Description
This paper explain using competencies to improve school turnaround principal success.
Using Competencies
to Improve School
Turnaround Principal
Success
by Lucy Steiner and
Emily Ayscue Hassel
About the Authors
LUCY STEINER is a senior consultant with Public
Impact. She researches and consults on a variety of critical
education issues, including teacher and leader policy,
school restructuring, charter school policy, and teacher
professional development. Ms. Steiner both conducts her
own work and leads project teams to deliver research,
training, and consulting. Her work ofen provides a bridge
between district leadership, school leadership, and instruc-
tion. A former high school English teacher, Ms. Steiner
holds a master’s degree in education and social policy from
Northwestern University and a B.A. with highest honors
from the University of North Carolina at Chapel Hill.
EMILY AYSCUE HASSEL is Co-Director of Public Im-
pact. She provides thought leadership and oversight to
Public Impact’s work on teacher and leader policy, or-
ganizational transformation, parental choice of schools,
and emerging opportunities for dramatic change in pre-K
to grade 12 education. Her work has appeared in Educa-
tion Week, Education Next, and other publications. She
previously worked for the Hay Group, a leading human
resources consulting frm. Ms. Hassel received her law and
master in business administration degrees from the Uni-
versity of North Carolina at Chapel Hill.
Acknowledgements
This report was made possible by the University of Vir-
ginia’s Darden/Curry Partnership for Leaders in Educa-
tion and its School Turnaround Specialist Program. The
authors are grateful to Julie Kowal and Bryan Hassel of
Public Impact and to LeAnn Buntrock and William Rob-
inson of the UVA School Turnaround Specialist Program
for their feedback on early drafs. We would also like to
thank Sharon Kebschull Barrett for careful editing, and
April Leidig-Higgins for the design of the report.
Using Competencies to Improve School Turnaround
Principal Success was made possible by the support of
© 2011 Public Impact, Chapel Hill, NC
© 2011 University of Virginia’s Darden/Curry Partnership
for Leaders in Education, Charlottesville, VA
Public Impact is a national education policy and manage-
ment consulting frm based in Chapel Hill, NC. We are a
team of researchers, thought leaders, tool-builders, and on-
the-ground consultants who help education leaders and
policymakers improve student learning in K–12 education.
For more on Public Impact and our research, please visit:
www.publicimpact.com.
Public Impact encourages the free use, reproduction, and
distribution of this working paper for noncommercial use.
We require attribution for all use. For more information
and instructions on the commercial use of our materials,
please contact us at www.publicimpact.com.
Please cite this report as:
Steiner, L., & Hassel, E. A. (Public Impact). (2011). Using
competencies to improve school turnaround principal success.
Charlottesville: University of Virginia’s Darden/Curry
Partnership for Leaders in Education. Retrieved from
www.DardenCurry.org
• 1
Introduction
I
n a time when student outcomes matter more
than ever, many states, districts, and reformers
are considering whether and how turnarounds of
chronically failing schools can achieve superior
results. In fact, we already know much about when and
how successful turnarounds work, both from other sec-
tors and from recent experience in education.
Two major factors afect turnaround success: the
characteristics and actions of the turnaround leader,
and the support for dramatic change that the leader
and staf receive from the district, state, and/or other
governing authority. Although leadership accounts for
25 percent of school efects in most schools,1 in a turn-
around the leader is paramount. It is almost unheard
of for turnarounds to occur without a special breed of
leader at the helm — one who engages and focuses the
whole community on achieving dramatic improvement
goals fast.2
This paper aims frst to shed light on one element
of leadership: the characteristics — or “competencies”
— of turnaround leaders who succeed in driving rapid,
dramatic change. Second, we recount the elements of
support that districts must provide these leaders to
enable and sustain a portfolio of successful school turn-
arounds. Fortunately, decades of experience in other
sectors and in education systems of other nations reveal
tools and techniques for understanding and using turn-
around leader competencies, and for governing turn-
around leaders successfully.
U.S. educators must act on this knowledge. Today,
few districts have an explicit strategy to select and em-
power school turnaround leaders using the best avail-
able techniques. Few provide the autonomy, support,
and accountability for rapid, dramatic change that
will attract, keep, and enable turnarounds by capable
leaders.
Here we explain what states, districts, and others
with an interest in school turnarounds need to know.
This paper:
• Describes how using competencies that predict
performance can improve turnaround principal
selection, evaluation, and development; and
• Summarizes prior research about how districts
can create the right environment to increase
school turnaround leader success.
In addition, the appendix provides details about
options for building valid competency models.
2 • usi ng competenci es to i mprove school tur naround pr i nci pal success
Turnaround Leader
Competencies
H
ere we describe the role of leader-
ship in organizational turnarounds
and how using competencies can allow
better selection, development, and
support of those leaders.
Turnaround Leaders:
A Special Breed of Leadership
Decades of research have documented that having
the right leader is an essential component of success-
ful turnarounds.3 Despite the intense national focus
on school turnarounds, signifcant barriers prevent
reformers from fnding and enabling leaders likely to
succeed in a turnaround.
Why? First, turnaround eforts are made when
organizations are in a state of entrenched failure. Lead-
ers who would otherwise succeed ofen fall short in a
turnaround. Turnaround success is a challenge even in
sectors where this strategy has been tried ofen. Stud-
ies across sectors suggest that only 30 percent of turn-
around eforts succeed.4 Even leaders who have excelled
in other circumstances may fail when faced with the
rapid, dramatic change required in a turnaround efort.
The current education leadership pool is unlikely to
have the number and type of candidates needed to lead
a large number of school turnaround eforts. Therefore,
recruitment must focus on candidates whose capabili-
ties ft the specifc demands of turnarounds.
Second, typical school district practices are not de-
signed to recruit and select talent for challenging schools,
including the bold leaders needed for turnaround
schools.5 Most districts base principal and staf hiring
on college degrees and years of experience. Numer-
ous research studies over several decades have shown
that degrees and experience (afer the frst few years of
teaching) are poor predictors of performance.6 As re-
formers focus on recruiting, they need new methods to
choose the right people for turnarounds.
Third, few districts measure performance diferences
among leaders and staf that would be useful for identify-
ing and developing internal candidates for school turn-
around leadership.7 Understanding the characteristics
needed to succeed in a turnaround would allow inter-
nal selection and development of high-potential candi-
dates from among current teachers and principals.
Understanding the Crucial Role
of Competencies
Performance diferences are large in difcult jobs:
research has found that the top 1 percent of jobhold-
ers in complex jobs produce results 127 percent better
than the average.8 Many organizations select employees
based on experience and degrees, hoping that these
indicators will predict success on the job. Yet many of
us have experienced the “experience and degrees” myth
frsthand. When two seemingly similar candidates are
hired — with the same level of education, experience,
and technical skills — one sometimes turns out to be an
outstanding performer, while the other struggles.
In the 1970s, a cognitive psychologist from Har-
vard University, David McClelland, tried to fnd out
why. He hypothesized and ultimately demonstrated
that habits of behavior and underlying motivations,
which he called “competencies,” diferentiate workers’
performance outcomes. As a result of his and others’
subsequent research, employers can understand not just
what employees do to be successful, but how they do it.9
By examining candidates’ competencies, employers can
usi ng competenci es to i mprove school tur naround pr i nci pal success • 3
uncover diferences likely to afect performance, help-
ing to choose between candidates who may otherwise
seem identical.
Competency-based performance management re-
mains relatively rare in education. But many organiza-
tions in other sectors (public, nonproft, and private)
use competencies for selection, development, and career
planning — even pay.10 Other nations, including Singa-
pore and the United Kingdom, now use competencies
throughout their education systems.11 To learn how
competency-based human capital systems work well,
we therefore need to turn to other sectors and nations
that have used them for many years.
For the past 10 years, competency-based evalua-
tion has been the bedrock of Singapore’s educational
system, one of the strongest in the world.12 In the early
2000s, Singapore implemented a competency-based
performance management system for the three major
roles in Singaporean schools — teachers, principals, and
school specialists.13 School ofcials use the competency
model in conjunction with the achievement of positive
student outcomes and other measurable results to set
annual competency targets, evaluate competency levels
throughout the year, match each educator to a career
path, and determine annual bonuses. In the decade
since introducing this system, Singapore has continued
to raise student performance and narrow the achieve-
ment gaps among its ethnic groups.14
By examining candidates’
competencies — habits of
behavior and underlying
motivations — employers can
understand how employees
succeed.
4 • usi ng competenci es to i mprove school tur naround pr i nci pal success
Research Base on Competencies
When Dr. McClelland began his research in the early
1970s, he was reacting to studies fnding that standard
ways of evaluating job candidates — IQ tests and other
tests of academic aptitude, knowledge content tests,
school grades, and academic credentials — did not fully
predict job performance and were ofen biased against
minorities, women, and people in poverty.15 In the
course of his research, he coined the term “competency”
to describe the behavioral characteristics that he found
could predict performance.
Although the term competency ofen describes any
work-related skill, in this context competencies refers
to the underlying motives and habits — patterns of
thinking, feeling, acting, and speaking — that cause a
person to be successful in a specifc job or role.16 Mc-
Clelland compared the competency patterns of very
high performers to those of typical performers. When
analyzing these high performers, he found that un-
derlying characteristics (e.g., persistence, achievement
motivation, self-confdence) led to actions (e.g., calcu-
lated risk taking, goal setting, planning) that in turn
led to better outcomes (e.g., efectiveness, productivity,
innovation).17 Research conducted in the decades since
McClelland’s original study has provided further evi-
dence that underlying competencies enable successful
performance in a given job or role.18
In the course of their research, McClelland and his
colleagues also developed a methodology for identify-
ing and validating the competencies for particular jobs
and roles. This interview technique — known as a be-
havior event interview (BEI) — combined elements of an
existing technique called the “critical incident method”
with probes about motivation that McClelland’s team
refned over several decades.19 Instead of asking people
to provide hypothetical responses to interview ques-
tions, BEIs ask them to walk interviewers through past
incidents step by step, as though the interviewee is
reliving the experience. This helps interviewees reveal
what they were thinking, saying, and doing at the time,
and makes it hard to claim credit for actions that the
individual did not take.20
Studies since McClelland’s original research indi-
cate that structured interviews such as the BEI that
probe for information about past events are highly
correlated with later job performance.21 For example,
an independent, comparative study of behavior-based
interviews and unstructured interviews at a large life
insurance company found that the behavior-based
interviews yielded a validity coefcient of .48 using
supervisor ratings as the performance criterion, and
.61 using sales dollars as the performance criterion. In
contrast, the standard interview yielded a validity co-
efcient of .08 and .05, respectively.22 McClelland’s last
published study found that 65 percent to 86 percent
of managerial candidates who met a threshold level of
competence when selected using a model constructed
with BEIs ended up in the top third of performers,
compared with 11 perent to 20 percent of candidates
who scored lower in competence during the selection
process.23
Behavior event interviews
probe for information about
past events to predict future job
performance.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 5
Imagine the consequence for children of selecting
turnaround principals as accurately: candidates meet-
ing threshold levels of competencies could be far more
likely to succeed in turnaround attempts than candi-
dates who fall short.
The “iceberg model” (Figure 1) was developed by
the Hay Group, an international human resource frm
where McClelland spent the latter part of his career,
to demonstrate how competencies relate to observable
qualifcations, knowledge, and skills. As this model
demonstrates, competencies are more difcult to de-
tect than qualifcations, skills, and knowledge, but
they largely infuence these observable behaviors.24
Competency research further suggests that outstand-
ing performance in complex jobs — ones in which
most candidates have a similar educational history
and signifcant autonomy over daily work tasks — is
driven more by underlying competencies than by read-
ily observed skills and knowledge. Individuals in these
complex jobs, such as school principals, use their similar
content knowledge very diferently to accomplish work
goals. For example, some proactively set difcult goals
and stick to them, while others do only what is asked
by superiors or give up when a goal proves difcult to
achieve. Some try to do all the work themselves, while
others identify colleagues’ strengths and put them into
roles where they will succeed. This wide variation in
how people work produces greatly varying results un-
explained by prior knowledge, degrees, and experience,
making competency-based performance management
practices especially critical.25
In the past 40 years, dozens of competencies have
been identifed as success distinguishers in diferent
jobs and roles, in combinations that are unique to each
particular role. Nevertheless, two competencies appear
critical to high levels of success in most complex leader-
ship jobs: “achievement” and “impact and infuence.”26
• Achievement is defned as “the drive and actions to
set challenging goals and reach a high standard of
performance.” In a leader, achievement includes “set-
ting high performance goals for the organization,
prioritizing activities to achieve the highest beneft
relative to inputs, and working to meet goals using
direct action, staf, and other available resources.”27
• Impact and infuence is “acting with the purpose
of afecting the perceptions, thinking and actions
of others. It includes empathizing with others and
Figure 1: Observable characteristics versus underlying competencies (Adapted from “The Iceberg Model” in Spencer &
Spencer, Competence at Work, p. 11.)
qualifications
Certifcation
Degrees {
{
Knowledge and skills
Teacher practices
competencies
Recurring patterns of thought and action
Example: Achievement Drive {
6 • usi ng competenci es to i mprove school tur naround pr i nci pal success
anticipating likely responses to situations, tailoring
actions and words to create an intended impact, and
giving and withholding information to obtain spe-
cifc responses.”28
Leaders in diferent situations — such as start-ups,
large organizations, and turnaround eforts — might
need to use these competencies in difering degrees
and ways.29 But a literature review of the actions that
successful turnaround leaders take indicates that high
levels of competence in both achievement and impact
and infuence are most likely essential for school turn-
around principals.30
Identifying Distinguishing Competencies:
Building a Valid Model
What makes competency-based performance manage-
ment strategies powerful is the potential to correlate
qualitative characteristics with performance outcomes
in a statistically valid manner. A good competency
model includes descriptions of both the competencies
that are needed to succeed in a job and the increasing
levels of performance within each competency. Most
important, in a valid model, the competencies and
increasing levels correlate with performance outcomes,
such as student learning gains.
Organizations should use competency models built
using an approach that is as valid and predictive of per-
formance as circumstances allow.31
• Building a competency model fom scratch. A model
can be built from scratch when there is an accessible,
large set of jobholders who have been on the job for
several years and who can be classifed as outstand-
ing and average.
• Building a model fom related, validated models —
the stepladder approach. Emerging roles, jobs in
emerging sectors, and jobs spread out among many
smaller organizations may not have enough acces-
sible performers for a data set of outstanding and
typical performers whom researchers can compare.
In these cases, a model can be extrapolated by map-
ping the actions needed for success in the job to
similar jobs for which validated competency models
are available.
In either case, a model can be validated and refned
over time by comparing competency ratings during se-
lection to later performance outcomes. Subsequent hir-
ing can emphasize competencies that most accurately
sort high and typical performers. Appendix A provides
more detail about the various options for building and
validating competency models.
Using Competencies to Ensure Efective
School Turnaround Leadership
Organizations can use competencies for many pur-
poses. Here we briefy describe three important uses
for school turnarounds: 1) hiring efective turnaround
principals; 2) evaluating principal performance; and 3)
providing targeted development for school turnaround
principals. Using performance-predictive competencies
at each of these critical stages of an employee’s career
increases the likelihood of improving employee perfor-
mance in key results areas.
Hiring efective school turnaround principals
Selecting people who already have most of the compe-
tencies needed for turnaround leadership, rather than
relying on long-term development, may be the best way
to achieve the rapid results a turnaround demands.32
A competency model that will be used for selec-
tion should include competency descriptions, levels,
and tools for rating and comparing candidates. For
example, some states and the UVA School Turnaround
Specialist Program use Public Impact’s competency
model for selecting school turnaround principals.33 It
includes: 1) short, broad defnitions of the competen-
cies that distinguish high performance; 2) rating scales
of increasingly efective levels of behavior within each
competency; 3) competency level targets for the job of
school turnaround principal; and 4) selection steps and
guidelines for assessing candidates’ competency levels
using the behavior event interview (BEI).34 Appendix B
provides more detail about how hirers can use the BEI
to assess candidate’s competencies.
Evaluating school turnaround principals
Considerable evidence indicates that current principal
evaluation practices are similar in quality to the inefec-
usi ng competenci es to i mprove school tur naround pr i nci pal success • 7
tual teacher evaluation practices that have been widely
criticized.35 According to researchers, principal evalua-
tion too ofen consists of a binary rating sheet that asks
the evaluator to check of “satisfactory” or “unsatisfac-
tory” on a number of items such as “time management”
or “demonstrates efective organizational skills.”36 This
kind of evaluation fails to deliver the information that
districts need to evaluate principals’ performance accu-
rately and that principals need to improve.
Including measurable results in principal evaluation
could correct this situation. The arguments for and
against this shif are beyond the scope of this report,
but basing at least part of any evaluation on student
outcomes would be consistent with the best practices of
many high-performing organizations in both the pub-
lic and private sectors.37
Evaluations rarely hinge solely on results, though.
Outcomes alone do not give employers information to
help with promotion and job placement decisions, nor
do they give employees information about how to im-
prove.38 A complete evaluation system includes not only
measurable results, but also professional skills, such as
curriculum planning, and of course the competencies
that are critical for achieving results. Understanding
the competencies a principal demonstrates during a
school turnaround efort and rating the principal’s per-
formance against these competencies in an evaluation
can help employers understand why a leader is succeed-
ing or falling short — and whether ultimate success is
likely. When the competency gaps are large and early
indicators of progress are poor, a leader may need to be
replaced rather than developed.39
Developing school turnaround principals
Waiting to develop a leader on the job means a sure
recipe for failure when fast results are essential. But
even very competent turnaround principals will have
areas of competency weakness, and they also will need
to develop turnaround competencies among the staf
leaders on their teams. Fortunately, some research in-
Principal evaluation too ofen
consists of a binary rating
sheet that asks the evaluator
to check of “satisfactory” or
“unsatisfactory.”
When the competency gaps are
large and early indicators of
progress are poor, a leader may
need to be replaced rather than
developed.
8 • usi ng competenci es to i mprove school tur naround pr i nci pal success
dicates that leaders can continue improving individual
competencies with the right training, when it is clearly
linked to outcome goals.40
Indeed, early turnaround results can provide the es-
sential motivation principals and staf leaders need to
improve.41 In a school turnaround, leaders must rapidly
identify failing approaches and try new tactics.42 Those
who understand their own challenges are in a better
position to make these changes fast.
The superintendent (or other direct supervisor)
and the principal can identify the principal’s com-
petency levels — for example, with ratings based on a
combination of supervisor, staf, parent, and student
input — and compare these to levels needed for supe-
rior performance. The principal can then focus on
closing his gaps and using his strengths more ofen. A
valid model with progressively more efective levels of
competence aids development by giving school lead-
ers specifc next-step actions needed for better student
outcomes.43
For example, a principal might have a current rating
of “4” on the impact and infuence competency, which
indicates that she regularly “thinks ahead about the
likely reaction of the audience and takes two or more
steps that are calculated to obtain desired impact.” In
order to improve, she and her supervisor might set a
goal for her to move to level 5 by more efectively and
consistently “using others (e.g., parents, staf members)
to obtain desired impact.”44 Appendix C provides more
details about how principals can continue improving
their individual competencies with the right training.
But even the best leader’s eforts can be thwarted
by an environment unsupportive of change. So next
we summarize the elements of a district environment
in which turnarounds are most likely to occur — and
succeed.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 9
Creating an External
Environment that
Supports Turnarounds
T
he research on turnarounds in educa-
tion and other sectors suggests that multiple
environmental factors infuence an orga-
nization’s ability to improve rapidly and
that even the best leader’s eforts can be frustrated and
diminished by an unsupportive environment. Districts
that want to increase the odds of successful school
turnarounds should take an active leadership and sup-
port role. The following steps for districts are drawn
from a research brief written by Public Impact that ap-
plies cross-sector and education research to the district
role in turnarounds:45
• Commit to success. Policymakers overseeing the
turnaround efort — state department ofcials,
district leaders, school board members — must pri-
oritize student learning needs over the customs, rou-
tines, and established relationships that can stand
in the way of necessary change. They must view
turnarounds not as a one-time solution but as part
of a sustained efort to eliminate chronic low perfor-
mance, and must be willing to stay the course even
when some frst attempts fail. Policymakers need
to assess their own capacity to oversee and support
dramatic and sometimes disruptive change before
committing to this strategy.
• Choose the right schools. Turnarounds are a neces-
sary step in schools where student performance is
extremely and chronically low and where incremen-
tal eforts to improve student outcomes have failed.
• Give leaders the “big yes.” Successful turnaround
leaders ofen achieve results by working around
rules, asking for forgiveness afer their strategy has
worked rather than seeking permission before-
hand.46 By giving turnaround leaders the “big yes”
on critical autonomies — stafng decisions, schedul-
ing, budgeting, and other operational issues — poli-
cymakers can help support successful turnarounds.
Districts that want to increase
the odds of successful school turn-
arounds should take an active
leadership and support role.
Policymakers overseeing the turn-
around efort must prioritize
student learning needs over the
routines that can stand in the way
of necessary change.
10 • usi ng competenci es to i mprove school tur naround pr i nci pal success
•Proactively engage the community. Dramatic
change requires active communication with local
stakeholders. Successful eforts to engage the com-
munity are characterized by public acknowledge-
ment of past failures coupled with a forceful, posi-
tive vision for the future.47 Publicizing early “wins”
can also send a powerful message that change is pos-
sible and turnarounds can work.48
• Hold leaders accountable for results. Policymakers
must hold turnaround leaders to high standards and
a short timeline for results. The research literature
does not indicate an exact timeline required to turn
an organization around, but in successful turn-
around eforts, fast, focused changes occur in the
frst few months, and substantial improvements in
the frst year.49
• Develop a talent pipeline. District leaders need
to build their supply of turnaround leaders and
teachers through proactive recruitment, careful
selection, targeted training, and strategic placement
in turnaround schools. The skills and abilities of
principals and teachers who succeed in turnarounds
difer from those of their peers who succeed in less-
challenging schools. Competency screening should
be a critical step in the hiring process.
Publicizing early “wins” can
also send a powerful message
that change is possible and turn-
arounds can work.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 11
A Call to Action
T
he U.S. desperately needs a strong
cadre of school leaders who can turn around
persistently low-performing schools. But
today, this cadre is far too small. States
and districts that are serious about eliminating broad-
scale failure in schools must use the very best tools
available to select, evaluate, and develop these school
turnaround leaders. Current practices — inconsistent
hiring, uneven support, and weak evaluation — are
severely inadequate. Competency-based people-man-
agement, coupled with the right district environment,
can signifcantly increase the number and performance
of school turnaround leaders. What steps would make
this urgent priority a reality?
Disseminate information about competency-
based practices to key stakeholders in education.
Too few education leaders and policymakers know
about competency-based selection, evaluation, and
development — strategies commonly used in other sec-
tors. Even fewer know that districts can validate and
improve competency models by comparing competency
ratings with performance outcomes. Similarly few
realize how diferent the competencies needed for the
role of turnaround leader are from those needed in
traditional principal jobs. Audiences that would beneft
from better awareness include district leaders, state
policymakers, philanthropists, parent and community
advocacy organizations, and national and local school
leadership training programs.
Invest in competency models for critical school
leadership (and teaching) roles. The best competency
models are based on data from behavioral event inter-
views and correlate with performance outcomes. States
and districts in the U.S. have not invested in this type
of rigorous model building, because of the investment
involved and low awareness of the value. If the United
States is going to dramatically improve teacher and
leader performance, then some combination of dis-
tricts, states, the federal government, and private foun-
dations must invest in the research, development, and
improvement of competency models for critical roles
in education. As the pool of serious school turnaround
attempts grows, validation and improvement of com-
petency models for leaders and teachers in this context
will be possible.
Select school turnaround leaders for competence.
Our nation must identify far more leaders to turn
around persistently failing schools. Competency-based
selection for critical leadership (and other important)
positions would enable selection from a much wider
labor pool — turnaround leaders from other sectors
and emerging teacher-leaders, for example. Other sec-
tors and nations have used this approach widely, yet it
remains rare in U.S. education.
Develop strong competency-based training pro-
grams for school turnaround principals. A critical ap-
plication of competency models is development. Educa-
tion leaders should add and expand training programs
for school turnaround principals that incorporate
competency-based practices, such as the University of
Virginia School Turnaround Specialist Program.
Evaluate and publicize results. As schools, districts,
and states undertake more school turnarounds, they
must collect data to compare successful and less-suc-
cessful leaders in these unique settings, and publicize
this information widely. These analyses can validate in-
dividual competencies — and perhaps identify new ones
as more data emerge — as well as provide rich examples
of competencies in action.
• 13
Appendix A
Identifying Distinguishing Competencies
Building a Valid Model
Two aspects of building a competency model are criti-
cal to validity, regardless of approach. The frst is to use
a sample of employees that compares truly outstanding
performers with people who are average performers in
their jobs. The second is to collect data about compe-
tencies that give more weight to behaviors that jobhold-
ers have actually displayed on the job than to what they
or others think they should do.50
In the following section, we describe two approaches
to building a valid competency model, noting areas
where model-builders have a choice of methods.
Building a competency model from scratch
This section describes the most customized way to
build a valid competency model.51 Using this method,
however, requires a large data set of jobholders who
have been on the job for several years and who can
clearly be classifed as outstanding or average. While
this might be possible someday for turnaround leaders,
the techniques described in the next section for
building — and then validating — models with limited
data sets are more appropriate today.
We are not aware of any public education organiza-
tion in the United States that has built a competency
model from scratch using the rigorous method de-
scribed here — for any job, including jobs that, unlike
turnaround leader, are already very prevalent. Because
several of the steps described below are costly and
require expertise, building a model that meets this
high standard of rigor and efectiveness represents a
considerable investment of resources. In positions with
numerous jobholders, such as traditional principal and
teaching roles, the cost per job of this method would
be small. Leaders of states and large districts, take note:
one large investment could provide valid, performance-
predictive tools to nearly every school in a district or
state — or nationwide.
From-scratch: Building a model with behavior
event interviews:52
• Determine performance criteria. First, determine
what constitutes outstanding performance. Do this
in consultation with a range of people who under-
stand the role, and, ideally, include data both quan-
titative (e.g., the magnitude and speed of student
learning gains) and qualitative (e.g., parent and staf
ratings).
• Select criterion sample. Researchers should then
select two groups of current jobholders, one that
has displayed average performance according to the
performance criteria, and another that has displayed
truly outstanding performance on the same set of
measures. The samples need to be large enough to
allow for statistical analysis.
• Collect data. The most efective method for col-
lecting data is the structured interview technique
mentioned earlier — the behavior event interview
(BEI).53 Unlike other interview techniques that ask
candidates to respond to hypothetical situations,
the BEI elicits detailed stories of past events that
reveal how top performers difer from more typical
or lower-performing jobholders. During this stage
of the model-building process, avoid bias by ensur-
ing that neither the interviewee nor the interviewer
knows if they are in the “outstanding” or “average”
sample. Because the BEI is also a highly efective
method to use for actual candidate selection afer
the model has been validated, it is described in detail
in the section below on selection.
• Develop model. Analyzing interview transcripts
to determine the diferences between average and
outstanding performers is the most complex stage
of the process, one that requires qualitative re-
view and coding as well as statistical analysis. The
goal is to determine what outstanding performers
14 • appendi x a
do (e.g., actions, thoughts, feelings) that average
performers do not do, and vice versa. Research-
ers then convert these fndings into interval scales
that identify a “threshold” level that describes the
minimum requirements for average performance in
each competency, as well as descriptions of levels of
increasingly efective behaviors associated with this
competency.54
For example, the competency called “Initiative
and Persistence” may be a critical competency for
turnaround leaders. It involves having the “drive and
actions to do more than is expected or required in
order to accomplish a challenging task.” As the scale
increases, so does the complexity of actions associ-
ated with this competency, from “voluntarily initi-
ates and follows through on new work project that
is not assigned by others” to “acts without formal or
explicit authority, takes personal or career risks and
bends organization norms or rules to accomplish a
work objective.”55
Afer the descriptors are written, they are then
tested for inter-rater reliability and refned as
needed. The fnal competency list, competency de-
scriptions, descriptions of diferent levels of perfor-
mance within a competency, and information about
coding are included in a document that becomes the
“competency model” for a particular job.
• Validate model. While validation is ideal, many
organizations building models from scratch rely
on the rigor of the initial process to produce a
valid model. When time and funds allow, several
methods can validate a competency model. Choos-
ing a second sample of top and typical performers,
conducting BEIs, and analyzing correlation of their
competencies with actual performance outcomes
is one method. Another rigorous method is to as-
sess incoming candidates using data from BEIs,
then analyze whether those who scored higher in
the selection process perform better in their jobs
according to the performance criteria.56 Because
the competency model is designed to predict actual
performance on the job, testing the model against
performance results is the most powerful way to
validate the model. The other advantage of this
method is it allows the model designers to revise and
refne based on actual performance.57
Stepladder approach: Building a model
from related, validated models
Two barriers can prevent building models from scratch:
cost and limited past data about high performers. First,
in many cases like school turnaround leaders, limited
data are available to build a competency model from
scratch. Emerging roles, jobs in emerging sectors, and
jobs spread out among many smaller organizations may
not have enough accessible performers for a data set of
outstanding and typical performers whom researchers
can compare. Second, models are expensive to build
from scratch, particularly when jobholders are spread
out geographically. What follows is a description of
how to achieve a valid competency model at a lower
cost and with limited data and access to jobholders.
• Determine performance criteria. This step is criti-
cal and should be implemented as described above.
• Select criterion sample. In selecting the two groups
for analysis, model builders have some leeway with
regard to sample sizes. The question to ask is, what is
the minimum sample size needed to produce a valid
result? Expert opinion suggests that it is better to in-
clude a larger sample of star performers, if possible,
because they are the best source for detailed infor-
mation about outstanding performers.58 Sometimes,
for example in an entirely new role, model builders
will need to envision expected actions of performers
likely to achieve outstanding results. In this case,
focus groups may still be useful, but extrapolation
from existing, valid models of related jobs will be
essential.
• Collect data. In addition to BEIs (referenced above),
there are other data collection techniques:
— Focus groups. This involves asking groups made
up of people who know the job well to identify
the competencies that are critical in a given job
or role. Experience indicates that about half the
competencies identifed by focus groups are vali-
dated by a full competency study using BEIs.59
appendi x a • 15
—360-degree surveys. Surveys ask superiors, peers,
subordinates, and external people who interact
with jobholders to rate whether particular com-
petencies are important for superior performance,
how ofen they are needed, and whether failure is
likely if someone does not have this competency.
Surveys are useful because they are quick and
cost-efective to administer, and they can provide
enough data for valid statistical analysis. How-
ever, survey designers may miss critical competen-
cies in creating the survey.
— Extrapolate from existing models of similar
jobs. Sometimes, top and typical performers
are not accessible or available in numbers large
enough to use the from-scratch method. This is
especially likely in new roles and in organizations
undergoing signifcant change. In these situations,
model builders can look at validated competencies
that distinguish performers in other roles and use
these to identify actions needed for job success.
School turnaround leaders are one example: until
an identifable, accessible population of perform-
ers who can clearly be labeled superior or typical is
available for study, extrapolated models will have
to sufce.60 This approach is a relatively quick
way to build a model, but still requires deep un-
derstanding of the job, the related jobs for which
validated competency models have already been
built, and the available competency models or
dictionaries. However, without access to detailed
information about superior and average perform-
ers in this exact role, the model may be less valid
initially than a from-scratch model would be, if it
were feasible. Validation is especially important
for models built from related job models.
• Develop model. Experts analyze the data from
focus groups, surveys, and existing competency
models or dictionaries to choose the competencies
likely to distinguish top performers and to identify
the “threshold” levels of behavior for solid and out-
standing performance. This stage requires expertise:
thorough understanding of the jobs, thoughtful
analysis of the data, and, when surveys are used,
the ability to conduct statistical analysis. However,
model builders do not identify new competencies
as they would when analyzing BEIs. The fnal out-
come — a competency model — is similar but less
customized than a from-scratch model.
• Validate model. Models built this way can be re-
fned over time, with use and deliberate validation
analyses. There are several options for validating the
competency model beyond comparing BEI results
with actual job performance. For example, design-
ers can conduct BEIs on incoming jobholders. Once
enough have been interviewed for statistical valid-
ity and enough time has passed to observe their
performance, experts can determine the extent to
which each competency in the model accurately
distinguishes outstanding and average performers.
Conducting BEIs for validation requires the same
expertise as BEIs for model building and thus is rela-
tively costly.
Another less-expensive approach when large
numbers of jobholders are available is to design
questionnaires that ask them — both outstanding
and average performers — to respond to questions
based on the model (e.g., how ofen in the past
two months have you taken on a voluntary task at
work?).61 If the model was well-designed, then out-
standing and average performers may be accurately
identifed by their responses and the competency
model validated. 360-degree assessments by peers,
subordinates, and supervisors can be used similarly
to validate.
16 •
Appendix B
Hiring Efective School Turnaround Principals
Hiring people who already have most of the competen-
cies needed for turnaround leadership, rather than rely-
ing on long-term development, may be the best way to
achieve the rapid results a turnaround demands.62
Afer prescreening for other requirements, hirers
can assess candidate competencies using these steps:63
• Step 1: Conduct behavior event interview. In a BEI,
the interviewer’s goal is to understand in detail how
candidates perform various aspects of their work. To
do this, the interviewer asks candidates to recall past
events when they have felt successful or have dealt
with specifc situations at work (e.g., a time when he
or she infuenced another person, or led a team of
people to accomplish work that was satisfying). The
candidate should spend 15 minutes or more describ-
ing the incident in great detail, with the interviewer
probing insistently for the information needed to
understand exactly what a person was thinking
or doing at the time. According to Hay Group re-
searchers, sample probes include: “What led up to
the situation? Who was involved? What did you
think about, feel, want to have happen in the situa-
tion? What did you do? What was the outcome?”64
The interviewer should: 1) probe insistently for
detail; 2) keep the candidate focused on past events
rather than refecting on hypothetical situations or
using generalities; and 3) take comprehensive notes
or record what candidates say so that their responses
can be used later for scoring.
• Step 2: Rate candidate’s competency levels. The
interview team then closely reviews the candidate’s
responses and notes any examples of “codable data,”
or data that is valid for scoring against the compe-
tency model. To be codable, responses must be in
the frst person (“I did this” rather than “we did
it”), be about real rather than hypothetical actions
and feelings, be volunteered by the candidate inde-
pendently, and be about past rather than present
feelings or behavior. Afer noting the codable com-
ments, the interview team compares them against
the competency level descriptions and rates the can-
didate on each competency.
• Step 3: Make hiring decisions. Afer each can-
didate has been rated on the competencies, hirers
compare the strengths (and weaknesses) of the can-
didates who meet all or most competency thresholds
to determine whom to hire. The number of slots
available compared with the number of qualifed
candidates — and the level of challenge in each
school — might afect how many are hired and for
which schools.
• Step 4: Collect performance data and revise in-
terview. Afer an initial round of hires, hirers can
compare competency scores at selection with actual
performance. Future hiring can focus on the compe-
tencies that best predict performance.65
In addition, most selected candidates will fnd
feedback about their competency scores helpful both
for using strengths with confdence and for choos-
ing staf and assigning roles that balance each leader’s
weaknesses.
• 17
Appendix C
Developing School Turnaround Principals
Even very competent turnaround principals will have
areas of competency weakness, and they also will need
to develop turnaround competencies among the staf
leaders on their teams. Fortunately, some research in-
dicates that leaders can continue improving individual
competencies with the right training, when it is clearly
linked to outcomes goals.66
Options to help school turnaround principals de-
velop specifc competencies include:
• Competency training. Competency training ap-
pears to be most efective when trainers: 1) present
compelling evidence that competency improvement
will make employees better at their jobs; 2) give
feedback to employees on how their own levels of
competence compare with outstanding performers;
3) give employees opportunities to practice compe-
tency behaviors; and 4) expect employees to set com-
petency development goals with action plans.67
• Self-development resource guides. Resource
guides instruct principals about how to develop
role-specifc competencies. They can include written
cases or video clips highlighting examples of compe-
tency behaviors at diferent levels, suggestions about
practice activities, and instructions on how to access
training and mentoring opportunities to improve
specifc competencies.68
• “Stretch” roles or assignments. Aiming some ac-
tivities toward improving weaknesses can hasten
development. For example, a principal who lacks
self-confdence might establish a goal to give several
presentations at community meetings where he
practices “openly stating his own expertise or com-
paring himself positively with others.”69
• Mentoring. A mentor who is very strong in a prin-
cipal’s areas of weakness can provide rapid feedback
and guidance about improvement. Districts should
assign mentors with the explicit expectation that the
mentor will coach the principal in specifed areas
needing development. Research suggests that devel-
opment cannot be imposed on another person.70
• 19
Notes
1. Leithwood, K., Seashore Louis, K., Anderson, S., &
Wahlsrom, K. (2004). How leadership infuences student
learning. NY: Wallace Foundation. Retrieved fromhttp://mt.educarchile.cl/MT/jjbrunner/archives/libros/
Leadership.pdf
2. Hassel, E. A., & Hassel, B. C. (2009). The big u-turn:
How to bring schools from the brink of failure to stellar
success. Education Next, 9(1), 21–27. Retrieved from http://
educationnext.org/the-big-uturn/
3. Herman, R., Dawson, P., Dee, T., Greene, J., Maynard,
R., Redding, S., et al. (2008). Turning around chronically
low-performing schools: A practice guide. Washington,
DC: National Center for Education Evaluation and
Regional Assistance, Institute of Education Sciences, U.S.
Department of Education. Retrieved fromhttp://ies.ed.gov/
ncee/wwc/pdf/practiceguides/Turnaround_pg_04181.pdf;
Bossidy, L. (2001, March). The job no CEO should delegate.
Harvard Business Review, 79(3), 47–49; Brenneman, G.
(1998, September-October). Right away and all at once:
How we saved Continental. Harvard Business Review,
76(5), 162–179; Buchanan, L. (2003, December). The
turning of Atlanta. Harvard Business Review, 81(12), 18–19;
Hamel, G. (2000, July-August). Waking up IBM: How
a gang of unlikely rebels transformed Big Blue. Harvard
Business Review, 78(4), 137–146; Hirschhorn, L. (2002,
July) Campaigning for change. Harvard Business Review,
80(7), 6–11; Joyce, P. (2004, August). The role of leadership
in the turnaround of a local authority. Public Money &
Management, 24(4), 235–242; Kanter, R.M. (2003, June).
Leadership and the psychology of turnarounds. Harvard
Business Review, 81(6), 58–67; Kim, W.C. & Mauborgne, R.
(2003, April). Tipping point leadership. Harvard Business
Review, 81(4), 60–69; Riesiner, R.A.(2002, February).
When a turnaround stalls. Harvard Business Review, 80(2),
45–52.
4. Beer, M. & Nohria, N. (2000). Breaking the code of
change. Cambridge, MA: Harvard Business School Press.
5. For more information on how school district practices
ofen fail to support efective hiring practices, see: The
Teaching Commission. (2006). Teaching at risk: Progress &
potholes. The Teaching Commission; National Council on
Teacher Quality. (2007). 2007 State teacher policy yearbook.
Retrieved fromhttp://www.nctq.org/stpy/reports/
stpy_national.pdf; The New Teacher Project. (2007).
Hiring, assignment, and transfer in Chicago public schools.
Washington, DC: The New Teacher Project. Retrieved
fromhttp://www.tntp.org/fles/TNTPAnalysis-Chicago.
pdf; The New Teacher Project. (2009). The widget efect:
Our national failure to acknowledge and act on diferences in
teacher efectiveness. New York: The New Teacher Project.
6. Goldhaber, D. D. & Brewer, D. J. Does teacher
certifcation matter? High school teacher certifcation status
and student achievement. Educational Evaluation and Policy
Analysis, 22(2), 129-145; Kane, T. J., Rockof, J. E., & Staiger,
D. O. (2006). What does certifcation tell us about teacher
efectiveness? Evidence fom New York City (NBER Working
Paper No. W12155). Cambridge, MA: National Bureau
of Economic Research. Retrieved fromhttp://gseweb.
harvard.edu/news/features/kane/nycfellowsmarch2006.
pdf; Walsh, K., & Tracy, C. O. (2005). Increasing the odds:
How good qualities can yield better teachers. Washington,
DC: National Council on Teaching Quality; Goldhaber,
D. (2002). The mystery of good teaching. Education Next.
Stanford, CA: Hoover Institution; Kane, T. J., Rockof,
J.E., & Staiger, D. O. Photo fnish: Teacher certifcation
doesn’t guarantee a winner. Education Next , 7(1); Chaney,
B. (1995). Student outcomes and the professional preparation
of eighth-grade teachers in science and mathematics. National
Science Foundation. For more on false assumptions about
performance predictiveness of traditional measures, see
McClelland, D. (1973). Testing for competency rather than
intelligence. American Psychologist, 28, 1–14.; Goldhaber, D.,
& Brewer, D. J. (2000).
7. Ibid.
8. Hunter, J. E., Schmidt, F. L., & Judiesch, M. K. (1990).
Individual diferences in output variability as a function of
job complexity. Journal of Applied Psychology, 75, 28–42.
9. Spencer, L. M. & Spencer, S. M. (1993). Competence
at work: Models for superior performance. New York: John
Wiley & Sons, Inc.
10. Examples of education organizations in the U.S. that
incorporate competencies in their selection process include
the University of Virginia School Turnaround Specialist
Program, Teach for America, The New Teacher Project,
and The Academy of Urban School Leadership.
11. Steiner, L. (2010). Using competency-based evaluation
to drive teacher excellence: Lessons fom Singapore.
Chapel Hill, NC: Public Impact. Retrieved from http://
opportunityculture.org/images/stories/singapore_
lessons_2010.pdf; Hobby, R., Crabtree, S., & Ibbetson, J.
(2004). The school recruitment handbook: A guide to
attracting, selecting and keeping outstanding teachers.
London and New York: RoutledgeFalmer.
12. For the past four years (2006–2009), Singapore has
ranked among the top four countries in the world on the
Trends in International Mathematics and Science Study
(TIMSS) science and math tests and the Progress in
International Reading Literacy Study (PIRLS) reading test.
20 • notes
See: Mullis, I. V. S., Martin, M. O., & Foy, P. (2008). TIMSS
2007 international mathematics report: Findings fom IEA’s
trends in international mathematics and science study at
the fourth and eighth grades. Cambridge, MA: TIMSS
& PIRLS International Study Center, Boston College.
Retrieved fromhttp://timss.bc.edu/TIMSS2007/intl_
reports.html; Martin, M.O., Mullis, I.V.S., & Foy, P. (2008).
TIMMS 2007 international science report: Findings fom
IEA’s trends in international mathematics and science study
at the fourth and eighth grades. Cambridge, MA: TIMSS
& PIRLS International Study Center, Boston College.
Retrieved fromhttp://timss.bc.edu/TIMSS2007/intl_
reports.html; Mullis, I.V.S., Martin, M.O., Kennedy, A.M.,
& Foy, P. (2006). IEA’s progress in international reading
literacy study in primary school in 40 countries. Cambridge,
MA: TIMSS & PIRLS International Study Center, Boston
College. Retrieved fromhttp://timss.bc.edu/pirls2006/
intl_rpt.html
13. Steiner, L. (2010).
14. Mourshed, M., Chijioke, C., & Barber, M. (2010).
How the world’s most improved school systems keep getting
better. McKinsey. Retrieved fromhttp://ssomckinsey.
darbyflms.com/reports/EducationBook_A4%20
SINGLES_DEC%202.pdf
15. Spencer & Spencer. (1993).
16. Spencer & Spencer. (1993).
17. Spencer, L. M., McClelland, D. C., & Spencer, S. M.
(1992). Competency assessment methods: History and state of
the art. Hay/McBer Research Press.
18. Lees, A. & Barnard, D. (1999). Highly efective
headteachers: An analysis of a sample of diagnostic data
fom the Leadership Programme for Serving Headteachers.
London: The Hay Group; Boyatzis, R. E. (1999). Developing
emotional intelligence (unpublished paper). Cleveland:
Case Western Reserve University, Department of
Organizational Behavior; Goleman, D. (1998). Working with
emotional intelligence. NY : Batnam; Jacobs, R. L. (2001).
Using human resource functions to enhance emotional
intelligence. In C. Cherniss & D. Goleman (Eds.), The
emotionally intelligent workplace. San Francisco: Jossey-
Bass; Nygren, D. J. and Ukeritis, M. D. (1993). The future of
religious orders in the United States. NY: Praeger; Spencer,
L. M. (2001). The economic value of emotional intelligence
competencies and EIC-based HR programs. In C. Cherniss
& D. Goleman (Eds.), The emotionally intelligent workplace.
San Francisco: Jossey-Bass; Williams, D. (1994). Leadership
for the 21
st
Century: Life insurance leadership study. Boston:
Hay Group; Boyatzis, R. E., Goleman, D., & Rhee, K.
(2000). Clustering competence in emotional intelligence:
Insights from the Emotional Competence Inventory (ECI)
s. In R. Bar-On and J. D. A. Parker (Eds.), Handbook of
emotional intelligence. San Francisco: Jossey-Bass, 343–362.
19. The critical incident method was developed by J. C.
Flanagan. See Flanagan, J. C. (1954). The critical incident
technique. Psychological Bulletin, 51, 327–358. McClelland’s
motivation probes are described in McClelland, D. (1989).
Human motivation. Cambridge, UK: Cambridge University
Press.
20. Spencer & Spencer. (1993).
21. Taylor, P. J. & Small, B. (2002). Asking applicants
what they would do versus what they did do: A meta-
analytic comparison of situational and past behavior
employment interview questions. Journal of Occupational
and Organizational Psychology, 75, 277–294; Krajewski,
H. T., Gofn, R. D., McCarthy, J. M., Rothstein, M. G., and
Johnston, N. (2006). Comparing the validity of structured
interviews for managerial-level employees: Should we look
to the past or focus on the future? Journal of Occupational
and Organizational Psychology, 79, 411–432; Pulakos, E. D.
and Schmitt, N. (1995). Experience-based and situational
interview questions: Studies of validity. Personnel Psychology,
48, 289–308; Boyatzis, R. E. (1982). The competent manager:
A model for efective performance. New York: Wiley-
Interscience; McDaniel, M. A., Whetzel, D. L., Schmidt,
F. L., & Maurer, S. D. (1994). The validity of employment
interviews: A comprehensive review and meta-analysis.
Journal of Applied Psychology, 79(4), 599–616; Motowidlo,
S. J., Carter, G. W., Dunnette, M. D., Tippins, N., Werner,
S., Burnett, J. R., & Vaughan, M. J. (1992). Studies of
the structured behavioral interview. Journal of Applied
Psychology, 77(5), 571–587; Campion, M.A., Campion, J.E.
(1994). Structured interviewing: A note on incremental
validity and alternative question types. Journal of Applied
Psychology, 79(6), 998–1002.
22. Orpen, C. (1985). Patterned behavior description
interviews versus unstructured interviews: A comparative
validity study. Journal of Applied Psychology, 70, 774–776.
23. McClelland, D. (1998). Identifying competencies with
behavioral event interviews. Psychological Science, 9, 331–339.
24. Hay Group. (2003). Using competencies to identify
high performers: An overview of the basics. Hay Group
Research Press. Retrieved fromhttp://www.haygroup.com/
downloads/uk/Competencies_and_high_performance.pdf.
25. Spencer & Spencer. (1993).
26. Spencer, McClelland, & Spencer. (1992); Spencer &
Spencer. (1993).
27. All competency descriptions provided here derive
from Competence at work, Spencer & Spencer. (1993). See p.
26 for Achievement Orientation Scale.
28. All competency descriptions provided here derive
from Competence at work, Spencer & Spencer. (1993). See p.
46 for Impact and Infuence Scale.
29. Goleman, D. (1998, November/December).
What makes a leader. Harvard Business Review.
Retrieved fromhttp://hbr.org/product/what-
makes-a-leader-harvard-business-review/an/
R0401H-PDF-ENG
30. Public Impact. (2007). School turnarounds: A review
notes • 21
of the cross-sector evidence on dramatic organizational
improvement. Lincoln, IL: Public Impact for The Center
for Comprehensive School Reform and Improvement.
Retrieved fromhttp://www.centerii.org/survey/downloads/
Turnarounds-Color.pdf
31. Hay Group. (2003).
32. Schmidt, R. & Hunter, J. (1998). The validity and
utility of selection methods in personnel psychology:
Practical and theoretical implications of 85 years of research
fndings. Psychological Bulletin, 124(2), 262–274; Spencer &
Spencer. (1993).
33. The Public Impact competency model for school
turnaround leaders that is used by the UVA School
Turnaround Specialist Program was developed using the
“stepladder approach” of building a model from related,
validated models. The competencies included in this
model stem primarily from validated competency studies
of highly successful leaders in analogous leadership roles
(entrepreneurs and managers in complex organizations).
These studies are reported in Spencer & Spencer, 1993. The
“stepladder approach” is described in detail in Appendix A.
34. Public Impact. (2008).
35. The New Teacher Project. (2009); Toch, T. &
Rotherham, R. (2008). Rush to judgment: Teacher evaluation
in public education. Washington, DC: Education Sector;
Reeves, D. (2009). Assessing educational leaders: Evaluating
performance for improved individual and organizational
results. Thousand Oaks, CA: Corwin Press.
36. Reeves, D. (2009); Stine, D. O. Developing an
evaluation system to improve principal performance and
accountability. Paper presented at the annual meeting of the
American Educational Research Association, Seattle, April
2001. 28 pages. ED 452 278.
37. Thomas, S. L. & Bretz, R. D., Jr. (1994, Spring).
Research and practice in performance appraisal: Evaluating
employee performance in America’s largest companies. SAM
Advanced Management Journal; The Hay Group (2010,
June). View point: Performing in uncertain times. Author:
Issue 3.
38. Spencer & Spencer. (1993); Norton, R. S. & Kaplan,
D. P. (1996). Using the balanced scorecard as a strategic
management system. Harvard Business Review; The Hay
Group (2010, June).
39. See Public Impact (2009). Try, try again: How to triple
the number of fxed failing schools without getting any better
at fxing schools. Retrieved fromhttp://publicimpact.com/
publications/Public_Impact_Try_Try_Again_Slide_
August_2009.pdf
40. Burke, M. J. & Day, R. R. (1986). A cumulative
study of the efectiveness of managerial training. Journal of
Applied Psychology, 71, 232–245.; Spencer & Spencer, 1993.
41. Individuals must feel motivated to improve. Hobby,
Crabtree, & Ibbetson. (2004).
42. Herman et al., (2008); Public Impact (2007).
43. McClelland, D. (1998).
44. All competence descriptions derived from Spencer &
Spencer, Competence at work. (1993).
45. Kowal, J., Hassel, E. A., & Hassel, B. C. (2009).
Successful school turnarounds: Seven steps for district
leaders. Washington, DC: Public Impact for The Center
for Comprehensive School Reform and Improvement.
Retrieved fromhttp://www.centerforcsri.org/fles/
CenterIssueBriefSept09.pdf
46. Duke, D. L., Tucker, P. D., Belcher, M., Crews, D.,
Harrison-Coleman, J., Higgins, J., et al. (2005). Lif-of:
Launching the school turnaround process in 10 Virginia
schools. Charlottesville, VA: Darden-Curry Partnership for
Leaders in Education. Retrieved fromhttp://www.darden.
virginia.edu/web/uploadedFiles/Darden/Darden_Curry_
PLE/UVA_School_Turnaround/LifOf.pdf; Public
Impact. (2007).
47. Calkins, A., Guenther, W., Belfore, G., & Lash,
D. (2007). The turnaround challenge: Why America’s best
opportunity to improve student achievement lies in our
worst-performing
schools. Boston: Mass Insight. Retrieved from http://
www.massinsight.org/publications/turnaround/51/fle/1/
pubs/2010/04/15/TheTurnaroundChallenge_MainReport.
pdf; Herman et al. (2008); Public Impact. (2007).
48. Herman et al. (2008); Public Impact. (2007).
49. Public Impact. (2007); Herman, et al. (2008).
50. Spencer, McClelland, & Spencer. (1992).
51. Descriptions of model development are derived from
several sources: Hay Group, 2003; Spencer, McClelland, &
Spencer, 1992; Spencer & Spencer, 1993; McClelland, 1998.
52. Spencer & Spencer. (1993); Hay Group. (2003).
53. Spencer & Spencer. (1993).
54. Spencer, McClelland, & Spencer. (1992).
55. Competency defnitions used here are derived from
Spencer & Spencer, Competence at work. (1993).
56. Spencer & Spencer. (1993).
57. Spencer & Spencer. (1993).
58. Hay Group. (2003).
59. Spencer, McClelland, & Spencer. (1992).
60. For example, the turnaround teacher and leader
models created by Public Impact were developed frst by
examining the research base on successful turnaround leader
actions and then mapping these to validated competency
models of similar roles (for leaders: managers in complex
organizations and entrepreneurs; for teachers: models
of teachers from other nations and studies of teachers
successful with high-poverty populations). Future validation
and editing — or from-scratch modeling — would be ideal,
when a larger pool of school turnaround leaders and
turnaround teachers with measured results are available for
study. Seehttp://www.publicimpact.com/human-capital/
competencies-of-high-performers for the most recent models
and selection tools.
22 • notes
61. Hay Group. (2003).
62. Schmidt, R. & Hunter, J. (1998). The validity and
utility of selection methods in personnel psychology:
Practical and theoretical implications of 85 years of research
fndings. Psychological Bulletin, 124(2), 262–274; Spencer &
Spencer. (1993).
63. For step-by-step guidance on conducting a BEI
interview for a school turnaround leader, see Public Impact’s
School turnaround leaders: Selection toolkit, available athttp://schoolturnarounds.org/.
64. Spencer, L. M., McClelland, D. C., & Spencer, S. M.
(1992). Competency assessment methods: History and state of
the art. Hay/McBer Research Press, p. 4
65. Spencer & Spencer. (1993).
66. Burke & Day. (1986); Spencer & Spencer. (1993).
67. Spencer & Spencer. (1993). p. 290.
68. Spencer & Spencer. (1993).
69. All competency descriptions provided here derive
from Spencer & Spencer, Competence at work (1993).
70. Spencer & Spencer. (1993).
doc_805435937.pdf
This paper explain using competencies to improve school turnaround principal success.
Using Competencies
to Improve School
Turnaround Principal
Success
by Lucy Steiner and
Emily Ayscue Hassel
About the Authors
LUCY STEINER is a senior consultant with Public
Impact. She researches and consults on a variety of critical
education issues, including teacher and leader policy,
school restructuring, charter school policy, and teacher
professional development. Ms. Steiner both conducts her
own work and leads project teams to deliver research,
training, and consulting. Her work ofen provides a bridge
between district leadership, school leadership, and instruc-
tion. A former high school English teacher, Ms. Steiner
holds a master’s degree in education and social policy from
Northwestern University and a B.A. with highest honors
from the University of North Carolina at Chapel Hill.
EMILY AYSCUE HASSEL is Co-Director of Public Im-
pact. She provides thought leadership and oversight to
Public Impact’s work on teacher and leader policy, or-
ganizational transformation, parental choice of schools,
and emerging opportunities for dramatic change in pre-K
to grade 12 education. Her work has appeared in Educa-
tion Week, Education Next, and other publications. She
previously worked for the Hay Group, a leading human
resources consulting frm. Ms. Hassel received her law and
master in business administration degrees from the Uni-
versity of North Carolina at Chapel Hill.
Acknowledgements
This report was made possible by the University of Vir-
ginia’s Darden/Curry Partnership for Leaders in Educa-
tion and its School Turnaround Specialist Program. The
authors are grateful to Julie Kowal and Bryan Hassel of
Public Impact and to LeAnn Buntrock and William Rob-
inson of the UVA School Turnaround Specialist Program
for their feedback on early drafs. We would also like to
thank Sharon Kebschull Barrett for careful editing, and
April Leidig-Higgins for the design of the report.
Using Competencies to Improve School Turnaround
Principal Success was made possible by the support of
© 2011 Public Impact, Chapel Hill, NC
© 2011 University of Virginia’s Darden/Curry Partnership
for Leaders in Education, Charlottesville, VA
Public Impact is a national education policy and manage-
ment consulting frm based in Chapel Hill, NC. We are a
team of researchers, thought leaders, tool-builders, and on-
the-ground consultants who help education leaders and
policymakers improve student learning in K–12 education.
For more on Public Impact and our research, please visit:
www.publicimpact.com.
Public Impact encourages the free use, reproduction, and
distribution of this working paper for noncommercial use.
We require attribution for all use. For more information
and instructions on the commercial use of our materials,
please contact us at www.publicimpact.com.
Please cite this report as:
Steiner, L., & Hassel, E. A. (Public Impact). (2011). Using
competencies to improve school turnaround principal success.
Charlottesville: University of Virginia’s Darden/Curry
Partnership for Leaders in Education. Retrieved from
www.DardenCurry.org
• 1
Introduction
I
n a time when student outcomes matter more
than ever, many states, districts, and reformers
are considering whether and how turnarounds of
chronically failing schools can achieve superior
results. In fact, we already know much about when and
how successful turnarounds work, both from other sec-
tors and from recent experience in education.
Two major factors afect turnaround success: the
characteristics and actions of the turnaround leader,
and the support for dramatic change that the leader
and staf receive from the district, state, and/or other
governing authority. Although leadership accounts for
25 percent of school efects in most schools,1 in a turn-
around the leader is paramount. It is almost unheard
of for turnarounds to occur without a special breed of
leader at the helm — one who engages and focuses the
whole community on achieving dramatic improvement
goals fast.2
This paper aims frst to shed light on one element
of leadership: the characteristics — or “competencies”
— of turnaround leaders who succeed in driving rapid,
dramatic change. Second, we recount the elements of
support that districts must provide these leaders to
enable and sustain a portfolio of successful school turn-
arounds. Fortunately, decades of experience in other
sectors and in education systems of other nations reveal
tools and techniques for understanding and using turn-
around leader competencies, and for governing turn-
around leaders successfully.
U.S. educators must act on this knowledge. Today,
few districts have an explicit strategy to select and em-
power school turnaround leaders using the best avail-
able techniques. Few provide the autonomy, support,
and accountability for rapid, dramatic change that
will attract, keep, and enable turnarounds by capable
leaders.
Here we explain what states, districts, and others
with an interest in school turnarounds need to know.
This paper:
• Describes how using competencies that predict
performance can improve turnaround principal
selection, evaluation, and development; and
• Summarizes prior research about how districts
can create the right environment to increase
school turnaround leader success.
In addition, the appendix provides details about
options for building valid competency models.
2 • usi ng competenci es to i mprove school tur naround pr i nci pal success
Turnaround Leader
Competencies
H
ere we describe the role of leader-
ship in organizational turnarounds
and how using competencies can allow
better selection, development, and
support of those leaders.
Turnaround Leaders:
A Special Breed of Leadership
Decades of research have documented that having
the right leader is an essential component of success-
ful turnarounds.3 Despite the intense national focus
on school turnarounds, signifcant barriers prevent
reformers from fnding and enabling leaders likely to
succeed in a turnaround.
Why? First, turnaround eforts are made when
organizations are in a state of entrenched failure. Lead-
ers who would otherwise succeed ofen fall short in a
turnaround. Turnaround success is a challenge even in
sectors where this strategy has been tried ofen. Stud-
ies across sectors suggest that only 30 percent of turn-
around eforts succeed.4 Even leaders who have excelled
in other circumstances may fail when faced with the
rapid, dramatic change required in a turnaround efort.
The current education leadership pool is unlikely to
have the number and type of candidates needed to lead
a large number of school turnaround eforts. Therefore,
recruitment must focus on candidates whose capabili-
ties ft the specifc demands of turnarounds.
Second, typical school district practices are not de-
signed to recruit and select talent for challenging schools,
including the bold leaders needed for turnaround
schools.5 Most districts base principal and staf hiring
on college degrees and years of experience. Numer-
ous research studies over several decades have shown
that degrees and experience (afer the frst few years of
teaching) are poor predictors of performance.6 As re-
formers focus on recruiting, they need new methods to
choose the right people for turnarounds.
Third, few districts measure performance diferences
among leaders and staf that would be useful for identify-
ing and developing internal candidates for school turn-
around leadership.7 Understanding the characteristics
needed to succeed in a turnaround would allow inter-
nal selection and development of high-potential candi-
dates from among current teachers and principals.
Understanding the Crucial Role
of Competencies
Performance diferences are large in difcult jobs:
research has found that the top 1 percent of jobhold-
ers in complex jobs produce results 127 percent better
than the average.8 Many organizations select employees
based on experience and degrees, hoping that these
indicators will predict success on the job. Yet many of
us have experienced the “experience and degrees” myth
frsthand. When two seemingly similar candidates are
hired — with the same level of education, experience,
and technical skills — one sometimes turns out to be an
outstanding performer, while the other struggles.
In the 1970s, a cognitive psychologist from Har-
vard University, David McClelland, tried to fnd out
why. He hypothesized and ultimately demonstrated
that habits of behavior and underlying motivations,
which he called “competencies,” diferentiate workers’
performance outcomes. As a result of his and others’
subsequent research, employers can understand not just
what employees do to be successful, but how they do it.9
By examining candidates’ competencies, employers can
usi ng competenci es to i mprove school tur naround pr i nci pal success • 3
uncover diferences likely to afect performance, help-
ing to choose between candidates who may otherwise
seem identical.
Competency-based performance management re-
mains relatively rare in education. But many organiza-
tions in other sectors (public, nonproft, and private)
use competencies for selection, development, and career
planning — even pay.10 Other nations, including Singa-
pore and the United Kingdom, now use competencies
throughout their education systems.11 To learn how
competency-based human capital systems work well,
we therefore need to turn to other sectors and nations
that have used them for many years.
For the past 10 years, competency-based evalua-
tion has been the bedrock of Singapore’s educational
system, one of the strongest in the world.12 In the early
2000s, Singapore implemented a competency-based
performance management system for the three major
roles in Singaporean schools — teachers, principals, and
school specialists.13 School ofcials use the competency
model in conjunction with the achievement of positive
student outcomes and other measurable results to set
annual competency targets, evaluate competency levels
throughout the year, match each educator to a career
path, and determine annual bonuses. In the decade
since introducing this system, Singapore has continued
to raise student performance and narrow the achieve-
ment gaps among its ethnic groups.14
By examining candidates’
competencies — habits of
behavior and underlying
motivations — employers can
understand how employees
succeed.
4 • usi ng competenci es to i mprove school tur naround pr i nci pal success
Research Base on Competencies
When Dr. McClelland began his research in the early
1970s, he was reacting to studies fnding that standard
ways of evaluating job candidates — IQ tests and other
tests of academic aptitude, knowledge content tests,
school grades, and academic credentials — did not fully
predict job performance and were ofen biased against
minorities, women, and people in poverty.15 In the
course of his research, he coined the term “competency”
to describe the behavioral characteristics that he found
could predict performance.
Although the term competency ofen describes any
work-related skill, in this context competencies refers
to the underlying motives and habits — patterns of
thinking, feeling, acting, and speaking — that cause a
person to be successful in a specifc job or role.16 Mc-
Clelland compared the competency patterns of very
high performers to those of typical performers. When
analyzing these high performers, he found that un-
derlying characteristics (e.g., persistence, achievement
motivation, self-confdence) led to actions (e.g., calcu-
lated risk taking, goal setting, planning) that in turn
led to better outcomes (e.g., efectiveness, productivity,
innovation).17 Research conducted in the decades since
McClelland’s original study has provided further evi-
dence that underlying competencies enable successful
performance in a given job or role.18
In the course of their research, McClelland and his
colleagues also developed a methodology for identify-
ing and validating the competencies for particular jobs
and roles. This interview technique — known as a be-
havior event interview (BEI) — combined elements of an
existing technique called the “critical incident method”
with probes about motivation that McClelland’s team
refned over several decades.19 Instead of asking people
to provide hypothetical responses to interview ques-
tions, BEIs ask them to walk interviewers through past
incidents step by step, as though the interviewee is
reliving the experience. This helps interviewees reveal
what they were thinking, saying, and doing at the time,
and makes it hard to claim credit for actions that the
individual did not take.20
Studies since McClelland’s original research indi-
cate that structured interviews such as the BEI that
probe for information about past events are highly
correlated with later job performance.21 For example,
an independent, comparative study of behavior-based
interviews and unstructured interviews at a large life
insurance company found that the behavior-based
interviews yielded a validity coefcient of .48 using
supervisor ratings as the performance criterion, and
.61 using sales dollars as the performance criterion. In
contrast, the standard interview yielded a validity co-
efcient of .08 and .05, respectively.22 McClelland’s last
published study found that 65 percent to 86 percent
of managerial candidates who met a threshold level of
competence when selected using a model constructed
with BEIs ended up in the top third of performers,
compared with 11 perent to 20 percent of candidates
who scored lower in competence during the selection
process.23
Behavior event interviews
probe for information about
past events to predict future job
performance.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 5
Imagine the consequence for children of selecting
turnaround principals as accurately: candidates meet-
ing threshold levels of competencies could be far more
likely to succeed in turnaround attempts than candi-
dates who fall short.
The “iceberg model” (Figure 1) was developed by
the Hay Group, an international human resource frm
where McClelland spent the latter part of his career,
to demonstrate how competencies relate to observable
qualifcations, knowledge, and skills. As this model
demonstrates, competencies are more difcult to de-
tect than qualifcations, skills, and knowledge, but
they largely infuence these observable behaviors.24
Competency research further suggests that outstand-
ing performance in complex jobs — ones in which
most candidates have a similar educational history
and signifcant autonomy over daily work tasks — is
driven more by underlying competencies than by read-
ily observed skills and knowledge. Individuals in these
complex jobs, such as school principals, use their similar
content knowledge very diferently to accomplish work
goals. For example, some proactively set difcult goals
and stick to them, while others do only what is asked
by superiors or give up when a goal proves difcult to
achieve. Some try to do all the work themselves, while
others identify colleagues’ strengths and put them into
roles where they will succeed. This wide variation in
how people work produces greatly varying results un-
explained by prior knowledge, degrees, and experience,
making competency-based performance management
practices especially critical.25
In the past 40 years, dozens of competencies have
been identifed as success distinguishers in diferent
jobs and roles, in combinations that are unique to each
particular role. Nevertheless, two competencies appear
critical to high levels of success in most complex leader-
ship jobs: “achievement” and “impact and infuence.”26
• Achievement is defned as “the drive and actions to
set challenging goals and reach a high standard of
performance.” In a leader, achievement includes “set-
ting high performance goals for the organization,
prioritizing activities to achieve the highest beneft
relative to inputs, and working to meet goals using
direct action, staf, and other available resources.”27
• Impact and infuence is “acting with the purpose
of afecting the perceptions, thinking and actions
of others. It includes empathizing with others and
Figure 1: Observable characteristics versus underlying competencies (Adapted from “The Iceberg Model” in Spencer &
Spencer, Competence at Work, p. 11.)
qualifications
Certifcation
Degrees {
{
Knowledge and skills
Teacher practices
competencies
Recurring patterns of thought and action
Example: Achievement Drive {
6 • usi ng competenci es to i mprove school tur naround pr i nci pal success
anticipating likely responses to situations, tailoring
actions and words to create an intended impact, and
giving and withholding information to obtain spe-
cifc responses.”28
Leaders in diferent situations — such as start-ups,
large organizations, and turnaround eforts — might
need to use these competencies in difering degrees
and ways.29 But a literature review of the actions that
successful turnaround leaders take indicates that high
levels of competence in both achievement and impact
and infuence are most likely essential for school turn-
around principals.30
Identifying Distinguishing Competencies:
Building a Valid Model
What makes competency-based performance manage-
ment strategies powerful is the potential to correlate
qualitative characteristics with performance outcomes
in a statistically valid manner. A good competency
model includes descriptions of both the competencies
that are needed to succeed in a job and the increasing
levels of performance within each competency. Most
important, in a valid model, the competencies and
increasing levels correlate with performance outcomes,
such as student learning gains.
Organizations should use competency models built
using an approach that is as valid and predictive of per-
formance as circumstances allow.31
• Building a competency model fom scratch. A model
can be built from scratch when there is an accessible,
large set of jobholders who have been on the job for
several years and who can be classifed as outstand-
ing and average.
• Building a model fom related, validated models —
the stepladder approach. Emerging roles, jobs in
emerging sectors, and jobs spread out among many
smaller organizations may not have enough acces-
sible performers for a data set of outstanding and
typical performers whom researchers can compare.
In these cases, a model can be extrapolated by map-
ping the actions needed for success in the job to
similar jobs for which validated competency models
are available.
In either case, a model can be validated and refned
over time by comparing competency ratings during se-
lection to later performance outcomes. Subsequent hir-
ing can emphasize competencies that most accurately
sort high and typical performers. Appendix A provides
more detail about the various options for building and
validating competency models.
Using Competencies to Ensure Efective
School Turnaround Leadership
Organizations can use competencies for many pur-
poses. Here we briefy describe three important uses
for school turnarounds: 1) hiring efective turnaround
principals; 2) evaluating principal performance; and 3)
providing targeted development for school turnaround
principals. Using performance-predictive competencies
at each of these critical stages of an employee’s career
increases the likelihood of improving employee perfor-
mance in key results areas.
Hiring efective school turnaround principals
Selecting people who already have most of the compe-
tencies needed for turnaround leadership, rather than
relying on long-term development, may be the best way
to achieve the rapid results a turnaround demands.32
A competency model that will be used for selec-
tion should include competency descriptions, levels,
and tools for rating and comparing candidates. For
example, some states and the UVA School Turnaround
Specialist Program use Public Impact’s competency
model for selecting school turnaround principals.33 It
includes: 1) short, broad defnitions of the competen-
cies that distinguish high performance; 2) rating scales
of increasingly efective levels of behavior within each
competency; 3) competency level targets for the job of
school turnaround principal; and 4) selection steps and
guidelines for assessing candidates’ competency levels
using the behavior event interview (BEI).34 Appendix B
provides more detail about how hirers can use the BEI
to assess candidate’s competencies.
Evaluating school turnaround principals
Considerable evidence indicates that current principal
evaluation practices are similar in quality to the inefec-
usi ng competenci es to i mprove school tur naround pr i nci pal success • 7
tual teacher evaluation practices that have been widely
criticized.35 According to researchers, principal evalua-
tion too ofen consists of a binary rating sheet that asks
the evaluator to check of “satisfactory” or “unsatisfac-
tory” on a number of items such as “time management”
or “demonstrates efective organizational skills.”36 This
kind of evaluation fails to deliver the information that
districts need to evaluate principals’ performance accu-
rately and that principals need to improve.
Including measurable results in principal evaluation
could correct this situation. The arguments for and
against this shif are beyond the scope of this report,
but basing at least part of any evaluation on student
outcomes would be consistent with the best practices of
many high-performing organizations in both the pub-
lic and private sectors.37
Evaluations rarely hinge solely on results, though.
Outcomes alone do not give employers information to
help with promotion and job placement decisions, nor
do they give employees information about how to im-
prove.38 A complete evaluation system includes not only
measurable results, but also professional skills, such as
curriculum planning, and of course the competencies
that are critical for achieving results. Understanding
the competencies a principal demonstrates during a
school turnaround efort and rating the principal’s per-
formance against these competencies in an evaluation
can help employers understand why a leader is succeed-
ing or falling short — and whether ultimate success is
likely. When the competency gaps are large and early
indicators of progress are poor, a leader may need to be
replaced rather than developed.39
Developing school turnaround principals
Waiting to develop a leader on the job means a sure
recipe for failure when fast results are essential. But
even very competent turnaround principals will have
areas of competency weakness, and they also will need
to develop turnaround competencies among the staf
leaders on their teams. Fortunately, some research in-
Principal evaluation too ofen
consists of a binary rating
sheet that asks the evaluator
to check of “satisfactory” or
“unsatisfactory.”
When the competency gaps are
large and early indicators of
progress are poor, a leader may
need to be replaced rather than
developed.
8 • usi ng competenci es to i mprove school tur naround pr i nci pal success
dicates that leaders can continue improving individual
competencies with the right training, when it is clearly
linked to outcome goals.40
Indeed, early turnaround results can provide the es-
sential motivation principals and staf leaders need to
improve.41 In a school turnaround, leaders must rapidly
identify failing approaches and try new tactics.42 Those
who understand their own challenges are in a better
position to make these changes fast.
The superintendent (or other direct supervisor)
and the principal can identify the principal’s com-
petency levels — for example, with ratings based on a
combination of supervisor, staf, parent, and student
input — and compare these to levels needed for supe-
rior performance. The principal can then focus on
closing his gaps and using his strengths more ofen. A
valid model with progressively more efective levels of
competence aids development by giving school lead-
ers specifc next-step actions needed for better student
outcomes.43
For example, a principal might have a current rating
of “4” on the impact and infuence competency, which
indicates that she regularly “thinks ahead about the
likely reaction of the audience and takes two or more
steps that are calculated to obtain desired impact.” In
order to improve, she and her supervisor might set a
goal for her to move to level 5 by more efectively and
consistently “using others (e.g., parents, staf members)
to obtain desired impact.”44 Appendix C provides more
details about how principals can continue improving
their individual competencies with the right training.
But even the best leader’s eforts can be thwarted
by an environment unsupportive of change. So next
we summarize the elements of a district environment
in which turnarounds are most likely to occur — and
succeed.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 9
Creating an External
Environment that
Supports Turnarounds
T
he research on turnarounds in educa-
tion and other sectors suggests that multiple
environmental factors infuence an orga-
nization’s ability to improve rapidly and
that even the best leader’s eforts can be frustrated and
diminished by an unsupportive environment. Districts
that want to increase the odds of successful school
turnarounds should take an active leadership and sup-
port role. The following steps for districts are drawn
from a research brief written by Public Impact that ap-
plies cross-sector and education research to the district
role in turnarounds:45
• Commit to success. Policymakers overseeing the
turnaround efort — state department ofcials,
district leaders, school board members — must pri-
oritize student learning needs over the customs, rou-
tines, and established relationships that can stand
in the way of necessary change. They must view
turnarounds not as a one-time solution but as part
of a sustained efort to eliminate chronic low perfor-
mance, and must be willing to stay the course even
when some frst attempts fail. Policymakers need
to assess their own capacity to oversee and support
dramatic and sometimes disruptive change before
committing to this strategy.
• Choose the right schools. Turnarounds are a neces-
sary step in schools where student performance is
extremely and chronically low and where incremen-
tal eforts to improve student outcomes have failed.
• Give leaders the “big yes.” Successful turnaround
leaders ofen achieve results by working around
rules, asking for forgiveness afer their strategy has
worked rather than seeking permission before-
hand.46 By giving turnaround leaders the “big yes”
on critical autonomies — stafng decisions, schedul-
ing, budgeting, and other operational issues — poli-
cymakers can help support successful turnarounds.
Districts that want to increase
the odds of successful school turn-
arounds should take an active
leadership and support role.
Policymakers overseeing the turn-
around efort must prioritize
student learning needs over the
routines that can stand in the way
of necessary change.
10 • usi ng competenci es to i mprove school tur naround pr i nci pal success
•Proactively engage the community. Dramatic
change requires active communication with local
stakeholders. Successful eforts to engage the com-
munity are characterized by public acknowledge-
ment of past failures coupled with a forceful, posi-
tive vision for the future.47 Publicizing early “wins”
can also send a powerful message that change is pos-
sible and turnarounds can work.48
• Hold leaders accountable for results. Policymakers
must hold turnaround leaders to high standards and
a short timeline for results. The research literature
does not indicate an exact timeline required to turn
an organization around, but in successful turn-
around eforts, fast, focused changes occur in the
frst few months, and substantial improvements in
the frst year.49
• Develop a talent pipeline. District leaders need
to build their supply of turnaround leaders and
teachers through proactive recruitment, careful
selection, targeted training, and strategic placement
in turnaround schools. The skills and abilities of
principals and teachers who succeed in turnarounds
difer from those of their peers who succeed in less-
challenging schools. Competency screening should
be a critical step in the hiring process.
Publicizing early “wins” can
also send a powerful message
that change is possible and turn-
arounds can work.
usi ng competenci es to i mprove school tur naround pr i nci pal success • 11
A Call to Action
T
he U.S. desperately needs a strong
cadre of school leaders who can turn around
persistently low-performing schools. But
today, this cadre is far too small. States
and districts that are serious about eliminating broad-
scale failure in schools must use the very best tools
available to select, evaluate, and develop these school
turnaround leaders. Current practices — inconsistent
hiring, uneven support, and weak evaluation — are
severely inadequate. Competency-based people-man-
agement, coupled with the right district environment,
can signifcantly increase the number and performance
of school turnaround leaders. What steps would make
this urgent priority a reality?
Disseminate information about competency-
based practices to key stakeholders in education.
Too few education leaders and policymakers know
about competency-based selection, evaluation, and
development — strategies commonly used in other sec-
tors. Even fewer know that districts can validate and
improve competency models by comparing competency
ratings with performance outcomes. Similarly few
realize how diferent the competencies needed for the
role of turnaround leader are from those needed in
traditional principal jobs. Audiences that would beneft
from better awareness include district leaders, state
policymakers, philanthropists, parent and community
advocacy organizations, and national and local school
leadership training programs.
Invest in competency models for critical school
leadership (and teaching) roles. The best competency
models are based on data from behavioral event inter-
views and correlate with performance outcomes. States
and districts in the U.S. have not invested in this type
of rigorous model building, because of the investment
involved and low awareness of the value. If the United
States is going to dramatically improve teacher and
leader performance, then some combination of dis-
tricts, states, the federal government, and private foun-
dations must invest in the research, development, and
improvement of competency models for critical roles
in education. As the pool of serious school turnaround
attempts grows, validation and improvement of com-
petency models for leaders and teachers in this context
will be possible.
Select school turnaround leaders for competence.
Our nation must identify far more leaders to turn
around persistently failing schools. Competency-based
selection for critical leadership (and other important)
positions would enable selection from a much wider
labor pool — turnaround leaders from other sectors
and emerging teacher-leaders, for example. Other sec-
tors and nations have used this approach widely, yet it
remains rare in U.S. education.
Develop strong competency-based training pro-
grams for school turnaround principals. A critical ap-
plication of competency models is development. Educa-
tion leaders should add and expand training programs
for school turnaround principals that incorporate
competency-based practices, such as the University of
Virginia School Turnaround Specialist Program.
Evaluate and publicize results. As schools, districts,
and states undertake more school turnarounds, they
must collect data to compare successful and less-suc-
cessful leaders in these unique settings, and publicize
this information widely. These analyses can validate in-
dividual competencies — and perhaps identify new ones
as more data emerge — as well as provide rich examples
of competencies in action.
• 13
Appendix A
Identifying Distinguishing Competencies
Building a Valid Model
Two aspects of building a competency model are criti-
cal to validity, regardless of approach. The frst is to use
a sample of employees that compares truly outstanding
performers with people who are average performers in
their jobs. The second is to collect data about compe-
tencies that give more weight to behaviors that jobhold-
ers have actually displayed on the job than to what they
or others think they should do.50
In the following section, we describe two approaches
to building a valid competency model, noting areas
where model-builders have a choice of methods.
Building a competency model from scratch
This section describes the most customized way to
build a valid competency model.51 Using this method,
however, requires a large data set of jobholders who
have been on the job for several years and who can
clearly be classifed as outstanding or average. While
this might be possible someday for turnaround leaders,
the techniques described in the next section for
building — and then validating — models with limited
data sets are more appropriate today.
We are not aware of any public education organiza-
tion in the United States that has built a competency
model from scratch using the rigorous method de-
scribed here — for any job, including jobs that, unlike
turnaround leader, are already very prevalent. Because
several of the steps described below are costly and
require expertise, building a model that meets this
high standard of rigor and efectiveness represents a
considerable investment of resources. In positions with
numerous jobholders, such as traditional principal and
teaching roles, the cost per job of this method would
be small. Leaders of states and large districts, take note:
one large investment could provide valid, performance-
predictive tools to nearly every school in a district or
state — or nationwide.
From-scratch: Building a model with behavior
event interviews:52
• Determine performance criteria. First, determine
what constitutes outstanding performance. Do this
in consultation with a range of people who under-
stand the role, and, ideally, include data both quan-
titative (e.g., the magnitude and speed of student
learning gains) and qualitative (e.g., parent and staf
ratings).
• Select criterion sample. Researchers should then
select two groups of current jobholders, one that
has displayed average performance according to the
performance criteria, and another that has displayed
truly outstanding performance on the same set of
measures. The samples need to be large enough to
allow for statistical analysis.
• Collect data. The most efective method for col-
lecting data is the structured interview technique
mentioned earlier — the behavior event interview
(BEI).53 Unlike other interview techniques that ask
candidates to respond to hypothetical situations,
the BEI elicits detailed stories of past events that
reveal how top performers difer from more typical
or lower-performing jobholders. During this stage
of the model-building process, avoid bias by ensur-
ing that neither the interviewee nor the interviewer
knows if they are in the “outstanding” or “average”
sample. Because the BEI is also a highly efective
method to use for actual candidate selection afer
the model has been validated, it is described in detail
in the section below on selection.
• Develop model. Analyzing interview transcripts
to determine the diferences between average and
outstanding performers is the most complex stage
of the process, one that requires qualitative re-
view and coding as well as statistical analysis. The
goal is to determine what outstanding performers
14 • appendi x a
do (e.g., actions, thoughts, feelings) that average
performers do not do, and vice versa. Research-
ers then convert these fndings into interval scales
that identify a “threshold” level that describes the
minimum requirements for average performance in
each competency, as well as descriptions of levels of
increasingly efective behaviors associated with this
competency.54
For example, the competency called “Initiative
and Persistence” may be a critical competency for
turnaround leaders. It involves having the “drive and
actions to do more than is expected or required in
order to accomplish a challenging task.” As the scale
increases, so does the complexity of actions associ-
ated with this competency, from “voluntarily initi-
ates and follows through on new work project that
is not assigned by others” to “acts without formal or
explicit authority, takes personal or career risks and
bends organization norms or rules to accomplish a
work objective.”55
Afer the descriptors are written, they are then
tested for inter-rater reliability and refned as
needed. The fnal competency list, competency de-
scriptions, descriptions of diferent levels of perfor-
mance within a competency, and information about
coding are included in a document that becomes the
“competency model” for a particular job.
• Validate model. While validation is ideal, many
organizations building models from scratch rely
on the rigor of the initial process to produce a
valid model. When time and funds allow, several
methods can validate a competency model. Choos-
ing a second sample of top and typical performers,
conducting BEIs, and analyzing correlation of their
competencies with actual performance outcomes
is one method. Another rigorous method is to as-
sess incoming candidates using data from BEIs,
then analyze whether those who scored higher in
the selection process perform better in their jobs
according to the performance criteria.56 Because
the competency model is designed to predict actual
performance on the job, testing the model against
performance results is the most powerful way to
validate the model. The other advantage of this
method is it allows the model designers to revise and
refne based on actual performance.57
Stepladder approach: Building a model
from related, validated models
Two barriers can prevent building models from scratch:
cost and limited past data about high performers. First,
in many cases like school turnaround leaders, limited
data are available to build a competency model from
scratch. Emerging roles, jobs in emerging sectors, and
jobs spread out among many smaller organizations may
not have enough accessible performers for a data set of
outstanding and typical performers whom researchers
can compare. Second, models are expensive to build
from scratch, particularly when jobholders are spread
out geographically. What follows is a description of
how to achieve a valid competency model at a lower
cost and with limited data and access to jobholders.
• Determine performance criteria. This step is criti-
cal and should be implemented as described above.
• Select criterion sample. In selecting the two groups
for analysis, model builders have some leeway with
regard to sample sizes. The question to ask is, what is
the minimum sample size needed to produce a valid
result? Expert opinion suggests that it is better to in-
clude a larger sample of star performers, if possible,
because they are the best source for detailed infor-
mation about outstanding performers.58 Sometimes,
for example in an entirely new role, model builders
will need to envision expected actions of performers
likely to achieve outstanding results. In this case,
focus groups may still be useful, but extrapolation
from existing, valid models of related jobs will be
essential.
• Collect data. In addition to BEIs (referenced above),
there are other data collection techniques:
— Focus groups. This involves asking groups made
up of people who know the job well to identify
the competencies that are critical in a given job
or role. Experience indicates that about half the
competencies identifed by focus groups are vali-
dated by a full competency study using BEIs.59
appendi x a • 15
—360-degree surveys. Surveys ask superiors, peers,
subordinates, and external people who interact
with jobholders to rate whether particular com-
petencies are important for superior performance,
how ofen they are needed, and whether failure is
likely if someone does not have this competency.
Surveys are useful because they are quick and
cost-efective to administer, and they can provide
enough data for valid statistical analysis. How-
ever, survey designers may miss critical competen-
cies in creating the survey.
— Extrapolate from existing models of similar
jobs. Sometimes, top and typical performers
are not accessible or available in numbers large
enough to use the from-scratch method. This is
especially likely in new roles and in organizations
undergoing signifcant change. In these situations,
model builders can look at validated competencies
that distinguish performers in other roles and use
these to identify actions needed for job success.
School turnaround leaders are one example: until
an identifable, accessible population of perform-
ers who can clearly be labeled superior or typical is
available for study, extrapolated models will have
to sufce.60 This approach is a relatively quick
way to build a model, but still requires deep un-
derstanding of the job, the related jobs for which
validated competency models have already been
built, and the available competency models or
dictionaries. However, without access to detailed
information about superior and average perform-
ers in this exact role, the model may be less valid
initially than a from-scratch model would be, if it
were feasible. Validation is especially important
for models built from related job models.
• Develop model. Experts analyze the data from
focus groups, surveys, and existing competency
models or dictionaries to choose the competencies
likely to distinguish top performers and to identify
the “threshold” levels of behavior for solid and out-
standing performance. This stage requires expertise:
thorough understanding of the jobs, thoughtful
analysis of the data, and, when surveys are used,
the ability to conduct statistical analysis. However,
model builders do not identify new competencies
as they would when analyzing BEIs. The fnal out-
come — a competency model — is similar but less
customized than a from-scratch model.
• Validate model. Models built this way can be re-
fned over time, with use and deliberate validation
analyses. There are several options for validating the
competency model beyond comparing BEI results
with actual job performance. For example, design-
ers can conduct BEIs on incoming jobholders. Once
enough have been interviewed for statistical valid-
ity and enough time has passed to observe their
performance, experts can determine the extent to
which each competency in the model accurately
distinguishes outstanding and average performers.
Conducting BEIs for validation requires the same
expertise as BEIs for model building and thus is rela-
tively costly.
Another less-expensive approach when large
numbers of jobholders are available is to design
questionnaires that ask them — both outstanding
and average performers — to respond to questions
based on the model (e.g., how ofen in the past
two months have you taken on a voluntary task at
work?).61 If the model was well-designed, then out-
standing and average performers may be accurately
identifed by their responses and the competency
model validated. 360-degree assessments by peers,
subordinates, and supervisors can be used similarly
to validate.
16 •
Appendix B
Hiring Efective School Turnaround Principals
Hiring people who already have most of the competen-
cies needed for turnaround leadership, rather than rely-
ing on long-term development, may be the best way to
achieve the rapid results a turnaround demands.62
Afer prescreening for other requirements, hirers
can assess candidate competencies using these steps:63
• Step 1: Conduct behavior event interview. In a BEI,
the interviewer’s goal is to understand in detail how
candidates perform various aspects of their work. To
do this, the interviewer asks candidates to recall past
events when they have felt successful or have dealt
with specifc situations at work (e.g., a time when he
or she infuenced another person, or led a team of
people to accomplish work that was satisfying). The
candidate should spend 15 minutes or more describ-
ing the incident in great detail, with the interviewer
probing insistently for the information needed to
understand exactly what a person was thinking
or doing at the time. According to Hay Group re-
searchers, sample probes include: “What led up to
the situation? Who was involved? What did you
think about, feel, want to have happen in the situa-
tion? What did you do? What was the outcome?”64
The interviewer should: 1) probe insistently for
detail; 2) keep the candidate focused on past events
rather than refecting on hypothetical situations or
using generalities; and 3) take comprehensive notes
or record what candidates say so that their responses
can be used later for scoring.
• Step 2: Rate candidate’s competency levels. The
interview team then closely reviews the candidate’s
responses and notes any examples of “codable data,”
or data that is valid for scoring against the compe-
tency model. To be codable, responses must be in
the frst person (“I did this” rather than “we did
it”), be about real rather than hypothetical actions
and feelings, be volunteered by the candidate inde-
pendently, and be about past rather than present
feelings or behavior. Afer noting the codable com-
ments, the interview team compares them against
the competency level descriptions and rates the can-
didate on each competency.
• Step 3: Make hiring decisions. Afer each can-
didate has been rated on the competencies, hirers
compare the strengths (and weaknesses) of the can-
didates who meet all or most competency thresholds
to determine whom to hire. The number of slots
available compared with the number of qualifed
candidates — and the level of challenge in each
school — might afect how many are hired and for
which schools.
• Step 4: Collect performance data and revise in-
terview. Afer an initial round of hires, hirers can
compare competency scores at selection with actual
performance. Future hiring can focus on the compe-
tencies that best predict performance.65
In addition, most selected candidates will fnd
feedback about their competency scores helpful both
for using strengths with confdence and for choos-
ing staf and assigning roles that balance each leader’s
weaknesses.
• 17
Appendix C
Developing School Turnaround Principals
Even very competent turnaround principals will have
areas of competency weakness, and they also will need
to develop turnaround competencies among the staf
leaders on their teams. Fortunately, some research in-
dicates that leaders can continue improving individual
competencies with the right training, when it is clearly
linked to outcomes goals.66
Options to help school turnaround principals de-
velop specifc competencies include:
• Competency training. Competency training ap-
pears to be most efective when trainers: 1) present
compelling evidence that competency improvement
will make employees better at their jobs; 2) give
feedback to employees on how their own levels of
competence compare with outstanding performers;
3) give employees opportunities to practice compe-
tency behaviors; and 4) expect employees to set com-
petency development goals with action plans.67
• Self-development resource guides. Resource
guides instruct principals about how to develop
role-specifc competencies. They can include written
cases or video clips highlighting examples of compe-
tency behaviors at diferent levels, suggestions about
practice activities, and instructions on how to access
training and mentoring opportunities to improve
specifc competencies.68
• “Stretch” roles or assignments. Aiming some ac-
tivities toward improving weaknesses can hasten
development. For example, a principal who lacks
self-confdence might establish a goal to give several
presentations at community meetings where he
practices “openly stating his own expertise or com-
paring himself positively with others.”69
• Mentoring. A mentor who is very strong in a prin-
cipal’s areas of weakness can provide rapid feedback
and guidance about improvement. Districts should
assign mentors with the explicit expectation that the
mentor will coach the principal in specifed areas
needing development. Research suggests that devel-
opment cannot be imposed on another person.70
• 19
Notes
1. Leithwood, K., Seashore Louis, K., Anderson, S., &
Wahlsrom, K. (2004). How leadership infuences student
learning. NY: Wallace Foundation. Retrieved fromhttp://mt.educarchile.cl/MT/jjbrunner/archives/libros/
Leadership.pdf
2. Hassel, E. A., & Hassel, B. C. (2009). The big u-turn:
How to bring schools from the brink of failure to stellar
success. Education Next, 9(1), 21–27. Retrieved from http://
educationnext.org/the-big-uturn/
3. Herman, R., Dawson, P., Dee, T., Greene, J., Maynard,
R., Redding, S., et al. (2008). Turning around chronically
low-performing schools: A practice guide. Washington,
DC: National Center for Education Evaluation and
Regional Assistance, Institute of Education Sciences, U.S.
Department of Education. Retrieved fromhttp://ies.ed.gov/
ncee/wwc/pdf/practiceguides/Turnaround_pg_04181.pdf;
Bossidy, L. (2001, March). The job no CEO should delegate.
Harvard Business Review, 79(3), 47–49; Brenneman, G.
(1998, September-October). Right away and all at once:
How we saved Continental. Harvard Business Review,
76(5), 162–179; Buchanan, L. (2003, December). The
turning of Atlanta. Harvard Business Review, 81(12), 18–19;
Hamel, G. (2000, July-August). Waking up IBM: How
a gang of unlikely rebels transformed Big Blue. Harvard
Business Review, 78(4), 137–146; Hirschhorn, L. (2002,
July) Campaigning for change. Harvard Business Review,
80(7), 6–11; Joyce, P. (2004, August). The role of leadership
in the turnaround of a local authority. Public Money &
Management, 24(4), 235–242; Kanter, R.M. (2003, June).
Leadership and the psychology of turnarounds. Harvard
Business Review, 81(6), 58–67; Kim, W.C. & Mauborgne, R.
(2003, April). Tipping point leadership. Harvard Business
Review, 81(4), 60–69; Riesiner, R.A.(2002, February).
When a turnaround stalls. Harvard Business Review, 80(2),
45–52.
4. Beer, M. & Nohria, N. (2000). Breaking the code of
change. Cambridge, MA: Harvard Business School Press.
5. For more information on how school district practices
ofen fail to support efective hiring practices, see: The
Teaching Commission. (2006). Teaching at risk: Progress &
potholes. The Teaching Commission; National Council on
Teacher Quality. (2007). 2007 State teacher policy yearbook.
Retrieved fromhttp://www.nctq.org/stpy/reports/
stpy_national.pdf; The New Teacher Project. (2007).
Hiring, assignment, and transfer in Chicago public schools.
Washington, DC: The New Teacher Project. Retrieved
fromhttp://www.tntp.org/fles/TNTPAnalysis-Chicago.
pdf; The New Teacher Project. (2009). The widget efect:
Our national failure to acknowledge and act on diferences in
teacher efectiveness. New York: The New Teacher Project.
6. Goldhaber, D. D. & Brewer, D. J. Does teacher
certifcation matter? High school teacher certifcation status
and student achievement. Educational Evaluation and Policy
Analysis, 22(2), 129-145; Kane, T. J., Rockof, J. E., & Staiger,
D. O. (2006). What does certifcation tell us about teacher
efectiveness? Evidence fom New York City (NBER Working
Paper No. W12155). Cambridge, MA: National Bureau
of Economic Research. Retrieved fromhttp://gseweb.
harvard.edu/news/features/kane/nycfellowsmarch2006.
pdf; Walsh, K., & Tracy, C. O. (2005). Increasing the odds:
How good qualities can yield better teachers. Washington,
DC: National Council on Teaching Quality; Goldhaber,
D. (2002). The mystery of good teaching. Education Next.
Stanford, CA: Hoover Institution; Kane, T. J., Rockof,
J.E., & Staiger, D. O. Photo fnish: Teacher certifcation
doesn’t guarantee a winner. Education Next , 7(1); Chaney,
B. (1995). Student outcomes and the professional preparation
of eighth-grade teachers in science and mathematics. National
Science Foundation. For more on false assumptions about
performance predictiveness of traditional measures, see
McClelland, D. (1973). Testing for competency rather than
intelligence. American Psychologist, 28, 1–14.; Goldhaber, D.,
& Brewer, D. J. (2000).
7. Ibid.
8. Hunter, J. E., Schmidt, F. L., & Judiesch, M. K. (1990).
Individual diferences in output variability as a function of
job complexity. Journal of Applied Psychology, 75, 28–42.
9. Spencer, L. M. & Spencer, S. M. (1993). Competence
at work: Models for superior performance. New York: John
Wiley & Sons, Inc.
10. Examples of education organizations in the U.S. that
incorporate competencies in their selection process include
the University of Virginia School Turnaround Specialist
Program, Teach for America, The New Teacher Project,
and The Academy of Urban School Leadership.
11. Steiner, L. (2010). Using competency-based evaluation
to drive teacher excellence: Lessons fom Singapore.
Chapel Hill, NC: Public Impact. Retrieved from http://
opportunityculture.org/images/stories/singapore_
lessons_2010.pdf; Hobby, R., Crabtree, S., & Ibbetson, J.
(2004). The school recruitment handbook: A guide to
attracting, selecting and keeping outstanding teachers.
London and New York: RoutledgeFalmer.
12. For the past four years (2006–2009), Singapore has
ranked among the top four countries in the world on the
Trends in International Mathematics and Science Study
(TIMSS) science and math tests and the Progress in
International Reading Literacy Study (PIRLS) reading test.
20 • notes
See: Mullis, I. V. S., Martin, M. O., & Foy, P. (2008). TIMSS
2007 international mathematics report: Findings fom IEA’s
trends in international mathematics and science study at
the fourth and eighth grades. Cambridge, MA: TIMSS
& PIRLS International Study Center, Boston College.
Retrieved fromhttp://timss.bc.edu/TIMSS2007/intl_
reports.html; Martin, M.O., Mullis, I.V.S., & Foy, P. (2008).
TIMMS 2007 international science report: Findings fom
IEA’s trends in international mathematics and science study
at the fourth and eighth grades. Cambridge, MA: TIMSS
& PIRLS International Study Center, Boston College.
Retrieved fromhttp://timss.bc.edu/TIMSS2007/intl_
reports.html; Mullis, I.V.S., Martin, M.O., Kennedy, A.M.,
& Foy, P. (2006). IEA’s progress in international reading
literacy study in primary school in 40 countries. Cambridge,
MA: TIMSS & PIRLS International Study Center, Boston
College. Retrieved fromhttp://timss.bc.edu/pirls2006/
intl_rpt.html
13. Steiner, L. (2010).
14. Mourshed, M., Chijioke, C., & Barber, M. (2010).
How the world’s most improved school systems keep getting
better. McKinsey. Retrieved fromhttp://ssomckinsey.
darbyflms.com/reports/EducationBook_A4%20
SINGLES_DEC%202.pdf
15. Spencer & Spencer. (1993).
16. Spencer & Spencer. (1993).
17. Spencer, L. M., McClelland, D. C., & Spencer, S. M.
(1992). Competency assessment methods: History and state of
the art. Hay/McBer Research Press.
18. Lees, A. & Barnard, D. (1999). Highly efective
headteachers: An analysis of a sample of diagnostic data
fom the Leadership Programme for Serving Headteachers.
London: The Hay Group; Boyatzis, R. E. (1999). Developing
emotional intelligence (unpublished paper). Cleveland:
Case Western Reserve University, Department of
Organizational Behavior; Goleman, D. (1998). Working with
emotional intelligence. NY : Batnam; Jacobs, R. L. (2001).
Using human resource functions to enhance emotional
intelligence. In C. Cherniss & D. Goleman (Eds.), The
emotionally intelligent workplace. San Francisco: Jossey-
Bass; Nygren, D. J. and Ukeritis, M. D. (1993). The future of
religious orders in the United States. NY: Praeger; Spencer,
L. M. (2001). The economic value of emotional intelligence
competencies and EIC-based HR programs. In C. Cherniss
& D. Goleman (Eds.), The emotionally intelligent workplace.
San Francisco: Jossey-Bass; Williams, D. (1994). Leadership
for the 21
st
Century: Life insurance leadership study. Boston:
Hay Group; Boyatzis, R. E., Goleman, D., & Rhee, K.
(2000). Clustering competence in emotional intelligence:
Insights from the Emotional Competence Inventory (ECI)
s. In R. Bar-On and J. D. A. Parker (Eds.), Handbook of
emotional intelligence. San Francisco: Jossey-Bass, 343–362.
19. The critical incident method was developed by J. C.
Flanagan. See Flanagan, J. C. (1954). The critical incident
technique. Psychological Bulletin, 51, 327–358. McClelland’s
motivation probes are described in McClelland, D. (1989).
Human motivation. Cambridge, UK: Cambridge University
Press.
20. Spencer & Spencer. (1993).
21. Taylor, P. J. & Small, B. (2002). Asking applicants
what they would do versus what they did do: A meta-
analytic comparison of situational and past behavior
employment interview questions. Journal of Occupational
and Organizational Psychology, 75, 277–294; Krajewski,
H. T., Gofn, R. D., McCarthy, J. M., Rothstein, M. G., and
Johnston, N. (2006). Comparing the validity of structured
interviews for managerial-level employees: Should we look
to the past or focus on the future? Journal of Occupational
and Organizational Psychology, 79, 411–432; Pulakos, E. D.
and Schmitt, N. (1995). Experience-based and situational
interview questions: Studies of validity. Personnel Psychology,
48, 289–308; Boyatzis, R. E. (1982). The competent manager:
A model for efective performance. New York: Wiley-
Interscience; McDaniel, M. A., Whetzel, D. L., Schmidt,
F. L., & Maurer, S. D. (1994). The validity of employment
interviews: A comprehensive review and meta-analysis.
Journal of Applied Psychology, 79(4), 599–616; Motowidlo,
S. J., Carter, G. W., Dunnette, M. D., Tippins, N., Werner,
S., Burnett, J. R., & Vaughan, M. J. (1992). Studies of
the structured behavioral interview. Journal of Applied
Psychology, 77(5), 571–587; Campion, M.A., Campion, J.E.
(1994). Structured interviewing: A note on incremental
validity and alternative question types. Journal of Applied
Psychology, 79(6), 998–1002.
22. Orpen, C. (1985). Patterned behavior description
interviews versus unstructured interviews: A comparative
validity study. Journal of Applied Psychology, 70, 774–776.
23. McClelland, D. (1998). Identifying competencies with
behavioral event interviews. Psychological Science, 9, 331–339.
24. Hay Group. (2003). Using competencies to identify
high performers: An overview of the basics. Hay Group
Research Press. Retrieved fromhttp://www.haygroup.com/
downloads/uk/Competencies_and_high_performance.pdf.
25. Spencer & Spencer. (1993).
26. Spencer, McClelland, & Spencer. (1992); Spencer &
Spencer. (1993).
27. All competency descriptions provided here derive
from Competence at work, Spencer & Spencer. (1993). See p.
26 for Achievement Orientation Scale.
28. All competency descriptions provided here derive
from Competence at work, Spencer & Spencer. (1993). See p.
46 for Impact and Infuence Scale.
29. Goleman, D. (1998, November/December).
What makes a leader. Harvard Business Review.
Retrieved fromhttp://hbr.org/product/what-
makes-a-leader-harvard-business-review/an/
R0401H-PDF-ENG
30. Public Impact. (2007). School turnarounds: A review
notes • 21
of the cross-sector evidence on dramatic organizational
improvement. Lincoln, IL: Public Impact for The Center
for Comprehensive School Reform and Improvement.
Retrieved fromhttp://www.centerii.org/survey/downloads/
Turnarounds-Color.pdf
31. Hay Group. (2003).
32. Schmidt, R. & Hunter, J. (1998). The validity and
utility of selection methods in personnel psychology:
Practical and theoretical implications of 85 years of research
fndings. Psychological Bulletin, 124(2), 262–274; Spencer &
Spencer. (1993).
33. The Public Impact competency model for school
turnaround leaders that is used by the UVA School
Turnaround Specialist Program was developed using the
“stepladder approach” of building a model from related,
validated models. The competencies included in this
model stem primarily from validated competency studies
of highly successful leaders in analogous leadership roles
(entrepreneurs and managers in complex organizations).
These studies are reported in Spencer & Spencer, 1993. The
“stepladder approach” is described in detail in Appendix A.
34. Public Impact. (2008).
35. The New Teacher Project. (2009); Toch, T. &
Rotherham, R. (2008). Rush to judgment: Teacher evaluation
in public education. Washington, DC: Education Sector;
Reeves, D. (2009). Assessing educational leaders: Evaluating
performance for improved individual and organizational
results. Thousand Oaks, CA: Corwin Press.
36. Reeves, D. (2009); Stine, D. O. Developing an
evaluation system to improve principal performance and
accountability. Paper presented at the annual meeting of the
American Educational Research Association, Seattle, April
2001. 28 pages. ED 452 278.
37. Thomas, S. L. & Bretz, R. D., Jr. (1994, Spring).
Research and practice in performance appraisal: Evaluating
employee performance in America’s largest companies. SAM
Advanced Management Journal; The Hay Group (2010,
June). View point: Performing in uncertain times. Author:
Issue 3.
38. Spencer & Spencer. (1993); Norton, R. S. & Kaplan,
D. P. (1996). Using the balanced scorecard as a strategic
management system. Harvard Business Review; The Hay
Group (2010, June).
39. See Public Impact (2009). Try, try again: How to triple
the number of fxed failing schools without getting any better
at fxing schools. Retrieved fromhttp://publicimpact.com/
publications/Public_Impact_Try_Try_Again_Slide_
August_2009.pdf
40. Burke, M. J. & Day, R. R. (1986). A cumulative
study of the efectiveness of managerial training. Journal of
Applied Psychology, 71, 232–245.; Spencer & Spencer, 1993.
41. Individuals must feel motivated to improve. Hobby,
Crabtree, & Ibbetson. (2004).
42. Herman et al., (2008); Public Impact (2007).
43. McClelland, D. (1998).
44. All competence descriptions derived from Spencer &
Spencer, Competence at work. (1993).
45. Kowal, J., Hassel, E. A., & Hassel, B. C. (2009).
Successful school turnarounds: Seven steps for district
leaders. Washington, DC: Public Impact for The Center
for Comprehensive School Reform and Improvement.
Retrieved fromhttp://www.centerforcsri.org/fles/
CenterIssueBriefSept09.pdf
46. Duke, D. L., Tucker, P. D., Belcher, M., Crews, D.,
Harrison-Coleman, J., Higgins, J., et al. (2005). Lif-of:
Launching the school turnaround process in 10 Virginia
schools. Charlottesville, VA: Darden-Curry Partnership for
Leaders in Education. Retrieved fromhttp://www.darden.
virginia.edu/web/uploadedFiles/Darden/Darden_Curry_
PLE/UVA_School_Turnaround/LifOf.pdf; Public
Impact. (2007).
47. Calkins, A., Guenther, W., Belfore, G., & Lash,
D. (2007). The turnaround challenge: Why America’s best
opportunity to improve student achievement lies in our
worst-performing
schools. Boston: Mass Insight. Retrieved from http://
www.massinsight.org/publications/turnaround/51/fle/1/
pubs/2010/04/15/TheTurnaroundChallenge_MainReport.
pdf; Herman et al. (2008); Public Impact. (2007).
48. Herman et al. (2008); Public Impact. (2007).
49. Public Impact. (2007); Herman, et al. (2008).
50. Spencer, McClelland, & Spencer. (1992).
51. Descriptions of model development are derived from
several sources: Hay Group, 2003; Spencer, McClelland, &
Spencer, 1992; Spencer & Spencer, 1993; McClelland, 1998.
52. Spencer & Spencer. (1993); Hay Group. (2003).
53. Spencer & Spencer. (1993).
54. Spencer, McClelland, & Spencer. (1992).
55. Competency defnitions used here are derived from
Spencer & Spencer, Competence at work. (1993).
56. Spencer & Spencer. (1993).
57. Spencer & Spencer. (1993).
58. Hay Group. (2003).
59. Spencer, McClelland, & Spencer. (1992).
60. For example, the turnaround teacher and leader
models created by Public Impact were developed frst by
examining the research base on successful turnaround leader
actions and then mapping these to validated competency
models of similar roles (for leaders: managers in complex
organizations and entrepreneurs; for teachers: models
of teachers from other nations and studies of teachers
successful with high-poverty populations). Future validation
and editing — or from-scratch modeling — would be ideal,
when a larger pool of school turnaround leaders and
turnaround teachers with measured results are available for
study. Seehttp://www.publicimpact.com/human-capital/
competencies-of-high-performers for the most recent models
and selection tools.
22 • notes
61. Hay Group. (2003).
62. Schmidt, R. & Hunter, J. (1998). The validity and
utility of selection methods in personnel psychology:
Practical and theoretical implications of 85 years of research
fndings. Psychological Bulletin, 124(2), 262–274; Spencer &
Spencer. (1993).
63. For step-by-step guidance on conducting a BEI
interview for a school turnaround leader, see Public Impact’s
School turnaround leaders: Selection toolkit, available athttp://schoolturnarounds.org/.
64. Spencer, L. M., McClelland, D. C., & Spencer, S. M.
(1992). Competency assessment methods: History and state of
the art. Hay/McBer Research Press, p. 4
65. Spencer & Spencer. (1993).
66. Burke & Day. (1986); Spencer & Spencer. (1993).
67. Spencer & Spencer. (1993). p. 290.
68. Spencer & Spencer. (1993).
69. All competency descriptions provided here derive
from Spencer & Spencer, Competence at work (1993).
70. Spencer & Spencer. (1993).
doc_805435937.pdf