exams administered from January 2012 to September 2014, thepercentage of exam results that were equal to or less than each score point.1

For example, 79 percent of MCAT scores were equal to or less than 30across all exams administered from January 2012 to September 2014.

1Prior versions of these tables (pre‐2012), as well as current score reporting systems maintained by AMCAS report the "Percentile Rank Range" in these columns rather than a single percentile rank. The percentile rank columns in the current tables correspond to the upper bound of the percentile rank ranges provided in these other versions. The lower bounds in these other versions essentially serve to provide continuity across score points.

,

total

Figure 4. Summary of total and section scores from the old MCAT exam for tests administered from January 2012 through September 2014 (N = 287,494).

(continued on next page)

42


Physical Sciences (PS)PS Section PercentileScore Rank Mean = 8.315 100 Std. Deviation = 2.514 10013 9712 9511 8910 79 9 67 8 55 7 40 6 23 5 11 4 5 3 2 2 0 1 0

Verbal Reasoning (VR)VR Section Percentile

Score Rank Mean = 8.115 100 Std. Deviation = 2.514 10013 10012 9811 9510 84 9 67 8 52 7 37 6 27 5 15 4 10 3 4 2 2 1 1

Biological Sciences (BS)BS Section PercentileScore Rank Mean = 8.815 100 Std. Deviation = 2.514 9913 9812 9511 8810 76 9 56 8 41 7 25 6 18 5 10 4 6 3 3 2 1 1 1

PS Section Score

VR Section Score

BS Section Score

0

2

4

6

8

10

12

14

16

18

20

22

24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Percent

PS Section Score

0

2

4

6

8

10

12

14

16

18

20

22

24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Percent

VR Section Score

0

2

4

6

8

10

12

14

16

18

20

22

24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Percent

BS Section Score

Figure 4. Summary of total and section scores from the old MCAT exam for tests administered from January 2012 through September 2014 (N = 287,494). (Continued)

43


Figure 5 shows the distributions of total scores from the old MCAT exam for tests administered from January 2012 to September 2014 by sex and racial/ethnic group. It uses box-and-whisker plots, which show the 10th, 25th, 50th (median), 75th, and 90th percentile scores for each group. The 10th and 90th percentile scores are shown by the ends of the “whiskers,” the 25th and 75th percentile scores are shown by the “box” (the left edge of each box shows the 25th percentile score, and the right edge shows the 75th percentile score), and the median is shown by the vertical bar inside each box.

For example, the 10th, 25th, median, 75th, and 90th percentile scores for female examinees were 15, 20, 25, 29, and 32, respectively. The mean total score from the old MCAT exam for each group appears in parentheses by the group label.

There was variability in the median total scores from the old MCAT exam for examinees in different sex and racial/ethnic groups. However, there was a great deal of overlap in the score distributions for different groups. The similarities and differences in these data are similar to those reported in the literature for other admissions tests (Roth et al. 2001; Sackett and Shen 2010). Recent research suggests that these differences in MCAT total scores for racial/ethnic minorities do not reflect test bias (Davis et al. 2013).

Figure 5. MCAT total scores for exams administered from January 2012 through September 2014 by sex and race/ethnicity.

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45

American Indian (mean=23.5; N=2,698)

Asian (mean=26.0; N=69,701)

Black (mean=20.5; N=25,504)

Hispanic (mean=21.9; N=27,980)

Pacific Islander (mean=23.0; N=623)

White (mean=26.1; N=134,431)

Female (mean=24.2; N=148,145)

Male (mean=26.3; N=138,950)

Total Score

44


Validity Data on the Old MCAT Exam

How well do undergraduate GPAs and scores from the old MCAT exam predict students’ performance in medical school?

The value of scores from the old MCAT exam in predicting students’ performance in medical school has been well established (Donnon et al. 2007; Dunleavy et al. 2013; Julian 2005; Koenig and Wiley 1997; Kroopnick et al. 2013; Kuncel and Hezlett 2007). Studies show that undergraduate grades and scores from the old MCAT exam predict students’ grades in medical school, academic difficulty or distinction, time to graduation, scores on USMLE Step exams, and unimpededprogress toward graduation.2

The tables below show the relationships between undergraduate GPAs, total scores from the old MCAT exam, and severalof these outcomes for students who started medical school in 2007, 2008, and 2009. They show the rates at whichmedical students who entered with different combinations of undergraduate GPAs and scores from the old MCAT exam (1) graduated in four years; (2) graduated in five years; (3) withdrew or were dismissed for academic reasons; (4) passedthe USMLE Step 1, Step 2 CK, and Step 2 CS exams on their first attempt; and (5) eventually passed the Step exams.

Table 3 shows the relationships between undergraduate GPAs, total scores from the old MCAT exam, and four-‐year graduation rates. Overall, 87 percent of the medical students graduated in four years, an impressive result given the 62 percent average graduation rate from U.S. graduate and professional schools generally (Bradburn et al. 2006). The percentages in the cells of Table 3 show that higher undergraduate GPAs and scores from the old MCAT exam generallyare associated with higher four-‐year graduation rates.

Table 3. Percentage of 2007–2009 Students Who Graduated from Medical School in Four Years, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 72% 77% 87% 91% 91% 90% 89% 88% 90%

59/82 244/317 1,048/1,200 3,025/3,333 4,328/4,774 3,157/3,497 1,713/1,934 615/701 14,196/15,850 3.60–3.79 -- 58% 66% 74% 83% 89% 90% 89% 87% 90% 88%

7/12 81/122 294/395 1,010/1,220 2,911/3,269 4,086/4,530 2,525/2,833 1,027/1,174 233/258 12,175/13,816 3.40–3.59 -- 40% 69% 72% 80% 87% 89% 90% 88% 89% 87%

4/10 75/109 259/362 762/955 1,960/2,252 2,656/2,979 1,537/1,713 562/641 115/129 7,933/9,157 3.20–3.39 -- 55% 65% 65% 75% 85% 90% 89% 88% 84% 84%

6/11 60/93 178/272 418/557 905/1,066 1,202/1,336 676/760 247/280 41/49 3,734/4,427 3.00–3.19 -- 43% 38% 62% 71% 84% 87% 90% 90% 81% 80%

6/14 20/52 104/169 229/321 411/487 429/495 246/274 97/108 13/16 1,557/1,939 2.80–2.99 -- 38% 59% 62% 74% 85% 82% 88% -- 72%

11/29 54/92 91/147 120/162 128/150 77/94 15/17 508/706 2.60–2.79 -- 44% 47% 63% 80% 86% 85% -- -- 70%

8/18 18/38 32/51 45/56 42/49 22/26 171/246 2.40–2.59 -- -- 40% 61% 71% 86% -- -- 73%

4/10 11/18 15/21 18/21 61/84 2.20–2.39 -- -- -- -- -- -- -- 44%

8/18 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 50% 47% 62% 70% 81% 88% 90% 90% 88% 88% 87% 10/20 31/66 316/510 1,155/1,660 3,604/4,474 9,394/10,650 12,893/14,339 8,253/9,210 3,667/4,160 1,026/1,163 40,349/46,252

Notes: 1. Blue shading = graduation rates of 90–100%; green shading = graduation rates of 80–89%; orange shading = graduation rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. Students entering medical school with advanced standing from medical, graduate, or other programs; enrolled in joint programs (e.g., MD-PhD) at the time of

matriculation or graduation; participating in special research/nonresearch studies; or deceased are not included in this table. 4. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

2 The median corrected validity coefficient was 0.75 for predicting first-‐year medical school grades from undergraduate GPAs and total scores fromthe old MCAT exam, 0.71 for predicting second-‐year grades (Koenig and Wiley 1997), and 0.54 for predicting third-‐year grades (Julian 2005). The median corrected validity coefficient was 0.64 for predicting USMLE Step 1 scores from undergraduate GPAs and total scores from the old MCAT exam, 0.54 for Step 2 Clinical Knowledge scores, and 0.55 for Step 3 scores (Kroopnick et al. 2013).

Validity Data on the Old MCAT Exam

How well do undergraduate GPAs and scores from the old MCAT exam predict students’ performance in medical school?

The value of scores from the old MCAT exam in predicting students’ performance in medical school has been well established (Donnon et al. 2007; Dunleavy et al. 2013; Julian 2005; Koenig and Wiley 1997; Kroopnick et al. 2013; Kuncel and Hezlett 2007). Studies show that undergraduate grades and scores from the old MCAT exam predict students’ grades in medical school, academic difficulty or distinction, time to graduation, scores on USMLE Step exams, and unimpeded progress toward graduation.2

The tables below show the relationships between undergraduate GPAs, total scores from the old MCAT exam, and several of these outcomes for students who started medical school in 2007, 2008, and 2009. They show the rates at which medical students who entered with different combinations of undergraduate GPAs and scores from the old MCAT exam (1) graduated in four years; (2) graduated in five years; (3) withdrew or were dismissed for academic reasons; (4) passed the USMLE Step 1, Step 2 CK, and Step 2 CS exams on their first attempt; and (5) eventually passed the Step exams.

Table 3 shows the relationships between undergraduate GPAs, total scores from the old MCAT exam, and four-year graduation rates. Overall, 87 percent of the medical students graduated in four years, an impressive result given the 62 percent average graduation rate from U.S. graduate and professional schools generally (Bradburn et al. 2006). The percentages in the cells of Table 3 show that higher undergraduate GPAs and scores from the old MCAT exam generally are associated with higher four-year graduation rates.

2 The median corrected validity coefficient was 0.75 for predicting first-year medical school grades from undergraduate GPAs and total scores from the old MCAT exam, 0.71 for predicting second-year grades (Koenig and Wiley 1997), and 0.54 for predicting third-year grades (Julian 2005). The median corrected validity coefficient was 0.64 for predicting USMLE Step 1 scores from undergraduate GPAs and total scores from the old MCAT exam, 0.54 for Step 2 Clinical Knowledge scores, and 0.55 for Step 3 scores (Kroopnick et al. 2013).

45


Table 4 shows the relationships between undergraduate GPAs, scores from the old MCAT exam, and five-‐year graduation rates. These data show that 95 percent of the students who entered medical school between 2007 and 2009 graduated in five years. The pattern of results shows that most students graduated from medical school in five years, including those who entered with modest MCAT scores. The percentages in the cells of Table 4 show that higher undergraduate GPAs and scores from the old MCAT exam generally are associated with higher five-‐year graduation rates.

Table 4. Percentage of 2007–2009 Students Who Graduated from Medical School in Five Years, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 84% 90% 94% 96% 97% 97% 98% 97% 96%

69/82 284/317 1,127/1,200 3,211/3,333 4,612/4,774 3,401/3,497 1,888/1,934 677/701 15,280/15,850 3.60–3.79 -- 58% 87% 88% 93% 95% 97% 96% 97% 98% 96%

7/12 106/122 347/395 1,131/1,220 3,110/3,269 4,378/4,530 2,732/2,833 1,137/1,174 254/258 13,203/13,816 3.40–3.59 -- 60% 82% 86% 89% 95% 96% 96% 96% 97% 94%

6/10 89/109 310/362 854/955 2,141/2,252 2,864/2,979 1,642/1,713 614/641 125/129 8,651/9,157 3.20–3.39 -- 91% 83% 83% 89% 93% 97% 96% 97% 94% 94%

10/11 77/93 225/272 496/557 996/1,066 1,294/1,336 733/760 272/280 46/49 4,151/4,427 3.00–3.19 -- 71% 62% 79% 86% 94% 94% 95% 94% 88% 90%

10/14 32/52 133/169 276/321 456/487 467/495 261/274 101/108 14/16 1,752/1,939 2.80–2.99 -- 66% 78% 84% 88% 95% 90% 100% -- 87%

19/29 72/92 123/147 143/162 142/150 85/94 17/17 614/706 2.60–2.79 -- 67% 76% 82% 84% 92% 96% -- -- 84%

12/18 29/38 42/51 47/56 45/49 25/26 206/246 2.40–2.59 -- -- 80% 78% 86% 90% -- -- 87%

8/10 14/18 18/21 19/21 73/84 2.20–2.39 -- -- -- -- -- -- -- 56%

10/18 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 75% 70% 80% 85% 91% 95% 96% 97% 97% 97% 95% 15/20 46/66 407/510 1,409/1,660 4,068/4,474 10,125/10,650 13,825/14,339 8,892/9,210 4,035/4,160 1,126/1,163 43,948/46,252

Notes: 1. Blue shading = graduation rates of 90–100%; green shading = graduation rates of 80–89%; orange shading = graduation rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. Students entering medical school with advanced standing from medical, graduate, or other programs; enrolled in joint programs (e.g., MD-PhD) at the time of

matriculation or graduation; participating in special research/nonresearch studies; or deceased are not included in this table. 4. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

Table 4 shows the relationships between undergraduate GPAs, scores from the old MCAT exam, and five-year graduation rates. These data show that 95 percent of the students who entered medical school between 2007 and 2009 graduated in five years. The pattern of results shows that most students graduated from medical school in five years, including those who entered with modest MCAT scores. The percentages in the cells of Table 4 show that higher undergraduate GPAs and scores from the old MCAT exam generally are associated with higher five-year graduation rates.

46


Table 5 shows the relationships between undergraduate GPAs, total scores from the old MCAT exam, and withdrawal or dismissal from medical school for academic reasons. According to these data, only 1.4 percent of entrants withdrew or were dismissed from medical school for academic reasons. These results show a negative association between undergraduate GPAs, total scores from the old MCAT exam, and this outcome. In general, lower undergraduate GPAs and total scores from the old MCAT exam are associated with higher levels of academic withdrawal or dismissal from medical school for academic reasons.

Table 5. Percentage of 2007–2009 Students Who Withdrew or Were Dismissed from Medical School for Academic Reasons, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 2.3% 2.7% 1.8% 0.7% 0.7% 0.4% 0.2% 0.1% 0.7%

2/86 9/338 24/1,304 27/3,621 35/5,324 16/4,290 6/2,566 1/1,026 121/18,567 3.60–3.79 -- 14.3% 5.3% 1.7% 2.6% 1.8% 0.9% 0.9% 0.3% 0.6% 1.2%

2/14 7/131 7/406 34/1,317 64/3,543 43/4,978 30/3,344 5/1,505 2/355 194/15,596 3.40–3.59 -- 8.3% 4.2% 5.9% 4.7% 1.6% 1.2% 1.1% 0.6% 0% 1.8%

1/12 5/119 23/390 48/1,019 40/2,436 38/3,278 22/1,996 5/801 0/168 182/10,227 3.20–3.39 -- 0% 4.0% 5.9% 4.2% 1.9% 1.0% 1.2% 0.6% 1.8% 2.0%

0/11 4/100 17/290 25/597 22/1,156 14/1,453 10/843 2/324 1/57 95/4,834 3.00–3.19 -- 6.7% 10.7% 9.8% 5.6% 2.9% 1.7% 1.0% 1.6% 6.3% 3.6%

1/15 6/56 17/173 19/341 15/522 9/531 3/296 2/123 1/16 74/2,076 2.80–2.99 -- 11.1% 7.2% 6.5% 4.6% 1.9% 1.0% 0% -- 4.6%

4/36 7/97 10/155 8/173 3/161 1/98 0/21 35/756 2.60–2.79 -- 4.8% 7.3% 5.7% 8.6% 3.8% 0% -- -- 5.7%

1/21 3/41 3/53 5/58 2/52 0/26 15/261 2.40–2.59 -- -- 0% 11.1% 4.8% 4.3% 8.3% -- 5.5%

0/12 2/18 1/21 1/23 1/12 5/91 2.20–2.39 -- -- -- -- -- -- -- 30.0%

6/20 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 4.8% 12.7% 5.6% 4.8% 3.4% 1.6% 0.9% 0.8% 0.4% 0.3% 1.4% 1/21 9/71 31/554 85/1,753 165/4,809 183/11,535 146/15,806 83/10,909 20/5,349 5/1,632 728/52,439

Notes: 1. Blue shading = withdrawal/dismissal rates of 0–10.0%; green shading = withdrawal/dismissal rates of 10.1–20.0%; orange shading = withdrawal/dismissal rates of 20.1–30.0%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

Table 5 shows the relationships between undergraduate GPAs, total scores from the old MCAT exam, and withdrawal or dismissal from medical school for academic reasons. According to these data, only 1.4 percent of entrants withdrew or were dismissed from medical school for academic reasons. These results show a negative association between undergraduate GPAs, total scores from the old MCAT exam, and this outcome. In general, lower undergraduate GPAs and total scores from the old MCAT exam are associated with higher levels of academic withdrawal or dismissal from medical school for academic reasons.

47


Tables that show the same types of relationships between undergraduate GPAs, total scores from the old MCAT exam, and USMLE Step exam pass rates appear in Tables 6 to 11. These data show positive relationships between undergraduate GPAs, total scores from the old MCAT exam, and first-time and eventual pass rates on the Step exams. Refer to Tables 6, 8, and 10 for grids summarizing the relationships with first-time pass rates on USMLE Step 1, Step 2 CK, and Step 2 CS exams. Tables 7, 9, and 11 summarize these relationships for students’ eventual pass rates for these exams.

Taken together, the examination of graduation rates, withdrawal or dismissal from medical school for academic reasons, and USMLE Step exam pass rates suggest that applicants with a wide range of undergraduate GPAs and scores from the old MCAT exam can succeed in medical school. These data suggest the facility with which admissions committee members predict their students’ eventual success by combining information about candidates’ qualifications from applications, transcripts, letters, and interviews with information about the curricula, support services, and graduation requirements of their medical schools. These results also are an important testament to the academic and other support that faculty provide to their students to help them succeed in medical school.

48


Tables that show the same types of relationships between undergraduate GPAs, total scores from the old MCAT exam, and USMLE Step exam pass rates appear in Tables 6 to 11. The data show positive relationships between undergraduateGPAs, total scores from the old MCAT exam, and first-‐time and eventual pass rates on the Step exams. Refer to Tables 6, 8 , and 10 for grids summarizing the relationships with first-‐time pass rates on USMLE Step 1, Step 2 CK, and Step 2 CS exams. Tables 7, 9, and 11 summarize these relationships for students’ eventual pass rates for these exams.

Taken together, the examination of graduation rates, withdrawal or dismissal from medical school for academic reasons, and USMLE Step exam pass rates suggest that applicants with a wide range of undergraduate GPAs and scores from the old MCAT exam can succeed in medical school. These data suggest the facility with which admissions committee members predict their students’ eventual success by combining information about candidates’ qualifications from applications, transcripts, letters, and interviews with information about the curricula, support services, and graduation requirements of their medical schools. These results also are an important testament to the academic and other support that facultyprovide to their students to help them succeed in medical school.[Designer: Please put Tables 6 and 7 on the same page and separate them visually with space – or a line? Prefer not a line.]

Table 6. Percentage of 2007–2009 Students Passing the Step 1 Exam on the First Attempt, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 79% 84% 91% 96% 98% 98% 99% 100% 97%

65/82 270/322 1,148/1,267 3,411/3,553 5,106/5,219 4,157/4,222 2,523/2,537 1,015/1,017 17,701/18,230 3.60–3.79 -- 42% 68% 80% 88% 93% 96% 98% 99% 99% 95%

5/12 85/125 315/395 1,119/1,266 3,232/3,460 4,691/4,892 3,213/3,292 1,468/1,485 351/354 14,480/15,283 3.40–3.59 -- -- 69% 78% 83% 92% 95% 98% 98% 100% 93%

79/115 280/359 807/974 2,194/2,383 3,076/3,225 1,912/1,959 769/786 167/167 9,295/9,985 3.20–3.39 -- 45% 59% 76% 81% 90% 94% 97% 98% 96% 91%

5/11 57/96 210/275 460/566 1,018/1,129 1,356/1,435 804/829 314/320 53/55 4,277/4,719 3.00–3.19 -- 46% 50% 69% 82% 86% 92% 98% 98% 93% 87%

6/13 25/50 111/160 260/319 441/511 479/520 285/291 119/122 13/14 1,741/2,003 2.80–2.99 -- 48% 64% 77% 80% 90% 92% 95% -- 81%

16/33 56/88 116/150 131/163 141/156 89/97 20/21 582/721 2.60–2.79 -- 47% 67% 80% 87% 92% 100% -- -- 82%

9/19 26/39 41/51 46/53 48/52 26/26 204/249 2.40–2.59 -- -- 50% 65% 95% 95% 100% -- 82%

6/12 11/17 19/20 20/21 12/12 71/87 2.20–2.39 -- -- -- -- -- -- 53%

9/17 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 60% 52% 64% 77% 86% 93% 96% 98% 99% 100% 94% 12/20 31/60 337/523 1,275/1,656 3,965/4,615 10,495/11,276 14,921/15,526 10,502/10,732 5,222/5,280 1,609/1,617 48,369/51,305

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

Table 7. Percentage of 2007–2009 Students Eventually Passing the Step 1 Exam, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 98% 98% 99% 100% 100% 100% 100% 100% 100%

80/82 314/322 1,252/1,267 3,537/3,553 5,207/5,219 4,216/4,222 2,531/2,537 1,016/1,017 18,164/18,230 3.60–3.79 -- 92% 97% 98% 99% 99% 100% 100% 100% 100% 99%

11/12 121/125 389/395 1,248/1,266 3,433/3,460 4,872/4,892 3,280/3,292 1,479/1,485 354/354 15,189/15,283 3.40–3.59 -- -- 97% 97% 98% 99% 99% 100% 99% 100% 99%

111/115 349/359 954/974 2,368/2,383 3,206/3,225 1,954/1,959 782/786 167/167 9,908/9,985 3.20–3.39 -- 100% 95% 96% 98% 99% 100% 99% 100% 100% 99%

11/11 91/96 265/275 556/566 1,118/1,129 1,432/1,435 822/829 320/320 55/55 4,673/4,719 3.00–3.19 -- 100% 92% 94% 97% 99% 99% 100% 99% 100% 98%

13/13 46/50 151/160 311/319 504/511 515/520 291/291 121/122 14/14 1,969/2,003 2.80–2.99 -- 88% 97% 98% 96% 99% 100% 100% -- 98%

29/33 85/88 147/150 157/163 155/156 97/97 21/21 704/721 2.60–2.79 -- 95% 97% 100% 96% 98% 100% -- -- 98%

18/19 38/39 51/51 51/53 51/52 26/26 243/249 2.40–2.59 -- -- 83% 100% 100% 100% 100% -- 98%

10/12 17/17 20/20 21/21 12/12 85/87 2.20–2.39 -- -- -- -- -- -- 88%

15/17 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 100% 97% 95% 97% 98% 99% 100% 100% 100% 100% 99% 20/20 58/60 499/523 1,606/1,656 4,541/4,615 11,192/11,276 15,464/15,526 10,702/10,732 5,263/5,280 1,616/1,617 50,961/51,305


[Designer: Please put Tables 8 and 9 on the same page and separate them visually with space – or a line? Prefer not a line.]

Table 8. Percentage of 2007–2009 Students Passing the Step 2 CK Exam on the First Attempt, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 91% 95% 98% 98% 99% 100% 100% 100% 99%

71/78 288/302 1,198/1,224 3,413/3,467 4,968/5,000 3,877/3,883 2,230/2,239 831/831 16,887/17,0353.60–3.79 -- -- 91% 95% 96% 98% 99% 99% 100% 100% 99%

106/117 355/374 1,178/1,222 3,284/3,346 4,662/4,712 3,054/3,077 1,339/1,345 295/295 14,284/14,4993.40–3.59 -- -- 88% 95% 95% 98% 98% 99% 99% 100% 98%

91/104 324/340 883/926 2,246/2,299 3,053/3,100 1,825/1,844 698/706 149/149 9,281/9,4833.20–3.39 -- 100% 91% 88% 91% 97% 98% 99% 99% 98% 96%

10/10 77/85 217/248 490/538 1,047/1,080 1,369/1,395 780/791 295/298 51/52 4,339/4,5003.00–3.19 -- 62% 73% 91% 94% 94% 96% 98% 97% 100% 95%

8/13 27/37 130/143 276/294 463/490 477/496 270/276 113/116 14/14 1,780/1,8812.80–2.99 -- 88% 87% 90% 92% 97% 97% 100% -- 93%

21/24 68/78 121/135 143/155 146/151 87/90 21/21 620/6672.60–2.79 -- 92% 82% 91% 96% 89% 96% -- -- 91%

12/13 28/34 41/45 48/50 42/47 24/25 201/2202.40–2.59 -- -- -- 87% 95% 100% 100% -- 95%

13/15 19/20 20/20 10/10 75/792.20–2.39 -- -- -- -- -- -- 85%

11/132.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 89% 87% 89% 93% 95% 98% 99% 99% 99% 100% 98% 16/18 47/54 408/461 1,420/1,531 4,204/4,404 10,665/10,910 14,742/14,926 9,930/10,000 4,703/4,732 1,350/1,351 47,485/48,387

Notes: 1. Blue shading = passing rates of 90-100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations.3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

49


Table 7. Percentage of 2007–2009 Students Eventually Passing the Step 1 Exam, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 98% 98% 99% 100% 100% 100% 100% 100% 100%

80/82 314/322 1,252/1,267 3,537/3,553 5,207/5,219 4,216/4,222 2,531/2,537 1,016/1,017 18,164/18,2303.60–3.79 -- 92% 97% 98% 99% 99% 100% 100% 100% 100% 99%

11/12 121/125 389/395 1,248/1,266 3,433/3,460 4,872/4,892 3,280/3,292 1,479/1,485 354/354 15,189/15,2833.40–3.59 -- -- 97% 97% 98% 99% 99% 100% 99% 100% 99%

111/115 349/359 954/974 2,368/2,383 3,206/3,225 1,954/1,959 782/786 167/167 9,908/9,9853.20–3.39 -- 100% 95% 96% 98% 99% 100% 99% 100% 100% 99%

11/11 91/96 265/275 556/566 1,118/1,129 1,432/1,435 822/829 320/320 55/55 4,673/4,7193.00–3.19 -- 100% 92% 94% 97% 99% 99% 100% 99% 100% 98%

13/13 46/50 151/160 311/319 504/511 515/520 291/291 121/122 14/14 1,969/2,0032.80–2.99 -- 88% 97% 98% 96% 99% 100% 100% -- 98%

29/33 85/88 147/150 157/163 155/156 97/97 21/21 704/7212.60–2.79 -- 95% 97% 100% 96% 98% 100% -- -- 98%

18/19 38/39 51/51 51/53 51/52 26/26 243/2492.40–2.59 -- -- 83% 100% 100% 100% 100% -- 98%

10/12 17/17 20/20 21/21 12/12 85/872.20–2.39 -- -- -- -- -- -- 88%

15/172.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 100% 97% 95% 97% 98% 99% 100% 100% 100% 100% 99% 20/20 58/60 499/523 1,606/1,656 4,541/4,615 11,192/11,276 15,464/15,526 10,702/10,732 5,263/5,280 1,616/1,617 50,961/51,305

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations.3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.


Table 8. Percentage of 2007–2009 Students Passing the Step 2 CK Exam on the First Attempt, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 91% 95% 98% 98% 99% 100% 100% 100% 99%

71/78 288/302 1,198/1,224 3,413/3,467 4,968/5,000 3,877/3,883 2,230/2,239 831/831 16,887/17,035 3.60–3.79 -- -- 91% 95% 96% 98% 99% 99% 100% 100% 99%

106/117 355/374 1,178/1,222 3,284/3,346 4,662/4,712 3,054/3,077 1,339/1,345 295/295 14,284/14,499 3.40–3.59 -- -- 88% 95% 95% 98% 98% 99% 99% 100% 98%

91/104 324/340 883/926 2,246/2,299 3,053/3,100 1,825/1,844 698/706 149/149 9,281/9,483 3.20–3.39 -- 100% 91% 88% 91% 97% 98% 99% 99% 98% 96%

10/10 77/85 217/248 490/538 1,047/1,080 1,369/1,395 780/791 295/298 51/52 4,339/4,500 3.00–3.19 -- 62% 73% 91% 94% 94% 96% 98% 97% 100% 95%

8/13 27/37 130/143 276/294 463/490 477/496 270/276 113/116 14/14 1,780/1,881 2.80–2.99 -- 88% 87% 90% 92% 97% 97% 100% -- 93%

21/24 68/78 121/135 143/155 146/151 87/90 21/21 620/667 2.60–2.79 -- 92% 82% 91% 96% 89% 96% -- -- 91%

12/13 28/34 41/45 48/50 42/47 24/25 201/220 2.40–2.59 -- -- -- 87% 95% 100% 100% -- 95%

13/15 19/20 20/20 10/10 75/79 2.20–2.39 -- -- -- -- -- -- 85%

11/13 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 89% 87% 89% 93% 95% 98% 99% 99% 99% 100% 98% 16/18 47/54 408/461 1,420/1,531 4,204/4,404 10,665/10,910 14,742/14,926 9,930/10,000 4,703/4,732 1,350/1,351 47,485/48,387

Notes: 1. Blue shading = passing rates of 90-100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than 10 observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

Table 9. Percentage of 2007–2009 Students Eventually Passing the Step 2 CK Exam, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 97% 99% 100% 100% 100% 100% 100% 100% 100%

76/78 298/302 1,221/1,224 3,458/3,467 4,994/5,000 3,881/3,883 2,238/2,239 831/831 17,008/17,035 3.60–3.79 -- -- 97% 98% 100% 100% 100% 100% 100% 100% 100%

113/117 366/374 1,217/1,222 3,333/3,346 4,702/4,712 3,074/3,077 1,343/1,345 295/295 14,454/14,499 3.40–3.59 -- -- 94% 99% 99% 100% 100% 100% 100% 100% 100%

98/104 337/340 916/926 2,289/2,299 3,088/3,100 1,840/1,844 705/706 149/149 9,436/9,483 3.20–3.39 -- 100% 99% 98% 98% 99% 99% 100% 100% 100% 99%

10/10 84/85 242/248 527/538 1,070/1,080 1,388/1,395 789/791 298/298 52/52 4,463/4,500 3.00–3.19 -- 85% 92% 98% 99% 99% 99% 100% 98% 100% 99%

11/13 34/37 140/143 291/294 485/490 493/496 275/276 114/116 14/14 1,859/1,881 2.80–2.99 -- 92% 94% 97% 99% 100% 99% 100% -- 98%

22/24 73/78 131/135 153/155 151/151 89/90 21/21 653/667 2.60–2.79 -- 92% 91% 98% 100% 100% 100% -- -- 98%

12/13 31/34 44/45 50/50 47/47 25/25 215/220 2.40–2.59 -- -- -- 93% 100% 100% 100% -- 97%

14/15 20/20 20/20 10/10 77/79 2.20–2.39 -- -- -- -- -- -- 100%

13/13 2.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 94% 94% 96% 98% 99% 100% 100% 100% 100% 100% 100% 17/18 51/54 442/461 1,497/1,531 4,366/4,404 10,861/10,910 14,888/14,926 9,987/10,000 4,726/4,732 1,351/1,351 48,186/48,387

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; Orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than10n observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.


Table 10. Percentage of 2007–2009 Students Passing the Step 2 CS Exam on the First Attempt, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 94% 96% 98% 99% 99% 99% 99% 99% 99%

72/77 280/292 1,182/1,205 3,376/3,421 4,896/4,943 3,813/3,843 2,209/2,226 825/830 16,663/16,8483.60–3.79 -- -- 83% 95% 99% 98% 98% 99% 99% 99% 98%

90/108 344/362 1,175/1,191 3,244/3,311 4,575/4,651 2,992/3,035 1,326/1,338 288/290 14,044/14,2973.40–3.59 -- -- 85% 92% 97% 98% 98% 98% 99% 97% 98%

83/98 302/328 876/907 2,224/2,260 2,980/3,051 1,794/1,827 685/693 141/145 9,095/9,3233.20–3.39 -- 70% 90% 89% 96% 96% 98% 97% 97% 98% 97%

7/10 76/84 212/237 498/519 1,021/1,062 1,346/1,367 758/780 285/294 48/49 4,253/4,4053.00–3.19 -- 73% 71% 92% 94% 94% 97% 99% 96% 100% 95%

8/11 24/34 124/135 270/286 450/477 478/493 268/272 110/114 14/14 1,748/1,8382.80–2.99 -- 71% 88% 94% 95% 92% 94% 95% -- 92%

17/24 63/72 124/132 143/150 137/149 85/90 20/21 602/6512.60–2.79 -- 73% 79% 90% 94% 91% 96% -- 89%

8/11 22/28 37/41 45/48 43/47 25/26 184/2062.40–2.59 -- -- -- 100% 95% 100% 90% -- 92%

13/13 19/20 20/20 9/10 70/762.20–2.39 -- -- -- -- -- -- 77%

10/132.00–2.19 -- -- --

1.47–1.99 -- -- -- --

All 88% 77% 85% 92% 97% 98% 98% 99% 99% 99% 98% 15/17 40/52 372/439 1,354/1,464 4,178/4,299 10,525/10,752 14,480/14,726 9,748/9,887 4,641/4,693 1,325/1,337 46,678/47,666

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than10n observations; blank cells = cells with 0 observations.3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

50


Table 9. Percentage of 2007–2009 Students Eventually Passing the Step 2 CK Exam, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 97% 99% 100% 100% 100% 100% 100% 100% 100%

76/78 298/302 1,221/1,224 3,458/3,467 4,994/5,000 3,881/3,883 2,238/2,239 831/831 17,008/17,0353.60–3.79 -- -- 97% 98% 100% 100% 100% 100% 100% 100% 100%

113/117 366/374 1,217/1,222 3,333/3,346 4,702/4,712 3,074/3,077 1,343/1,345 295/295 14,454/14,4993.40–3.59 -- -- 94% 99% 99% 100% 100% 100% 100% 100% 100%

98/104 337/340 916/926 2,289/2,299 3,088/3,100 1,840/1,844 705/706 149/149 9,436/9,4833.20–3.39 -- 100% 99% 98% 98% 99% 99% 100% 100% 100% 99%

10/10 84/85 242/248 527/538 1,070/1,080 1,388/1,395 789/791 298/298 52/52 4,463/4,5003.00–3.19 -- 85% 92% 98% 99% 99% 99% 100% 98% 100% 99%

11/13 34/37 140/143 291/294 485/490 493/496 275/276 114/116 14/14 1,859/1,8812.80–2.99 -- 92% 94% 97% 99% 100% 99% 100% -- 98%

22/24 73/78 131/135 153/155 151/151 89/90 21/21 653/6672.60–2.79 -- 92% 91% 98% 100% 100% 100% -- -- 98%

12/13 31/34 44/45 50/50 47/47 25/25 215/2202.40–2.59 -- -- -- 93% 100% 100% 100% -- 97%

14/15 20/20 20/20 10/10 77/792.20–2.39 -- -- -- -- -- -- 100%

13/132.00–2.19 -- -- -- --

1.47–1.99 -- -- -- --

All 94% 94% 96% 98% 99% 100% 100% 100% 100% 100% 100% 17/18 51/54 442/461 1,497/1,531 4,366/4,404 10,861/10,910 14,888/14,926 9,987/10,000 4,726/4,732 1,351/1,351 48,186/48,387

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; Orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than10n observations; blank cells = cells with 0 observations.3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.


Table 10. Percentage of 2007–2009 Students Passing the Step 2 CS Exam on the First Attempt, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 94% 96% 98% 99% 99% 99% 99% 99% 99%

72/77 280/292 1,182/1,205 3,376/3,421 4,896/4,943 3,813/3,843 2,209/2,226 825/830 16,663/16,848 3.60–3.79 -- -- 83% 95% 99% 98% 98% 99% 99% 99% 98%

90/108 344/362 1,175/1,191 3,244/3,311 4,575/4,651 2,992/3,035 1,326/1,338 288/290 14,044/14,297 3.40–3.59 -- -- 85% 92% 97% 98% 98% 98% 99% 97% 98%

83/98 302/328 876/907 2,224/2,260 2,980/3,051 1,794/1,827 685/693 141/145 9,095/9,323 3.20–3.39 -- 70% 90% 89% 96% 96% 98% 97% 97% 98% 97%

7/10 76/84 212/237 498/519 1,021/1,062 1,346/1,367 758/780 285/294 48/49 4,253/4,405 3.00–3.19 -- 73% 71% 92% 94% 94% 97% 99% 96% 100% 95%

8/11 24/34 124/135 270/286 450/477 478/493 268/272 110/114 14/14 1,748/1,838 2.80–2.99 -- 71% 88% 94% 95% 92% 94% 95% -- 92%

17/24 63/72 124/132 143/150 137/149 85/90 20/21 602/651 2.60–2.79 -- 73% 79% 90% 94% 91% 96% -- 89%

8/11 22/28 37/41 45/48 43/47 25/26 184/206 2.40–2.59 -- -- -- 100% 95% 100% 90% -- 92%

13/13 19/20 20/20 9/10 70/76 2.20–2.39 -- -- -- -- -- -- 77%

10/13 2.00–2.19 -- -- --

1.47–1.99 -- -- -- --

All 88% 77% 85% 92% 97% 98% 98% 99% 99% 99% 98% 15/17 40/52 372/439 1,354/1,464 4,178/4,299 10,525/10,752 14,480/14,726 9,748/9,887 4,641/4,693 1,325/1,337 46,678/47,666

Notes: 1. Blue shading = passing rates of 90–100%; green shading = passing rates of 80–89%; orange shading = passing rates of 70–79%. 2. Dashes = cells with fewer than10n observations; blank cells = cells with 0 observations. 3. For students who took the old MCAT exam multiple times, the most recent old MCAT total score at the time of matriculation was used in this analysis.

Table 11. Percentage of 2007–2009 Students Eventually Passing the Step 2 CS Exam, by Old MCAT Total Score and Undergraduate GPA Range


5–14 15–17 18–20 21–23 24–26 27–29 30–32 33–35 36–38 39–45 All 3.80–4.00 -- -- 96% 98% 100% 100% 100% 100% 100% 100% 100%

74/77 287/292 1,202/1,205 3,419/3,421 4,938/4,943 3,841/3,843 2,224/2,226 830/830 16,826/16,848 3.60–3.79 -- -- 92% 99% 100% 100% 100% 100% 100% 100% 100%

99/108 358/362 1,191/1,191 3,303/3,311 4,645/4,651 3,031/3,035 1,336/1,338 290/290 14,264/14,297 3.40–3.59 -- -- 93% 98% 100% 100% 100% 100% 100% 100% 100%

91/98 321/328 905/907 2,259/2,260 3,044/3,051 1,823/1,827 691/693 145/145 9,292/9,323 3.20–3.39 -- 90% 98% 95% 99% 99% 100% 100% 100% 100% 99%

9/10 82/84 225/237 515/519 1,056/1,062 1,366/1,367 779/780 294/294 49/49 4,377/4,405 3.00–3.19 -- 82% 88% 99% 98% 99% 100% 100% 99% 100% 99%

9/11 30/34 133/135 281/286 473/477 493/493 272/272 113/114 14/14 1,820/1,838 2.80–2.99 -- 83% 97% 99% 99% 100% 99% 100% -- 99%

20/24 70/72 131/132 149/150 149/149 89/90 21/21 642/651 2.60–2.79 -- 91% 93% 100% 98% 98% 96% -- 97%

10/11 26/28 41/41 47/48 46/47 25/26 200/206 2.40–2.59 -- -- -- 100% 95% 100% 100% -- 97%

13/13 19/20 20/20 10/10 74/76 2.20–2.39 -- -- -- -- -- -- 100%

13/13 2.00–2.19 -- -- --

1.47–1.99 -- -- -- --

All 88% 94% 93% 98% 100% 100% 100% 100% 100% 100% 100% 15/17 49/52 409/439 1,429/1,464 4,284/4,299 10,728/10,752 14,706/14,726 9,874/9,887 4,686/4,693 1,337/1,337 47,517/47,666


51


References

Bradburn, E.M., Nevill, S., Cataldi, E.F., and Perry, K. (2006). Where Are They Now? A Description of 1992–93 Bachelor’s Degree Recipients 10 Years Later (NCES 2007-159). Washington, DC: U.S. Department of Education, National Center for Education Statistics.

Davis, D., Dorsey, J.K., Franks, R.D., Sackett, P.R., Searcy, C.A., and Zhao, X. (2013). Do racial and ethnic group differences in performance on the MCAT exam reflect test bias? Academic Medicine, 88(5), 593–602.

Donnon, T., Paolucci, E.O., and Violato, C. (2007). The predictive validity of the MCAT for medical school performance and medical board licensing examinations: A meta-analysis of the published research. Academic Medicine, 82, 100–106.

Dunleavy, D.M., Kroopnick, M.H., Dowd, K.W., Searcy, C.A., and Zhao, X. (2013). The predictive validity of the MCAT exam in relation to academic performance through medical school: A national cohort study of 2001–2004 matriculants. Academic Medicine, 88(5), 666–671.

Julian, E. R. (2005). Validity of the medical college admission test for predicting medical school performance. Academic Medicine, 80, 910–917.

Koenig, J.A., and Wiley, A. (1997). Medical school admission testing. In R.F. Dillon (Ed.), Handbook on Testing (pp. 274–295). Westport, CT: Greenwood Press.

Kroopnick, M.H., Dunleavy, D.M., Dowd, K.W., Searcy, C.A., and Zhao, X. (2013). A comprehensive school-level analysis of the predictive validity of the Medical College Admission Test (MCAT). Paper presented at the annual conference of the American Educational Research Association, San Francisco.

Kuncel, N.R., and Hezlett, S.A. (2007). Standardized tests predict graduate students’ success. Science, 315, 1080–1081.

Roth, P.L., Bevier, C.A., Bobko, P., Switzer, F.S. III, and Tyler, P. (2001). Ethnic group differences in cognitive ability in employment and educational settings: A meta-analysis. Personnel Psychology, 54, 297–330.

Sackett, P.R., and Shen, W. (2010). Subgroup differences on cognitive tests in contexts other than personnel selection. In J.L. Outtz (Ed.), Adverse Impact: Implications for Organizational Staffing and High Stakes Selection (pp. 323–346). New York, NY: Taylor & Francis Group.


www.aamc.org

using mcat® data in 2016 medical student selection

Documents