iir*n ion-DeficaA

Hyperactivity
Disorder Test
A Method for ldentifying
lndividuals with ADHD

EXAMINER'S MANUAL

James E. Gilliam

I
 Attention-Deficit/
Hyperactivity Disorder Test
A Method for ldentifying
lndividuals with ADHD-

EXAMINER'S MANUAL

James E. Gilliam

@e,r,*,g,sl
8700 Shoal Creek Boulevard
Austin, Texas 7 87 57 -6897
800/897-3202 Fax 800/397-7633
0rder online at http://www.proedinc.com
 Contents

Acknowledgments o v

I Overview of the ADHDT o !

Information About ADHD o l
Description of ADHD t I
Assessment of Children Suspected of Having ADHD c !
Description of theADHDT o !
Uses of the ADHDT r j
SummarY o $

2 Administration and Scoring t f

General Administration Pr -,reiu:es c i
Specific Administration Pr':,cedures r J
Scoring Procedures . \
SummarY ' 10

3 Interpreting the Results . 1l

Completing the Summar)*,i Response Form r 1l
Interpreting Test Scores , 72
Using the Standard Error of Measurement o l3
Using ADHDT Scores to Identify Persons with ADHD . 13
Sharing the Results . 14
Summary . 15

4 Development and Technical Characteristics o 17

Item Selection r 17
Normative Procedures . 17

lll
 Reliability . 27
ValiditY . 23
Summary.3l

References o 33

Appendix: Normative Tables . Bs

Table A. Converting Raw Scores to Standard Scores and Percentiles (Males) . 36
Table B. Converting Raw Scores to Standard Scores and Percentiles (Females) . 36
Table C. Converting Sum of Standard Scores to ADHD Quotient and percentile Rank . 37

iv
 Acknowledgments

The development and norming of the ADHDT took Jeanne Warnecke, Andrew Champion, Patricia Fager-
approximately 2 years from conception to completion. quist, Donna Hook, Barbara Adams, Rayenell Beltran;
During that time period. I received generous advice and (Colorodo) Marcia Becker, Doug Hershey, Debra Hall
assistance from manv people. To ail of them, I owe much Ownbey, Maureen Nelson, Rebecca Martin, Mary Hill,
gratitude. I am especiallv indebted to specific individuals, Linda Adams; (Connecticul) Leslie Orendorf, Cathy
and I want to take this opportunitl, to acknowledge their Ziegler, Lori Monico, Gene E. Panciera, Janice Patzold,
unique contributions to the final version of the ADHDT. Jo Ann Z. Sebastianelli; (Delaware) Pauline Powell,
Kathleen Fad convinced me of the need for a test on Barbara Goode; (Florido) Marsha Strong, Charles Clifton,
Attention-Deficit,/Hyperactivitr. Disorder and encouraged Marla Bell, Karen Bachman, Jodie Adams, Parrish Bar-
me to write it. Ronnie Livingston rras involved in the test's wick, Rocky Fucile, Joan Gast, Janet M. Burgess, Denise
early development and contributed significantly to the M. Dickenson, Sarah Cripe, Rayann Mitchell, Sally M. Lutz,
initial pool of test items. Don Hammill. as always, pro- Elizabeth Edwards, Ingrid Lundquist, Kimberly Garcia,
vided his masterful touch and editorial st1'le to the final Patricia DiMeo, Cindy R. Burse, Michael P. Ferrentino,
product. Kueifen Nancy Yang rr.orked diligently crunch- Loretta Berry, Art Chiodo, Janet Barnes, P. A. Arnold, Lee
ing numbers and making statistical sense of huge amounts Sorenson, Carol Bernich, Joan Ewing; (Georgia) Larry
of data. To each of them. I ofier a verv special thank you. Mann, Lisa H. Baxter, E. Holloman, Jeania Howard,
I am also indebted to the dedicated practitioners and Connie Capehart, Paige Marlow, Barbara Merckling,
parents who contributed data in the standardization Alecia Green, Gloria Williams, Kathy Shook, Dawn Dur-
process. The norming of the .{DHDT could not have taken ham, Doris Drummond; (Hawaii) Mari Ann Lum Arveson;
place without the generous cooperation and assistance (ldaho) Carla Domitz, Shelley Sparkman, Tina Witty; (111r-
of the following people: (ALabama ) Randi Block, Debbie nols) Talene Carmody, Susan G. Bower, A. Youngmann,
Minor, Kelly B. Ryan, Angela \\ iglel'. \aomi Buckner, Chris Bohlmann, Dorothy Hartigan, Joann Tafoya, Amy
Brenda Brantley, Cindv Croft: \Alaska) Betsy Bartlet; Diepholz, Connie Fischer, Winifred M. Brown, Tammy
(Arizono) Luann M. Carpino. Tina Appleton, Sandra Mores, Marni Foderaro, Leslie Gordon, Barbara Green-
Fields, Julie Barrow, Nlarjorie Bohrer. Christina Allen, berg, Bette Bradley, Debra S. Honegger, Debra Tuccori,
Mr. Thompson; (Arkansas) Andrea Buck, Anne M. Crosby, Vickie Bowers, Celeste Flanagan, Kenneth E. Harris,
Jaimie Wilcox, Glenda Jenkin: (California) Donald G. Claudette Bialka, Tami Gelsosomo, Lindy Greenlee, Mary-
Huffman, Mary McCarthy-Bernard. Yvonne Scott, Frances anne Smiley, Lindalou Foellner, Donna Kastl, Kathleen
Felix, Martin R. Brown, AIIie L. Almore, Mary Smith, White, Jill Manley, Minerva Alvarez; (lndiona) Myra
Linda Arizmerdi, Ann Edelson. Cheng-Lin Chang, Leslie Akins, Mary Blocher, Beverly Wilson, Aron Hartford,
Banks, Erin Valley, Michaelann Dievendorf, John W. Merry Hiatt, Mary C. Hall, Donna F. Hixon, Cynthia Biggs;
Mossman, Barbara Beer, Louise G. Viales, Phyllis Field- (lowa) Elaine Baughman, Terre Hirst, LeaAnn Albers,
ing Campbell, J. Belcher, Mar-v Clemons. Doris A. Akin, Shirlee Johnson, Donna Smith, Amy Mullenberg, Joanne
Cheryl Kealy, Sheila Bechthold, Jennie Lee C. Kordes, Hinrichs, Kim Gess, Anita M. Hill, Judy Arner, Becky
a

Coppess, Rachelle Coomes, Paula Grant, Alice Terpening; Ashling, Lora Myers, Valerie Hare, Christopher S. Hull,
(Kansas) Heather Hunziker, Mary M. Geed, Catherine Pat Kloch, Rowland Nordma; (Pennsyluania)Sally Catch-
Whitcomb, B. J. Madewell, Teresa Miller, Sandra Harwick, pole, Dona Gilson, Carole Williamson, William L. Fox,
Michael R. Bruce, Retha Sleezer, Debra Camarena; (r(en- Merilyn Gills, Bill Calderwood, Cassandra K. Delong,
tucky) Janice Crick, Mary Freitag, Teresa Durkin, Mar- Carole Jeckell, Kimberly Griest, Dolores Brennan. Mary
garet Zuniga; (Louisiana) Robert E. Luttrell, Bobbi Jane Dawes, Susan Forrest, Paula Pagliai, Wendy Cee,
Brumfield, Jacqueline Ackel, Elvert M. Bijou; (Moine) S. Cwiklinski, Deborah Fischer Adamczyk, Melanie Bow-
Laurie Clark, Jacqueline Petrillo; (Maryland) Karen Dolan, ser, Carol Boone, Maryann Rihs: (Sourft Carolino) Patricia
Deborah C. Bunker, Sharon Beebe, Kelly Jenkins, Andrea Carson, Pamela Elaine Hardin, Wa1,ne Austin, Linda John-
Batchelder, Theodora E. Karczewski, Edna E. McPherson; son, Nona Boff, Margaret Jamison. Delores C. McAlister;
(Mossachuseffs) Ron Ackerman, Barbara A. Kelley, Vir- (South Dakoto) Charla O'Dea. Ruth Stabile (Tennessee)
ginia Kaylor, Noel-Sydney D avis; (M ic h i gan ) Dawn Donner, Amy Clark, Lisa Evans, Doris Richmond, Kaihy Edwards,
Jose Puente, Jan Baum, Michelle Braun, Nancy Garrett, Linda Milliron, Audrey Collins, Candyce H. Williams
Pam Rzepecki, Dorothy Berzins, Rondi Fry, Amy Koomier, Glaser, Becky Handy; (Texas) Denzil Flood, Jettie Win-
Linda Neal, Ray Franzen, Beverly Case, Connie Scripps, ston, Linda Kuisk, Kathy Mason, Tim Smith, Ivan Vance,
Beth Lascari, Jeff Mozdzierz, Dian Flynn, Sharon Har- Kristen Rinn, Minnie Edwards, Jereme Scott, Doneta
wood Kramer, Julie Aschbacher, Karen Cordes, Leslie J. Baughman, Gayla Gear, Marilyn Green, Myrlene Read,
Burton, Joanne Kellz; (Minnesota) Helen Peel, Dave Cathy Dawson, Susie Clark, La Juan Garrett. Beth Ker-
Barte, Susan Mclough, Lois L. Bruns, Sue Hynes, Carol zee, Tandy D. Girouard, Lynette Blough, Jeanette McKen-
Falkner, Duane Anderson, Paul Paetzel, Sue Witzel, Tracy non, Lauren Daleo, Grace Mueller. Lorna P. Benner,
Albrecht, Cathy Erickson; (Mississippi) Grace Dickerson, Jenny Greene, M. Benavides, April Proffitt, Susan M.
Jessie M. Adams, John Dorrough, Dena Adams; (l{issouri) Davis, Cindy T. Mclnnis, Sandra Moore, Eva Dominguez,
Judy Wollberg, Mary Ellen O'Hare, Ruth Littman Block, Judith Bowman, Rebecca Brauchie, Delia Parra, Mary
Nadine Barr, Meresa Dulinsky, Kristine Banderman, Shelba Alfert, Lucy Cerda; (Utah)Patricia Tucker, Shirley Crocker,
Knight, Lori L. Johnnson, Jackie Haynes, Lois Brandt, C. Russell B. Gardner, JoAnn Johnson, Yvonne S. Hansen;
Blien-Sharp; (trfiontana) Nancy Padon, Tammy Otten, Kara (Vtrginia) Jill Hockenbury, Susan Crenshaw, Jessica Bates,
Hansen, Erica Beyl, Jane C. Jindrich; (Nebraska) Holly Phillip C. Fellows, TerriAcbel; (Washington)Carol Robl,
Brooks, Mary Jo Hansen, R. Mangus, Nadine Karr, Donna Genese Harney, Kerry Arnold, Shannon D. Markley, M.
Johnson, M. J. Graham, Kari Stark, Pat Deutsch; (New Bean, Kathy Admire, David Gutschmidt, Cara Barringer,
Hampshire) Bette Stow; QVeu Jersey) Sue Evans, Linda Loren Imus, Lorraine Shank, Rebecca Shipley, Carol A.
Goldstein; Jeri Byrne, Anita Clark, Pat Marchant, Patricia Dyer, Kelly Stile, Gena Bradford, Judith Bahl, Gwen
Parisi, Gail Fazio, Janet Courtney, Kathleen Marasco; (Nert Farrow; (Wisconsin) Gina Interrante, Devora Gelin, LuAnn
Mexico) Susan Maynard, Byron Adair, Caroline Garrick, Alsleben, Vilma Bivens, Sue Long, Christiane M. Petra-
Kate Malone, S. E: Bourgeault; (I/ero York) Carol Aaron, kopoulos, Heidi M. Bragger, M. Croysdale, Sue Dal Santo,
Frances R. Brida, Pam Bezek, Liona Mattson, Bonita Jean Truesdale, Lynette Kasparek; (West Viginia) Lynne
Nadien, Colleen Mondi, Cathy Boudreau, Jan Acquisto, Dunmire, Barbara Albaugh; (Wyoming) Kathy A. Picker-
L. Trotta Boehler, Colleen Griffin, Jacqueline B. Faber, ing, Marilyn Horsley, Laura Dolezal.
Victor Acosta, Maureen Billey; (Neuado) Bette Ann Ander- Special thanks go to Nils Pearson and the staff mem-
son, Judith Barnes, Belinda Brumfield, Donna Spear;(North bers of PRO-ED's research and production departrnents
Carolina) Margaret Casey, Penny Loschin, Nancy C. for their suggestions, assistance, and encouragement in
Hiatt, Betty J. Armfield, Susan Carlton, Cynthia Hender- the development of the ADHDT.
son, Carol G. Hazel, Brenda M. Dunn, Mary H. Bryan;
(North Dakota) Elaine Broker, Peggy Erickson; (Ohio)
Debra Benedum, Peggy Back, Katie Fox, Cheryl Chesney,
Meg Clark, Monica Bettinger, Theresa Barbato, Nancy Note
Belleville, Lynn Callahan, Lynn Blakely, Phyllis Lammers,
Karen L. Chapin, Susan Freeman, Pam Conine, Charles Clinicians and researchers who use the ADHDT are
Massie, Alfreda Havens, Jennifer Walton, Mary Anne encouraged to send copies of their work to the author
Izenour, Brenda Zackery, Charlotte Helies, Kenna Fegan, in care of PRO-ED, 8700 Shoal Creek Boulevard, Austin,
Elois M. Dunham, Karen A. Kane, Therese A. Farber; Texas 78757. Appropriate and well-designed research
(Oklahoma) Patricia Fellabaum, Paula Burdette, Sandra using the ADHDT will be cited in future revisions of the
Cranfill, Armando L. Corral, Suzanne Dunn, Gazell Hughes, test and its manual. Suggestions for improving the instru-
Debra Cald; (Oregon) Liz Healy, Anne Brown, Ruth ment are also welcomed.

vl
 Overview of the ADHDT

The Attention-Deficit /Hy peractiuity Disorder Test (ADHDT) been present before age 7. Persons with ADHD generally
is a standardized. norm-referenced test that contrib- display impairments in all three problem areas but to
utes to the diagnosis of students with Attention-Deficit/ varying degrees. These impairments must be present in
Hyperactivity Disorder (ADHD). Based on the DSM-IV at least two environmental settings (e.g., home, school,
(American Psychiatric Association, 1994) definition of or work), and there must be clear evidence of interfer-
ADHD, the ADHDT is comprised of three subtests: Hyper- ence with developmentally appropriate social, academic,
activity, Impulsivity, and Inattention. The test is easily or occupational functioning.
administered in a short amount of time by parents. teach- ADHD has only recently been recognized as a dis-
ers, and others who have had sustained contact with the tinct disorder (Nussbaum & Bigler, 1990), but parents,
subject. Scoring is simple, and guidelines are provided to educators, and clinicians are becominq more aware and
assist the examiner in interpreting the results. Reliability knowledgeable about it. According to Shaywitz and Shay-
and validity ratings are exceptionally strong and support witz (1992), ADHD is one of the most frequently diagnosed
its use as a diagnostic instrument. neurobehavioral disorders in childhood, affecting chil-
This manual provides basic information about ADHD, dren from early infancy through school and adult life.
instructions for administering and scoring the test, proce- Ingersoll and Goldstein (1993) reported that conservative
dures for interpreting the results, and technical infor- estimates suggest the prevalence at 3% to 5% of all schorl-
mation about the instrument. In this particular chapter, age children. Safer and Krager (1988) stated that the dis-
general information about Attention-Def icit/Hyperactivity order is being diagnosed more frequently now than a
Disorder is provided. Diagnostic criteria for ADHD are decade ago.
presented, and suggestions for identifl,ing ADHD are Historically the diagnosis of ADHD has been difficult
described. Finally, a description of the ADHDT is given, to make because of disagreement concerning whether the
and its uses are discussed. condition was a biological brain disorder or a behavioral
response to certain environments such as school or other
situations where demands were placed on the child. Lack
of agreement about the definition of ADHD also con-
lnformation About ADHD tributed to the controversy. Labels such as minimal brain
injury, minimal brain dysfunction, and minimal cerebral
According to the American Psychiatric Association (199a), dysfunction are only a few of the terms that were used
"The essential f eature of Attention-Def icit/Hyperactivity to categorize children who manifested the disorder. Most
Disorder is a persistent pattern of inattention. impulsivity, of the early diagnostic labels had some connection with
and,/or hyperactivity-impulsivity that is more frequent neurological damage. This was due in part to the fact that
and severe than is typically observed in individuals at a children and adults who had suffered some type of brain
comparable level of development" (p. 78). The symptoms damage were often distractible, impulsive, and hyper-
of hyperactivity, impulsivity, or inattention must have active (Nussbaum & Bigler, 1990). However, no clear
r
neurological impairment has been demonstrated for most (e) often has difficulty organizing tasks and activities
children with ADHD, and no obvious disease exists (Bain, (f) often avoids, dislikes, or is reluctant to engage
1991). in tasks that require sustained mental effort (such
as schoolwork or homework)
Over time, the focus of attention shifted to the prob-
(g) often loses things necessary for tasks or activ-
lems these children presented: inattention, impulsivity,
ities (e.g., toys, school assignments, pencils,
and hyperactivity. These core symptoms were considered books, or tools)
to be the basis of attention-deficit disorder, and in 1980, (h) is often easily distracted bv extraneous stimuli
attention deficit disorder was included as a disorder in (i) is often forgetful in dailv activities
the DSM-lll (American Psychiatric Association, 1980). The
Iabel has since been modified, and the current diagnos- (2) six (or more) of the following symptoms of
tic nomenclature is "Attention-Deficit/Hyperactivity Dis- hyperactivity-impulsivity have persisted for at
least 6 months to a degree that is maladaptive and
order" (American Psychiatric Association, 1994).
inconsistent with der.elopnrentaI level:

Hyperactiuity
(a) often fidgets with hands or feet or squirms in seat
Description of ADHD (b) often leaves seat in classroom or in other situa-
tions in which remaining seated is expected
When discussing ADHD, most people rely on the descrip- (c) often runs about or climbs excessively in sit-
tion of the condition from the American Psychiatric Asso- uaticns in which lt is inappropriate (in adoles-
ciation (1994)published in the DSM-IV. Individuals who cents or adults, may be limited to subjective
assess ADHD need to be familiar with this description. feelings of restlessness)
In this section, the DSM-IV diagnostic criteria are pre- (d) often has difficulty playing or engaging in leisure
sented. In addition, the features that are frequently asso- activities quietly
ciated with ADHD, age at onset, and prevalence, sex ratio, (e) is often "on the go" or often acts as if "driven
by a motor"
and outcome are discussed.
(f) often talks excessively

Impulsiuity
Diagnostic Criteria for ADHD (g) often blurts out answers before questions have
been completed
The primary symptoms of ADHD are inattention, impul- (h) often has difiiculty awaiting turn
sivity, and hyperactivity (American Psychiatric Asso- (h) often interrupts or intrudes on others (e.g., butts
ciation, 1994). These primary symptoms may vary in into conversations or eames)
intensity, and other symptoms may coexist with these
core symptoms. People with ADHD generally manifest Some hyperactive-impulsive or inattentive symptoms
that caused impairment were present before age 7 years.
their symptoms in most situations, including school, home,
work, and social situations, but to varying degrees. C. Some impairment from the symptoms is present in two
The DSM-IV diagnostic criteria for ADHD (American or more settings (e.g., at school [or work] and at home).
Psychiatric Association, 1994) are presented here:
D. There must be clear evidence of clinically sigrrificant
A. Either (1) or (2): impairment in social, academic, or occupational func-
tioning.
(1) six (or more) of the following symptoms of inatten-
tion have persisted for at least 6 months to a degree E. The symptoms do not occur exclusively during the
that is maladaptive and inconsistent with develop- course of a Pervasive Developmental Disorder, Schizo-
mental level: phrenia or other Psychotic Disorder, and are not better
accounted for by another mental disorder (e.g., Mood
Inattention Disorder, Anxiety Disorder, Dissociative Disorder or a
(a) often lails to give close attention to details or Personality Disorder).
makes careless mistakes in schoolwork, work.
or other activities
Code based on type:
(b) often has difficulty sustaining attention in tasks
314.01 Attention-Deficit/Hyperactivity Disorder, Com-
or play activities
bined Tlpe: if both Criteria A1 and A2 are met
(c) often does not seem to listen when spoken to
for the past 6 months
directly
(d) often does not follow through on instructions and 3f 4.0O Attention'Deficit/Hyperactivity Disorder, Pre-
fails to finish schoolwork, chores, or duties in the dominantly Inattentive Type: if Criterion A1
workplace (not due to oppositional behavior or is met but Criterion A2 is not met for the past
failure to understand instructions) 6 months
 also occur'
severe problems of antisocial behavior may
3 I 4.0 f Attention'Deficit'/Hyperactivity Disorder' Pre' stealing' and per-
dominantly Hyperactive'Impulsive Tlpe: if for example, aggression towards others,
forms of behavior may
Criterion A2 is met but criterion A1 is not met
{or sistent ruie breaking. These severe
the Past 6 months result in a diagnosis of conduct disorder'
Conduct disorder is sometimes associated with
ADHD
and 1994)' According to
Coding note: For individuals {especiall.v adolescents (American Psychiatric Association,
that no longer meet disorders and ADHD
adults)"who currentlv have svmptoms bain (1991), the overlap of conduct
full criteria. ln partial Remission" should be specified' may be as high as one-third to one-half ' Conduct
disorder

314.9 Attention'Deficit/Hyperactivity Disorder Not is iiagnosed when a child demonstrates a persistent'

Otherwise SPecified .".u.rIng pattern of rule violations that violates the
This category is for disorders with prominant symptoms basic riglti of others' violates the law' and is signif-
of inattention or hyperactivity-impulsivity that do not icantly different from age-appropriate social norms'
meet criteria for Attention-Deficit'/Hyperactivity Disorder' Juvenile delinquency, alcoholism, and substance
abuse
problems that result
(pp. 83-85) are common in this disorder' The
from the combination of ADHD and conduct disorders
make education and treatment difficult' Early identifica-
tion and treatment are essential for children who have
Associated Features of ADHD
both disorders.
Other disorders that have been noted with ADHD
are
that
Associated features of ADHD are those characteristics Tourette's Syn-
include coordination problems, tic disorders and
are often, but not invariably, present' These may
and coexisting disorders' Persons with drome, epilepsy, and mental retardation' The author
both symptoms
have other handicapping conditions and may cannot possibly describe all of the behavioral disorders
ADHDmay ADHD'
vary and psychiatric conditions that might coexist with
demonstrate a variety of associated problems that with ADHD
of self-esteem are common' as well Uut [fre reader should recognize that persons
with age. Problems mood' and
of mood, low frustration tolerance' can manifest other problems of personality'
u, ,upid fluctuations Exami-
anxiety, as well as other psychiatric conditions'
and temper tantrums. Problems in school, particularly are
ners should seek information and counsel when they
academic underachievement, are characteristic of
chil-
other
such as oppositional concerned about the possibility of the existence of
dren with ADHD. Other disorCers
disorders with ADHD.
defiant disorder, conduct disorder, and specific develop-
mental disorders may occur with ADHD'
As a result of their core symptoms, children with
ADHD often have difficulty relating and interacting with Age at Onset
other children. The impulsivity characteristic is the malor
problem. Wanting to be first in line' taking things that One of the diagnostic criterion for ADHD is onset
before
tth"r, are using, and quickly losing interest in activities the age of 7. One subgroup of children with ADHD n'ay
them and
are characteristic of children with ADHD and cause age (betweenZVz 5)'
be diignosed at a much earlier
to be unpopular with their peers, often leading to aliena-
This subgroup is comprised of children whose most prom-
tion. Such alienation can lead to reiection and negative inent syhptom is hyperactivity' These children are
self-esteem for the child. reported to demonstrate increased levels of motor activ-
The majority of children with ADHD also experience ity beginning at birth or early infancy and may display
academic problems. The inability to maintain attention ,tbUO symptoms throughout childhood and adolescence
and stay on task causes these children to have difficulty (Wendei, tggZ). I\4ost children with ADHD are not diag-
in schotl, particularly in activities that require focused nosed until they enter school (around the age of 5
or 6)'
attention, ictive listening, and completing tasks within
of
specific time periods (Zentall, 1993)' The incidence
population is high; some esti-
learning disabilities in this
Prevalence, Sex Ratio, and Outcome
mate as high as 30% to 40% (Bain, 1991)' However' one
must be cautious and not assume that when a child has
Nussbaum and Bigler (1990) pointed out that authori-
ADHD he or she will automatically have learning prob-
ties disagree about the prevalence of ADHD' Shaywitz
lems or a specific learning disability'
and Shaywit z (1992) stated that estimates suggest that
Some children with ADHD are oppositional or non-
ADHD aifects i0% to 20% of the school-age population'
compliant-refusing to comply with teachers' or parents'
Ingersoll and Goldstein (1993) estimated that at Ieast
requests, defying authority figures, and throwing temper
tanirums. Generally, these children's problems stem from
lX to Sy' of children under the age of 18 have ADHD'
According to Barkley (1981), a2o/o to 3% prevalence fig-
their core symptoms and not from malicious intent' More
3
 ure is generally accepted by most professionals. Males are
estimated to outnumber females about 6 to 1 (Nussbaum
Description of the ADHDT
& Bigler, 1990). Symptoms (particularty hyperactivity) The ADHDT is a behavior checklist used to identif,
seem to lessen as the child matures and often disappear
persons with Attention-Deficit/Hyperactivity Disordei
around puberty. Problems of attention and impulsivity
The test was normed on l,2Tg subjects with ADHD fron
may continue into later years. The child with ADHD
47 states and Canada. In the following section, the char
whose symptoms persist into adolescence and adulthood
acteristics of the ADHDT are listed, the individual sub
is at risk for developing emotional problems (Nussbaum
tests are described, and the components of the test art
& Bigler, 1990). As a result of his or her learning difficul- presented.
ties, the student with ADHD may have limited educational
opportunities and may not receive positive educational
and vocational experiences. Characteristics of the ADHDT

To achieve the stated purposes of the ADHDT. the tesl

has the following characteristics:
Assessment of Children Suspected o Thirty-six items describing behaviors
and characteristics
of Having ADHD of persons with ADHD are included.
. Items are based on the most common behavioral
Because ADHD is a behavioral syndrome, a person sus- problems of ADHD reported in the DSM_IV and profes_
pected of having this disorder should be systematically
sional literature.
observed in several different environments (e.e., home,
school, playground, and work). The purpose of observa_
. Scores from the ADHDT provide data for diagnosis of
tion in different environments is to rule out the effects ADHD and other behavioral disorders.
of a specific environment (such as school) on behavior. o The test is designed for use
by parents and professionals
He or she should also be observed in different activities and can be used at school and at home.
because some persons suspected of having ADHD may
o Both validity and reliability of the instrument
not demonstrate core symptoms when they are engaged are clearly
in pleasurable or nondemanding activities. It is also impor_
supportive of its stated purposes.
tant for the examiner to observe several average students o Behaviors are rated based on the
severity of the per_
who are the same age as the referred student to docu_ son's problems.
ment that the problem behaviors occur more frequently o It is designed for use with persons from ages 3 to 23.
or are more severe than those seen in his or her aver_
age peers. o Norms, based on a nationally representative
sample
The child suspected of ADHD should also be assessed of persons with ADHD, are presented for comparison
by a psychologist who can evaluate the child by using purposes.
psychological and neuropsychological tests that con_ o The amount of time necessary to
administer the test is
tribute to the diagnosis of ADHD. psychological testing minimal (approximately 5 to l0 minutes in most cases).
is important for gathering clinical information about the
o The test is easily completed by teachers, parents, and
child's emotional and behavioral problems and the rela-
tionship of these problems to his or her learning difficul_ others who have the most knowledge of the subject,s
ties. Psychological testing is also useful for differentiating behavior or the greatest opportunity to observe him
an attention problem from other learning disorders. or her.
Parents or caregivers should also be involved in the o Standard scores are provided for comparing the sub_
assessment. Because they have had the most sustained ject rated with persons with ADHD from the normative
contact, parents are an excellent source of information sample.
about their child's behavior. They have also had the most o A table is provided for determining the likelihood that
opportunity to observe the child in different situations
a subject has ADHD and the severity of the disorder.
under a variety of circumstances. Given adequate instruc_
tion and appropriate tests to use, parents can make sig-
nificant contributions to the diagnosis of ADHD. The Description of the Subtests
ADHDT was designed for use by parents, and a large
number of parents contributed information in the norm- The ADHDT is comprised of three subtests totalling
ing of the test. 36 items. The items are related to the three core symp-
 toms of ADHD discussed in the literature (Bain, 1991: test scores are then transferred to the space on the
Barkley, 1981; Nussbaum & Bigler. 1990). The first sub- Summary/Response Form designated Section II, Score
test, Hyperactivity, measures excessive motor movement Summary. Raw scores are converted to standard scores
and is comprised of items 1 through 13. Trvo examples and percentiles based upon normative data found in the
from this subtest are "Excessir.e running. lumping, climb- Appendix. To assist with interpreting the meaning of test
ing" and "Excessive talking. 'The second subtest, Impul- scores, an interpretation guide is provided in Section III.
sivit1,. assesses the problems of inhibiting behavior and The Profile of Scores in Section IV provides a graphic view
delaving making a response; it contains items 14 through of the test's standard scores. The results of other tests
23 An erample from this subtest is "Acts before think- that have been used with the subject are also recorded
ing." Inattention is the third subtest, which measures a in Section IV. The individual subtests are found in Sec-
person's problems of focusing and paying attention to the tion V, Response Form. The examiner records his or her
important features of a task. This subtest is comprised of score for each item on the test in this section. Instructions
items 24 through 36. Two examples from this subtest for administerinq each subtest are also found in this sec-
include "Fails to finish projects" and "Difficulty sustain- tion. Section VI provides key questions that the examiner
ing attention." needs to ask. Answers to these questions provide impor-
The test is completed by persons who are familiar tant information necessary for the diagnosis of ADHD.
with the subject. The ratings indicate the extent to which Section VII provides space for the examiner to record
the behaviors are seen as a problem for the individual. recommendations or other pertinent information regard-
The following guidelines are used to make the ratings: ing the subject.

0- Not a Problem
1= Mild Problem
)- Severe Problem Uses of the ADHDT
Scores are computed for each subtest. A total score is The ADHDT is well constructed and has strong psycho-
obtained by summing the stanCard scores for the subtests metric properties. Because of its sound construction and
and converting that value to a quotient. its strong technical characteristics, the ADHDT can be
used with confidence for the following purposes: (a) to
identify persons with ADHD, (b) to assess persons referred
Components of the ADHDT for behavioral problems, (c) to document progress in the
problem areas as a consequence of special intervention
The ADHDT consists of an examiner's manual and a set programs, (d) to target goals for change and intervention
of 50 Summary,/Response Forms. The following text pro- on the student's individualized education program (lEp),
vides a description of these components. and (e) to measure ADHD in research projects. Each of
these purposes is described in this section.
Examiner's Manual. This manual contains information
required to administer, score, and interpret the ADHDT.
Those who use the ADHDT rnust follow all instructions ldentification of Students with ADHD
and procedures described in the manual when administer-
ing, scoring, and interpreting results from the test. The One purpose of diagnosis is to identify those persons
manual also contains technical data pertaining to item who have a specific disorder. To DeMeyer, Hingtgen, and
analysis and selection, reliability, validitl,, and standardi- Jackson (1981), the requisites of a useful diagnostic sys-
zation of the instrument. Normative tables are published tem are that different diagnosticians using it will make
in the Appendix. the same diagnosis and that the results differentiate
related but separate conditions. A test should provide data
Summary,/Response Form. All pertinent informa- in the lorm of objective scores for determining those per-
tion related to the ADHDT is recorded on this form. In sons who are likely to have ADHD and those who are
Section I, demographic information about the subject is not. The ADHDT produces reliable and valid scores for
recorded, the rater and the examiner are identified, and this purpose. It can be used by a variety of individuals
the date the rating was made is given. The Summary,/ to accumulate data for identifying the person's behavioral
Response Form contains the three subtests and sections problems. Results from the ADHDT, along with other test
for recording, scoring, and interpreting results. When results, behavioral observations, case histories, parent
scoring the ADHDT, the examiner adds the item scores interviews, and so forth, provide valuable information for
and derives a total score for each subtest. The sub- diagnosing ADHD.
 Diagnosis is difficult when ADHD and other disorders
Target Goals for lndividualized Education
occur together. One of the strengths of the ADHDT is its
Programs (tEPs)
ability to differentiate those students with ADHD from
those students with learning disabilities, conduct dis_
The ADHDT is useful for determining specific behavioral
orders, and behavioral problems who do not have ADHD.
strengths and deficits for individual students. Used with
Research completed during the normative process demon-
other assessment data, information derived from the
strated the ability of the ADHDT to differentiate students
ADHDT should assist teachers and others to identify prob_
with ADHD from students with other diagnoses.
lems, set goals, and identify targets for intervention.
Improvement in behaviors described on individual test
items and improvement in ADHDT scores can be used
Assessment of Fersons Referred as goalsfor IEPs. Specific items from the ADHDT can be
for Behavioral problems used as behavior targets for interventions.

Regardless of the diagnosis, if an individual is referred

because of behavioral problems, some type of assessment Data Collection for Research purposes
is necessary to describe the nature of the person,s prob-
Iems. The assessment should identify or diagnose the Research studies require valid and reliable instruments
problem and help the examiner make decisions about for recording data. Data produced by the instruments
what treatments should be attempted. The ADHDT pro_ need to be objective and describe the behavior being
vides information about three types of behavioral prob_ studied in a precise and meaningful way. Because the
lems (i.e., hyperactivity, impulsivity, and inattention) that ADHDT uses standard scores, it provides useful data
are meaningful regardless of the diagnosis. Even if the for quantifying the severity of ADHD behavior. The for_
person is not diagnosed as having ADHD, information mat of ihe ADHDT permits accurate measurement by
will
be provided that will be useful in the diagnosis and treat_ teachers, parents, and others in and out of school and
ment of the person's behavioral problems. other environments. The psychometric properties of the
ADHDT, particularly reliability and validity ratings, are
strong, making it an excellent instrument for researchers.

Documenting Progress of probtem Behaviors

The importance of documenting progress is well recog_

Summary
nized. Documentation not only is important for account_
ability purposes, but is essential for program planning, The Attention-Deficit/Hyperactiuity Disorder lesf is a
decision making, and educational placement. The ADHDT
norm-referenced test designed to evaluate the behavior
results permit people to document behavioral progress
of persons with ADHD or persons with behavioral prob_
made by individuals. Because the test is completed by
lems. Based on the most current and relevant definitions
caregivers, it may be used as frequently as necessary to of ADHD, it is useful for evaluating behavioral problems
evaluate a person's performance. The ADHDT is well and contributing valuable information toward the diag_
suited for annual student evaluation in special education
nosis of ADHD. Easily completed by parents and profei^
and is particularly useful for accumulating data for deci_
sionals, the ADHDT provides information about problems
sions about extended school year instruction.
of hyperactivity, impulsivity, and inattention.
 Administration and Scoring

This chapter presents instructions for administering and proctor the administration of the ADHDT, but the profes-
scoring the ADHDT. The first two sections describe gen- sional examiner who will interpret and use the ADHDT
eral administration procedures and specific administra- results must be experienced.
tion procedures. The final section describes procedures
for computing raw scores and converting raw scores to
percentile ranks and standard scores. Rater Qualifications

L
The term rater ref.ers to the individual who answers the
questions contained on the ADHDT subtests, usually the
General Administration Procedures subject's teacher, teacher's aid, or parent. The questions
in Section VI, Key Questions, of the Summary,/Response
This section presents general administration procedures Form will be answered by the subject's parents or other
that describe the qualifications required of the people who primary caregivers.
administer the ADHDT (i.e., the examiner), the qualifica- Although no special training is required to administer
tions of the people who do the the actual ratings (i.e., the or score the ADHDT, examiners should make sure that
raters), and the amount of time required for testing. the rater knows how to respond to the items on the sub-
tests. The rater should read the items on the response
Examiner Qualifications form at least twice, think about the behavior described,
and rate how serious the problem is for the individual.
Throughout this manual, the term examiner refers to the
professional who is responsible for administering the
ADHDT, scoring the subtests, and interpreting the results.
Testing Time
This person must know all aspects of the examiner's
The ADHDT is not a timed test. Raters may set their own
manual, especially the test's conceptual framework, the
pace for completing the items. The three subtests can be
statistical characteristics of the test, and the psycho-
metric principles governing norm-referenced assessment completed in a single session or they can be completed
in general. Obviously, he or she must be proficient in one at a time. Most raters will complete all the subtests
administering and scoring the instrument and have a solid
in a single 5- to 10-minute session.
working knowledge of the guidelines for interpreting nor-
mative test data.
The ADHDT can be given by parents, classroom Specific Administration Procedures
teachers, educational diagnosticians, psychological associ-
ates, psychologists, and others who are familiar with the The ADHDT has three subtests: the Hyperactivity Sub-
test. A parent, classroom teacher, or teacher's aid may test, the Impulsivity Subtest, and the Inattention Subtest.
Each subtest is an independent measure and may be com- Computing Raw Scores
pieted in any order. Every effort should be made to
complete all three subtests. A total raw score is computed for each of the three sub-
The raters begin by reading through the ADHDT tests of the ADHDT. The total is achieved simply by add-
Summary/Response Form and then completing all the ing the raw scores for each item. This figure is written
items of which they are absolutely certain. In most cases, in the box at the bottom of that particular subtest. Fig-
raters will be certain and confident of how to responcl ure 2.1 demonstrates the correct scoring procedure. The
on each item and will move quickly from item to item, raw score for each subtest is also recorded in Section II
completing the entire test. Infrequently, raters may be on the ADHDT Summary/Response Form.
uncertain or question how to rate a behavior. In these
cases, the examiner shoulci delay the administration of
the ADHDT until the examiner can observe the subject Converting Raw Scores to Percentile Ranks,
and gain a more accurate rating of the behavior. Standard Scores, and the ADHD Quotient
Each rater who completes the ADHDT will need a
copy of the Summary/Response Form and a pen or pencil. Raw scores from the ADHDT subtests can be converted
The rater should write the name oi the subject who is to percentile ranks and to derived standard scores with
being rated and the date of the rating on the front page a mean of 10 and a standard deviation of 3. The sum of
of the Summary,/Response Form. General directions for the subtest standard scores can then be converted into
completing the subtests are printed on page 2 directly the ADHD Quotient and its percentile. Procedures for
above the Hyperactivity Subtest. They are reproduced obtaining these scores are described next.
here.
Determining Percentile Ranks and Standard Scores
DIRECTIONS: Please indicate which of the following for the Three Sutrtests. The raw score for a particu-
behaviors,/characteristics are a problem for this individ- lar subtest can be converted into a percentile rank (%)
ual. Mark or circle 0 if the behavior is not o problem or or standard score (SS) by means of Tables A and B in the
if you have not had the opportunity to observe the behav- Appendix of this manual. When using the conversion
ior. Mark or circle I if the item refers to a behavior that table, the examiner finds the column for the individual
is a mild problem. Mark or circle 2 if the item refers to subtest being converted. The examiner moves down that
a behavior that is a seL)ere problem for this individual. cclumn until the obtained raw score is located. By mov-
Do not skip any items. ing horizontally to the outside left column, the examiner
can determine the corresponding standard score. By
moving horizontally to the outside right column, the
0 Not a Problem The subject rarely demonstrates this
corresponding percentile rank is found. The standard
problem, and it does not impair his
score and percentile rank should be recorded in Sec-
or her functioning. tion II of the ADHDT Summary,/Response Form. This
I Mild Problem The subject sometimes demonstrates procedure is followed for each subtest completed.
this behavior, and it occasionally
causes problems and impairs his or
Determining the ADHD Quotient and Its Percentile
her functioning. Rank. The sum of the standard scores on the ADHDT
subtests can be converted into a full-scale percentile rank
2 Severe Problem The subject frequently demonstrates and ADHD Quotient. The ADHD Quotient is another type
this behavior, and it usually causes of standard score that has a mean of 100 and a standard
problems and impairs his or her deviation of 15. It represents the overall rating on the
functioning. ADHDT. Table C in the Appendix is used for converting
the sum of the subtest standard scores to the ADHD Quo-
tient. Using this table, the examiner locates the sum of
the subtest standard scores obtained on the ADHDT and
Scoring Procedures the corresponding percentile rank and ADHD Quotient.

Directions are given in this section for scoring each of

the ADHDT subtests. This includes computing raw scores, Plotting the Profile of Scores
converting raw scores to percentiles and standard scores
for the three subtests, and calculating an overall quotient Section IV on the ADHDT Summary/Response Form con-
by combining the standard scores of the subtests. tains a profile that graphically displays the standard scores
 Section Y. Response Form

DIRECTIONS: Please indicate which of the following behaviors/characteristics are a problem for this individ-
ual. Mark or circle 0 if the behavior is not a problem or il yau have not had the opportunity to observe the behavior.
Mark or circle 1 if the item refers to a behavior that is a mr'ld problem" Mark or circle 2 if the item refers to
a behavior that is a severe problem for this individual. Do not skip any items.

Hyperactivity Subtest
E
E -9
C) E .o
o
-g I
o-
o- g
(u o- Io
o I
z o0)
=
1. Loud 0 1
l,
2. Constantly "on-the-go" 0 1 (,
3. Excessive running, jumping, climbing 0 o 2

4. Twisting and wiggling in seat 0 1

@
5. Easily excited 0 1
@
6. Grabs objects 0 c 2

7. Excessive talking
@ 1 2

8. Difficulty remaining seated 0

o 2

9. Constantly manipulating
10. lnability to play quietly
objects
o 1 2

0 1
@
11. Fidgets 0 (1/ 2

12. Restless
@ 1 2

13. Squirms 0 1
@
Hyperactivity Sum I

Figure 2.1. Example of correct scoring procedure for the Hyperactivity Subtest.
for each subtest and the ADHD euotient for the total test. ers, parents, and others to evaluate subjects ages 3
After the raw scores have been converted, the standard through 23 who exhibit behavioral problems. The sub_
scores should be graphed on the profile. This is easily tests can be administered all at once or independently
accomplished and provides a graphic illustration of the of each other. Total administration time is less than
subject's performance on the ADHDT. 5 minutes for most raters. The subtests are easily scored.
and raw scores for each subtest can be converted to per_
centile ranks and standard scores.
Summary
The ADHDT battery includes three subtests: Hyperactiv_
ity, Impulsivity, and Inattention. It can be used by teach_

10
 Interpreting the Results

This chapter will help examiners interpret the scores testing, subject's date of birth, and subject's age at testing
obtained from the ratings. First, specific instructions are are recorded in this section. The examiner should be sure
given for completing the information on the ADHDT that everything recorded is correct.
Summary,/Response Form. The various kinds of scores
generated by the ADHDT and what they mean are dis-
cussed next. The following text examines the use of the Section ll. Score Summary
standard error of measurement when interpreting test
scores. How to use the ADHDT scores to identify persons In this section, the examiner records the raw scores, stan-
with ADHD is then explained, and helpful suggestions are dard scores, and percentiles for the three subtests. The
given for sharing the results of the ADHDT with parents normative table for converting the subtest raw scores to
and other interested professionals. percentiles and standard scores is Table A (for males) or
B (for females) in the Appendix of this manual. Instruc-
tions for converting raw scores to standard scores and
percentile ranks were given in Chapter 2. For conven-
Completing the ience, the standard errors of measurement for each
subtest have been recorded in this section. Informatron
Summary/Response Form about the standard errors of measurement is found in
Chapter 4. The ADHD Quotient and its percentile are also
Demographic information about the subject is recorded
recorded here.
in Section I of the Summary,/Response Form. The scores
from the three subtests are listed in Section II. The Inter-
pretation Guide provided in Section III helps interpret the Section lll. lnterpretation Guide
scores. In Section IV, the ADHDT scores are graphically
displayed. The specific test items are listed in Section V. From information in this section, the examiner can deter-
Directions for rating the items of the three subtests on mine the probability that the subject has ADHD and the
the ADHD'| are also provided in this section. Key ques- severity level of the behavioral disturbances. A range of
tions for accumulating diagnostic information are listeci values for the subtest standard scores and the ADHD Quo-
in Section VI. Section VII provides space for writing rec- tient is provided for estimating the probability of ADHD.
ommendations and comments.

Section lV. Profile of Scores

Section l. ldentifying lnlormation
A graph is provided in this section to plot the results of
The subject's name and address, rater's name and relation- the ADHDT and any other tests that might be available.
ship to the subject, examiner's name and title, date of This is a graphic representation of the test results that

l1
shows the relationship of the subtests' standard scores to scores, three kinds of normative scores are reported: per-
each other and to the overall assessment of the subject's centile ranks, standard scores for the subtests, and a quo-
behavior (i.e., the ADHD Quotient). Low scores are indic- tient representing overall performance on the ADHDT.
ative of persons with little or no behavior problems and
are plotted at the bottom in the unshaded portion of the
profile. High scores are more indicative of ADHD and are Raw Scores
plotted in the shaded portion of the profile. As the size
of the scores increases, the severity of ADHD or behavior Raw scores are the original numerical values associated
problems increases. Also, the probability of the sub- with the subject's test performance. They are the sum of
ject being diagnosed as ADHD increases as well. This is the ratings for the items of each subtest. Although raw
graphically represented by the darkening of the shaded scores are sometimes used in research, they have little
area of the profile. clinical value. The real value of raw scores is that they
can be converted into standard scores. Because raw
scores are the basis from which all standard scores are
Section V. Response Form
calculated, they must be accurate (i.e., they must be free
of errors and recorded correctly). For this reason, exam-
Section V, Response Form, contains the ADHDT subtests.
iners should double-check their raw score results before
Directions for rating each item in the subtests are pro-
calculating standard scores.
vided at the top of page 2. A space is provided at the
bottom of each subtest for recording the sum of the sub-
test ratings.
Percentile Ranks

Section Vl. Key Questions Percentile ranks (%) are commonly used in educational
and psychological evaluations. They indicate the percent-
Section VI, Key Questions, includes a series of questions age of scores in the normative sample that occur above
that provide information necessary for identifying per- or below a given score. If, for example, a score falls at
sons with ADHD. These questions document the evalua- the 63rd percentile, you know that 63% of the subjects
tions that have been done and by whom and any further in the normative sample fall below that score. Because
assessment that needs to be done. They can be used as percentiles provide a ranking from I to 100, their mean-
the basis for interviewins the subject's parents, caregivers, ing is easily understood. Their simplicity makes them use-
guardians, or significant i,rthers about matters that are ful for interpreting test performance to parents and others.
important diagnostically. However, one must understand two things about per-
The purpose of the Key Questions section is to docu- centile ranks.
ment certain facts. First, examiners should determine that First, percentiles apply to the characteristics of the
(a) the ADHD behaviors developed before the age of 7 normative group that was used in compiling the percentile
and have continued throughout the subject's lifetime, ranks for that particular test. In the case of the ADHDT.
(b) the behaviors are seen in various environments (e.g., for example, the normative group was a sample of
at home, at school, and in the community), (c) the per- persons with ADHD between the ages of 3 through 23.
son exhibits the behaviors regardless of who is present. When examiners compare a subject's percentile ranks
(d) the subject's behaviors are not specific to certain on the ADHDT, they are actually comparing the subject
activities or limited to certain situations, and (e) other to a group of subjects known to have ADHD. This is
handicapping conditions are not causing the subject's important information when identifying persons with
behavior. ADHD. A subject who does not have ADHD will score
lower than the normative group (i.e., have a iow per-
centile ranking). And the more characteristics of ADHD
Section Vll. Recommendations and Comments
the subject demonstrates, the higher that person's per-
centile rank will be.
Section VII, Recommendations and Comments, provides
A second consideration is that percentile ranks are
additional room for the examiner to write down any per-
not interval data. Equal differences in percentile ranks
tinent information regarding the subject.
do not represent equal differences in the attributes or
behaviors being measured. The distance between two
percentile ranks becomes much greater as those ranks
lnterpreting Test Scores are more distant from the mean. For example, the differ-
ence between the percentile ranks of 5 and 10 (or between
This section of text familiarizes examiners with the differ- 90 and 95) represents a much greater performance differ-
ent scores that accompany the ADHDT. In addition to raw ence than that between percentile ranks of 50 and 55.

t2
 Standard Scores way that an examiner can control statistically for some
of the error that may enter into the test situation. The
Standard scores are more useful than percentiles; the SEgy Ior each subtest standard score is reported in Chap-
ADHD Quotient is the most useful score because it pro- ter 4. The SEy for all subtests is about I point, and the
vides an overall rating of the subject's behavior. These SE74 for the ADHD Quotient is 3.
scores are calculated directll'from the raw score distribu- Examiners know that a test score is only an esti-
tion. Thev are then normalized into a standard score distri- mate of the subject's test performance. By adding and
bution rlith a predetermined mean and standard deviation. subtracting the SE14 to and from an individual's score, the
Standard scores are expressed as standard deviation units examiner can determine the upper and lower limits,
:o indicate a score's distance from the average perfor- respectively, of the range within which the subject's true
mance of the normative sample. For the ADHDT subtests, test score is likely to be found. Probability estimates are
the mean of the standard score distribution has been set applied. For instance, 68% of the time, a subject's true
at 10 and the standard deviation at 3. The larger the stan- test score is likely to fall within a range that is plus or
dard score. the more severe the ADHD behavior that it minus 1 SE74 Irom the obtained test score. By extending
represents. Standard scores are computed for all of the the range to plus or minus 2 SE7a, the examiner increases
ADHDT subtests. Depending on the gender of the sub- the conf idence interval to g5%. A range of plus or minus
ject, Table A or Table B in the Appendix is used to con- 3 SETa pushes the confidence interval up to g9%.
vert standard scores for all of the subtests. When examiners report ADHDT scores or interpret
Standard scores are considerably more valuable than them to others, they should consider each score's SE14.
percentile ranks, even though they both are computed In this way, subsequent users of the test score will know
directly from the raw score distribution. Standard scores the range within which the true score probably lies. The
are interval data and can be compared directly with other SEy of each subtest and ADHD Quotient is displayed in
standard scores with the same mean and standard devia- Section II of the ADHDT Summary/Response Form.
tion. This allows examiners to compare a subject's perfor-
mance with the performance of the normative group or to
compare the subject's performance on one subtest of the
ADHDT with another subtest of the ADHDT. Standard
scores can also be added, subtracted, and otherwise manip-
Using ADHDT Scores to ldentify
ulated statistically, which makes them ideal for research. Persons with ADHD
The ADHDT standard scores are particularly helpful
in profiling and comparing the subject's performance The previous sections of this chapter were concerned
associated with the three subtests that make up the instru- with the technical aspects of the scores produced by
ment. Usinq the standard scores, the examiner can iden- the ADHDT. Most examiners will probably want to use
tify the subject's relative strengths and weaknesses both the scores to diagnose ADHD. This section provides infor-
in comparison to the normative group and in comparison mation for judging the ADHDT standard scores relative
to his or her scores on the various subtests of the ADHDT. to the diagnosis of ADHD. Before a discussion of the inter-
pretation of the scores, however, a note of caution is
provided about using ADHDT scores or any scores for
The ADHD Quotient making important diagnostic decisions about ADHD. The
use of norm-referenced tests is discussed next, followed
The ADHD Quotient generated by the ADHDT is another by guidelines for interpreting subtest standard scores and
kind of standard score. The ADHD Quotient, which has the ADHD Quotient in identifying persons with ADHD.
a mean of 100 and a standard deviation of 15. is the most
reliable of all the scores generated on the ADHDT. It is
a
computed by calculating the sum of the standard scores
of the subtests and then converting that sum into a quo-
A Note of Caution
tient. The table for obtaining the ADHD Quotient can be
Remember that test scores alone do not diagnose any-
I found in Table C of the Appendix.
thing. They simply provide data about some characteris-
i
tics thought to be important in classifying something.
Therefore, the results of the ADHDT should never be the
Using the Standard Error single source of information used to diagnose ADHD. The
of Measurement ADHDT does provide important data about ADHD behav-
iors and norms for comparing a subject's scores with those
The standard error of measurement (SEl is an impor- of a national sample of persons known to have ADHD.
tant statistic for examiners to use when interpreting test Tests, like observations, interviews, or even a simple
scores. Attention to the meaning of this statistic is one review of records, yield valuable information that con"
I

i
I 13
I

I
I
I
I
t
 tributes to a diagnosis. But in the end, a competent exam- For this reason it provides the best prediction of ADHD
iner must decide whether or not an individual has ADHD. When comparing a subject's scores to scores on Table 3.1
the examiner should first look at the subject's ADHD
Quotient.
The Use of Norm-Referenced Tests If the subject's ADHD Quotient is 90 or above, the
person probably has ADHD. Standard scores of 8 through
In writing this manual, the author made a concerted effort 12 for the subtests or ADHD Quotients of 90 through 11{-r
to present evidence that the ADHDT is well built psycho- are within the average range for subjects with ADHD in
metrically. High standards were set for the ADHDT, and the normative sample. Approximately 50% of the subjects
for the most part, these standards were met. Its scores with ADHD scored in this range. Standard scores above
wili help a knowledgeable examiner document informa- 12 or ADHD Quotients equal to or greater than 111 are
tion about the subject that wiil help with diagnosis. One highly indicative of ADHD. The probabiliry of non-ADHD
of the ways the ADHDT helps examiners is by providing subjects receiving scores this high is very unlikely.
norms on persons with ADHD. Subtest standard scores of 6 or 7 or an ADHD euo-
A norm-referenced test can be used to compare a per- tient of 80 through 89 are beiow average for subjects with
son's test results with a sample of subjects with known ADHD and represent borderline scores in terms of the
characteristics-in this case, persons with ADHD. If the likelihood of ADHD. Persons who receive scores in this
normative sample is representative of the characteristics range may or may not have ADHD. In the normative
or attributes being measured (i.e., the characteristics of study. only 23% of the ADHD subjects scored g9 or lower.
ADHD), a point of reference is established. Given this A person who receives a quotient of 80 through 89 should
norm, one can compare a person's scores to it, ask cer- not be assumed to have ADHD; in such cases, additional
tain questions, and make certain ludgments. By com- evidence from other tests, parent interviews, and direct
paring a subject's ADHDT scores to the scores on the observation of the subject should be gathered to aid in
normative table, the examiner can determine if a subject's the diagnosis.
performance on the ADHDT is below average, average, Significantly low scores are standard scores below 6
or above average in relatir-rn to that of the ADHD sub- on any subtest or an ADHD Quotient below 80. In the
jects in the normative group. normative sample, less than 9% of the subjects with
ADHD scored this low. Ninety-eight percent of the sam-
ple had an ADHD Quotient of 70 or greater. If the ADHD
Using the ADHDT to Determine
Quotient is below 70, the person very probably does not
the Likelihood of ADHD have ADHD.

The question most examiners will probably ask is ,,How

likely is it that the subject has ADHD?" Estimates of like-
lihood that a subject will receive certain scores on the Sharing the Results
ADHDT are displayed in Table 3.1.
The best overall estimate of a subject's behavior is The results of the ADHDT should be shared with the peo-
the total test score, in the case of the ADHDT, the ADHD ple who are legally entitled to receive the inforrnation.
Quotient. This quotient takes into account all the symp- Written reports should be included in the subject's per-
tomatic behaviors of ADHD measured on the ADHDT. manent records. Oral reports should be presented at for-

TABLE 3.1
Guidelines for lnterpreting subtest Standard scores and the ADHD euotient

Subtest ADHD Percentile Percentage of Probability of

Standard Score Quotient Rank Normative Samplea ADHD

17 -"t9 131 + 99+ 2 Very High

15-16 121-130 92-98 7 High
13-14 1't1-120 76-91 't6 Above Average
8-12 90-1 1 0 25-75 s0 Average
6-7 80-89 9-24 16 Below Average
4*5 70-79 2-8 7 Low
1-3 <69 .1-1 2 Very Low
aThe normative
sample is composed entirely of individuals diagnosed with ADHD.

l4
 .,- ll l!

mal meetings (e.g., IEP meetings,-diagnosis and evalua- testing situation or even in the rater. Because of the
tion staffings, or parent conferences). When sharing the possibility for error, interpretation should be judicious.
results from the ADHDT, the examiner should always con- Alternative explanations for test results should be con-
sider the following points. sidered and reported when appropriate.
A thorough understanding of the purposes, content,
and construction of the ADHDT is necessary prior to any
presentation. The test manual should be made available
rvhen presenting results to people who are unfamiliar with 5ummary
the test. The section Uses of the ADHDT in Chapter 1 is
particularly useful as well as the data in the sections on The ADHDT is easy to complete and simple to score and
reliability and validity in Chapter 4. can be used by a variety of raters and examiners. Using
Any report should include the reasons why the sub- the tables in the appendix, raw scores are quickly con-
ject was evaluated and why the ADHDT was selected as verted to percentile ranks and standard scores which are
t
part of the evaluation. In the report, the components of recorded on the Summary/Response Form. Several fea-
the ADHDT should be described, along with the scores tures help with interpretation of the results. The interpre-
and their meanings. The psychometric characteristics of tation guide in Section III of the Summary,/Response Form
the test should also be explained. provides a range of standard score values for estimating
The examiner should discuss the implications of the the probability that the subject has ADHD and the severity
results for diagnosis and treatment and make suggestions of the behavioral problems. Section IV of the Summary,/
for changes in the current program. If necessary, recom- Response Form allows the examiner to plot the results
mendations for further testing may be appropriate. of the ADHDT and provides a graphic display of the test
The examiner should avoid overstating the meaning results showing the relationship of the subtest standard
of the ADHDT scores. The ADHDT has good psycho- scores and ADHD Quotient. Since the ADHDT is norm ref-
metric characteristics, and examiners can use it with con- erenced, the examiner can determine if a subject's scores
fidence. But all tests are subject to error. Some of the error on the ADHDT are below average, average, or above aver-
is inherent in the test and some may be inherent in the age compared with those of persons known to have ADHD.

15

I
.

I
,

Development and
Technical Characteristics

This chapter contains a description of the procedures used selected in a variety of ways. A mailing list of special edu-
to develop the ADHDT, including the data obtained in cation teachers who teach children with ADHD, learn-
establishing the statistical characteristics of the test. Topics ing disabilities, emotional disturbance, mental retardation,
discussed relate to item selection, normative procedures, and physical impairments was purchased from Market
reliability, and validity. Data Retrieval, a company that specializes in mailing lists.
From this list, approximately 5,000 teachers were ran-
domly selected by a computer. These teachers were
mailed a survey and asked if they would be interested
Item Selection in helping to norm the ADHDT on students they taught
who were diagnosed as having ADHD. Approximately
As mentioned previously, the items of the ADHDT corre-
600 teachers volunteered to help. From this group,424
spond to the definition of ADHD found in the DSM-IV.
teachers from various geographic locations were ran-
All the ADHDT items are based on behavioral descrip-
domly selected to participate in the norming and were
tions or examples of ADHD characteristics mentioned in
mailed copies of the ADHDT along with other tests fre-
the DSM-IV. Because the descriptions are behaviorally
quently used with students who have ADHD. Three hun-
specific and replete with examples, the relationship
dred ninety-eight of these teachers returned completed
between the description and the test items is easy to see.
checklists. To ensure geographical representation, an
additional group of teachers in special education and
general education who taught children with ADHD were
Normative Procedures contacted and sent testing materials.
c Parents of ADHD children were contacted in several
In this section, the procedures used to norm the ADHDT ways. A list of chapter coordinators for Children and
are described. The demographic characteristics of the Adults with Attention Deficit Disorders (CH.A.D.D.) was
normative sample are presented, along with a discussion obtained, and 50 of the chapter coordinators were ran-
of the types of normative scores that can be derived from domly selected and contacted by phone or mail and asked
the ADHDT. to participate in the normative study. The chapter coordi-
nators distributed and collected the ADHDT at their meet-
ings. Approximately 25 chapter coordinators returned the
Selecting the Normative Group
tests. Other parents completed the ADHDT at professional
The ADHDT was normed on a sample ol 1,279 children and parent group meetings.
and young adults who had a diagnosis of Attention- In many cases, teachers also rated students who had
Deficit/Hyperactivity Disorder. These subjects were other handicaps, such as learning disabilities, emotional

17
 disturbance, mental retardation, and other disabilities. the ADHD population. However, the following assump-
They also rated students who were not handicapped. tions about the ADHD population seem reasonable. The
Many parents also rated their other children who did not ADHD population is geographically distributed across the
have ADHD. As a result, the total sample of subjects in United States like other normally distributed characteris_
the standardization group was 2,696. Of this number i,27g tics of the total population. A review of the ADHD litera_
were previously diagnosed with ADHD, 976 had other ture revealed no findings that contradicted the assumption
handicapping conditions, and 541 were nonhandicapped. that ADHD is normally distributed relative to race or eth_
The data from the non-ADHD subjects were used in the nicity. One would expect, therefore, that the racial and
validity studies but were not included in the norms ethnic characteristics of persons with ADHD would be
reported in the next section of this chapter. similar to the U.S. census data.
Given what is known about ADHD and what can be
assumed, the sample used for norming the ADHDT
Demographic Characteristics of the appears representative of the ADHD population. On the
Normative Group demographic characteristics reported for the 1gg0 U.S.
Census, the characteristics of the normative sample
Subjects in the normative group represent the diversity approximate the census data statistics in race, ethnicity,
found among people with ADHD. persons with ADHD dis_ and geographical area.
play a wide range of symptoms with varying degrees of Normative samples should have at least 75 to 100 sub-
symptoms both within and between subjects. As a result jects at every age level and at least 750 to 1.000
of the variety of data collection methods, the normative in the
total sample (Hammill et al., 1992). The normative sam_
group was composed of subjects with the diverse char_ ple for the ADHDT is adequate for total sample size.
acteristics typical of ADHD. We failed to reach 75 subjects at some ages but were
Subjects in the normative sample covered a,*,ide geo_ not overly concerned because the ADHDT scores are only
graphical range. They came from 47 states and Canada.
minimally related to age. Compared with other tests
The ages of the sample also covered a wide range, from
designed to identify persons with ADHD, the ADHDT
3 through 23 years. This diversity adds to the strength represents a major improvement both in size and repre_
of the subtests and prcvides comparisons for a wide sentativeness of the sample. It is one of the few assess_
variety of relevant demographic characteristics. ment instruments available that was normed entirely on
According to Hammill, Brown, and Bryant (1992)test persons diagnosed with ADHD.
developers must demonstrate that the normative sample
In the normative study, more teachers (1y' 63g) com_
is representative of the specified group of people with =
pleted the ADHDT than parents (1{ 3g1). This ratio
whom the test developer intends the instrument to be =
seems appropriate because teachers and other profession_
used, in this case persons with behavioral disturbances.
alswill be the principal raters of the ADHDT. The sub
Evidence of representativeness requires that important
stantial number of respondents in the normative study
demographic characteristics of the normative sample who were parents of children with ADHD adds credibility
approximate those of the reference population as a whole.
to the use of the ADHDT by parents.
In the case of the ADHDT, for the test to be representa_
tive, it must conform to the characteristics of the ADHD
population, not the normal population.
Normative Scores
Table 4.i presents data on the demographic char,
acteristics of the normative sample. With a prevalence The subtests of the ADHDT are all norm referenced, based
rate of 3%, one would think that there would be a large
on the results from the subjects with ADHD in the stan_
amount of information about the demographic charac- dardization sample. The norms are reported in terms of
teristics of persons with ADHD, but unfortunately, this
standard scores and percentile ranks. The standard scores
is not the case. Little is known about the demographic are normally distributed and allow an examiner to make
characteristics of people with ADHD. And what is gener_
comparisons between an individual,s subtest scores and
ally accepted is usually an estimate. For example, the only the subtest scores of a nationally representative group of
demographic characteristic of ADHD that is ionsistently
subjects who were known to have ADHD. These scores
reported in the research is that the number of boys are described in this section. However, their interpreta_
diagnosed with ADHD is consistently higher than girls.
tion is explained in greater detail in Chapter 3.
According to Nussbaum and Bigler (1gg0), the exact ratio
of boys to girls has been the subject of some controversy,
Standard Scores. Standard score norms are expressed
but they report that the generally accepted figure is 6:1.
as standard deviation units that designate a score,s dis-
_ Other than the higher incidence of males to females,
little is known about the demographic characteristics of
tance from the average performance of the normative
sample by applying a predetermined mean and standard

18
 TABLE 4.1
Demographic Characteristics of the ADHD Subjects in the Normative Sample

% of School-age
Characteristic Sample Size 7o of Sample' Population"

Total Sample 1,279

lt/l a les 1,024 80 s1 (82)b
Females 255 20 49 (18)b
Race
I
t Eth
White
Black
Other
ncity
i
1,042
167
70
82
13
05
80
16
04

Native American 24 02 0'l

H ispan ic 115 09 'l'l
Asian 50 05 03
Af rican-American 164 13 14
Other; Everyone Else 916 72 71

Geographic Area
Northeast z'.t4 17 19
North Central 264 21 24
South 485 38 36
West 316 25 21

Urban/Rural
U rba n /Sub u rban 680 53 77
Rural 321 25 23
Unknown 278 22

Socioeconomic Status
Free Lunch 289 23
Reduced Lunch 90 07
Pays for Lunch 734 57
Unknown 166 13

Raters
Teachers 638 50
Pa rents 391 3I
Psych iatrists/Diag nosticia ns 104 08
Spouse 13 01
Other 133 't0

Taking Medication
Yes 7s2 s9
No 273 21

t
I Age
Unknown 254 20

I <3
4
8
20
<1
2
5 37 3
6 63 5
7 "t'17 9
8 15s 12
9 160 13
't0 133 10
11 't 03 8

19
ir TABLE 4.1. Continued

% of School-age
Characteristic Sample Size % of Sample" Populationa

Age (cont.)
12 123 10
13 105 8
14 87 7
't5 65 5
16 50 4
17 26 2
18 18 <1
19 6 <1
20 3 <1
21 6 <1
22 2 <1
23 4 <1
aPercentages are rounded
off to the next whole number; decimals have been ornitted. bPercentages estimated in ADHD populat
o.
(Nussbaum & Bigler, 1990).

deviation. For example, the mean and standard deviation for different ages and gender. As a result, standard scores
for z-scores are 0 and l, respectively; for l-scores, they were computed for the normative sample of subject-.
are 50 and 10; and so on. For the ADHDT subtests, the based on age and gender. These norms are located in the
mean has been set at l0 and the standard deviation at 3. Appendix of the manual.
Standard scores for the ADHDT subtests are derived
directly from a cumulative frequency table containing the
raw scores received by the normative sample. When nor- The ADHD Quotient. The ADHD Quotient is anotirer
mative tables are constructed, the raw scores are trans- type of normalized standard score. This quotient has a
formed into the desired derived distribution (i.e., into a mean of 100 and a standard deviation of 15 and represents
distribution with a mean of l0 and a standard deviation the examiner's overall assessment of the characteristics
of 3). Raw score means and standard deviations were of ADHD manifested by the subject. The ADHD Quotienr
computed for each age and gender. Partial correlation of is derived by summing the standard scores for all ihe
raw score totals for each subtest controlling for age and subtests of the ADHDT. This sum can be converted to a
gender and multivariate analysis of variance of ADHDT quotient using Table C in the Appendix.
subtest scores by age and gender revealed that there were
negligible but significant differences between subjects Percentile Ranks. Percentile ranks are reported for
based on age and sex. The correlation between hyper- each of ihe ADHDT subtests. These are useful and easilr
activity and age was -.19 (p < .01), between impulsivity understood scores that are frequently reported in educa-
and age -.07 (p < .05), and the total raw score and tional and psychological evaluations. They enjoy popular
age -.09 (p < .01). The correlation between age and use because their meaning is quickly grasped by psycho-
inattention was.05 and not significant. These findings indi- metrically naive individuals, as well as by teachers, par-
cate that there is a small relationship between ADHD and ents, and various other professionals. The only drawback
age, and as persons with ADHD increase in age, the inten- to percentile ranks is that they are not interval data.
sity of their hyperactivity and impulsivity decreases Because of this, unequal distances exist between score
slightly. points. For example, the distance between the percentile
To examine the relationship between the ADHDT raw ranks of 15 and 20 is not the same as the distance between
scores and gender, /-tests for independent samples were the percentile ranks of 85 and 90. For this reason, per-
performed on the normative sample. Results indicated centile ranks cannot be averaged or otherwise oper"ated
that males score slightly higher than females on all sub- on arithmetically"
tests as well as the total score of the ADHDT. These scores Percentile ranks, like standard scores, are deriveci
were statistically significant (p < .05). directly from the raw score distribution of a test. Thel'
Given the significant differences between age and indicate the percentage of scores in the normative group
gender on the raw scores, separate norms were required that are above or below the score in question. Examiner-q

20
 may use Table A (for males) or Table B (for females) to late with the subtest to which they belong or to the total
convert raw scores to percentile ranks for the ADHDT test score is the degree to which the test systematically
subtests. introduces error in the construct being measured. Because
the purpose of a test is to measure a certain trait, ability,
or content, the more items relate to each other, the
smaller the error in the test will be. If the items are
unrelated to each other, they are most likely measuring
Reliability different qualities, and the amount of test error due to
content sampling will be great.
A good test is reliable. That is, it measures consistently.
The internal consistency of the items on the ADHDT
Tests that have adequate reliability will yield more or less
was investigated using Cronbach's coefficient alpha (1951).
ti the same scores across different periods of time and across
different examiners. When unreliable tests are used,
This statistical procedure is one of the most rigorous for
determining reliability and is commonly reported in test
examiners get inconsistent results. Obviously, greater con-
manuals. Coefficient alphas were computed for all of the
fidence is given to the results of tests that are consistent
subtests of the ADHDT utilizing 754 ol the ADHD sub-
in their measurements. Tests that have good reliability jects from the normatization sample. The analyses were
have very little error associated with their scores. Because
performed separately for age and gender. The resulting
of this, the study of a test's reliability focuses on estimat-
coefficients and their corresponding standard errors of
ing the amount of error associated with its scores.
measurement are reported in Table 4.2.
According to Anastasi (1988), two important sources
As seen in Table 4.2,the ADHDT demonstrates strong
of error can be attributed to content sampling and time
estimates of internal consistency. All correlations are
sampling. Content sampling error arises from the content
above .90. These findings suggest that the items within
of the test itself and can be measured through internal
the subtests are homogeneous and that the overall mag-
consistency reliability studies. Time sampling error (the
nitude of the correlations indicate it has strong reliability.
difference between scores on the same test given at differ-
Because the ADHDT will be completed by a variety
ent times) is measured by studies of stability reliability.
of persons, an additional study was conducted to examine
The ADHDT's reliability was carefully documented so that
the effects of different types ol raters on the test's inter-
examiners could have faith in the results obtained. In the
nal consistency. Data from the normative sample were
following text, the studies of internal consistency, stan-
used to calculate coefficient alphas by the rater's role:
dard error of measurement, and stability reliabilitv are
teacher, parent, psychologist/diagnostician, or spouse.
reported.
Table 4.3 presents the results of this study.
As can be seen, the alphas are quite large. The stan-
dard errors of measurement for the standard scores are
lnternal Consistency Reliability also reported in Table 4.3, and they are appropriately low.
The results of this study demonstrate that the ADHDT has
Internal consistency reliability is concerned lvith the lva.v strong internal consistency when used by a variety of
items of a test contribute in a systematic wav to the sub- raters. It is a reliable instrument, and examiners can have
test score or total test score. In tests that have good confidence in the subtests when making decisions or inter.
internal consistency, all items correlate positivell, and preting the results from the ADHDT. All of the subtests
moderately with their respective subtest score and the are sufficiently reliable for contributing to important diag-
total test score. The degree to which items fail to corre- nostic decisions.

t
TABLE 4.2
Cronbach's Alpha Reliability Coefficients and Standard Errors of Measurement for the ADHDT

Males Females
ADHDT Age
Subtests Group Alpha 5E rvt n Alpha SEut

3-7 't22 .94 .73 30 .91 .90

Hyperactivity 8-23 632 .93 70 143 .93 .79
lmpulsivity 3-23 754 .92 .85 't73 .93 .79
lnattention 3-23 754 .93 .79 '173 .92 .85
ADHD Quotient 3-23 754 .97 2.60 173 .96 3.00

21
 TABLE 4.3
Cronbach's Alpha Reliability Coefficients and Standard Error of Measurement for the ADHDT by Type of Rater

Psychologist/
Parent Diagnostician

(n : a53) (n = 274) : (n 65) (n = 11)

Alpha sEut Af pha SEu Alpha

Hyperactivity .94 .73 .92 .85 .94 .73 .94 .73

lmpuisivity .93 "79 .89 .99 .92 .85 .94 .73
lnattention .93 .79 .93 .79 .92 .85 .97 .52
ADHD Quotient .97 2.60 .96 3.0 .97 3.0 .98 2.12

Standard Error of Measurement amount of time to pass (usually 2 weeks or less), and test-
ing the same group again. The results of the scores of the
Examiners know that a test score is not exact; it is an esti- two testings are correlated to determine the amount of
mate. The standard error of measurement (SEy) is a sta- stability reliability in the test. If a test fails to yield the
tistical indicator of the error variance associated with a same or similar scores for a subject on different occasions,
specific test score. It is directly proportional to the instru- something must be wrong with the test.
ment's reliability. By adding the standard error of mea- Two test-retest studies were done. In the first study,
surement to (or subtracting it from) an individual's the ADHDT was administered to a sample of 21 subjects
obtained test score, an examiner can determine the upper (13 males, 8 females; mean age 10-4). Thirteen of the
limits (or lower limits) of the range within which the true subjects were diagnosed as having ADHD, 4 were non-
test score is most likely to be found. For instance, 68% handicapped, and I student was diagnosed as learning
of the time a subject's true test score is likely to be found disabled. The subjects were rated on the ADHDT by their
within a range of plus or minus 1 standard error of mea- teachers at 2-week intervals. Raw scores for the two test-
surement unit from the obtained test score. A proba- ings were converted into subtest standard scores and
bility of 95% exists that the true score is between minus 2 overall quotients. The values were then correlated. The
and plus 2 standard errors of measurement. By extend- results reported in Table 4.4 provide evidence of the sta-
ing the range to plus 3 or minus 3 standard errors of mea- bility of the ADHDT when used with students in school.
surement, an examiner can extend the confidence level All test-retest correlation coefficients were beyond the
to 99%. .01 level of significance and of sufficient magnitude to
As a test's reliability increases, lts SEpl decreases. suggest that the ADHDT has adequate test-retest reli-
This functional relationship is apparent in the formula ability for use as an instrument for identifying persons
SEla= SD^,[l -, (SD = standarddeviation; r = re]i- with ADHD.
ability). Table 4.2 reports the SEy for the quotients and A second study was performed utilizing the ratings
subtests, calculated by inserting the coefficients alpha given by college students enrolled in teacher education
into the formula. The standard errors of measurement in classes in a department of special education. Twenty-one
Tables 4.2 and 4.3 are reported in standard score points. undergraduate students majoring in special education
The SEy lor the subtests of the ADHDT is approximately completed the ADHDT on students they were working
I standard score point, and the SEplIor the ADHD Quo- with as part of their training. Fifteen of the students rated
tient is 3. Because the ADHDT's reliability coefficients are were males, and 6 were females. Twelve were diagnosed
strong and the standard errors of measurement are small, as having ADHD, 4 as emotionally disturbed, and 5 as
one can conclude that the ADHDT demonstrates accept- learning disabled. The college students rated their sub-
able internal consistency reliability. jects at l-week intervals. Results of this study are reported
in Table 4.5.
This study, like the one previous one, resulted in sig-
Stability Reliability nificantly strong correlations attesting to the ADHDT's
stability over time. Given these results, examiners can
Stability reliability is most often determined by adminis- have confidence in the reliability of the scores from
tering the same test to a group ol subjects, allowing an the ADHDT.

22
 TABLE 4.4
Test-Retest Reliability of Teachers Utilizing the ADHDT at 2-Week lntervals

ADHDT Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

Hyperactivity .89*
lmpulsivity .91*
lnattention .85*
ADHD Quotient .92*

*p .01 .

TABLE 4.5
Test-Retest Reliability of College Students" Utilizing the ADHDT at 1-Week Intervals

ADHDT Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

Hyperactivity .92*
lmpulsivity .93*
lnattention .85*
ADHD Quotient .94*

uN
= 21.
*p < .01.

Validity logical and empirical methods to determine whether the

items of a test are representative examples of the con-
Most discussions of validity address the three major cate- struct that is being measured. Content validity must be
gories of content validity, criterion-related validitl'. and built into a test during its construction.
construct validity. In the following section, these three The process used to identify items was discussed
types of validity are discussed, and studies are described earlier in Chapter I and in this chapter. To review, items
to demonstrate the validity of the;\DHDT as a diagnostic for the ADHDT were derived from the DSM-IV descrip-
instrument. tion of ADHD. Behavioral characteristics of ADHD were
Validity refers to the extent that a test measures rvhat described in checklist fashion and organized into the three
it sets out to measure. Validity research provides the user categories of impairments specified in the DSM-IV (i.e.,
with evidence that the test in question measures what it hyperactivity, impulsivity, and inattention). Thirty-six
purports to measure, that it can be put to work for its items were written that seemed to capture the essence
stated or intended purposes, and that useful inferences of ADHD. Thirteen of the items described hyperactive
can be drawn from its results. The data presented in this behaviors, 10 items described impulsive behaviors, and
section demonstrate that the ADHDT is a valid instrument 13 items described disturbances of inattention. These
for identifying persons with ADHD from other behavioral items were evaluated to determine the discriminating
disorders. power of the test. Item discrimination and item difficulty
Evidence of a test's validity accrues over time. At this analyses were then conducted to confirm the validity of
point, the ADHDT is new and has not been in use long the test items.
enough to accrue a large body of research to attest to its Item discrimination coefficients are a specific use of
validity. However, sufficient preliminary evidence will be the correlation coefficient in which the items of a test are
presented in this manual for examiners to feel confident correlated to the total score. This procedure, although
about the ADHDT. It is hoped that independent research- simple in concept and technique, is actually quite powerful
ers will soon replicate some of the studies reported here and sophisticated. Item discrimination is essential in the
or will produce additional evidence of the scale's validity. construct validation of a scale and also helps to homoge-
nize the scale's items, ensuring its ultimate internal con-
Content Validity sistency reliability.
Item discrimination coefficients indicate the extent to
As implied by its name, content validity is concerned with which a given item discriminates subjects who exhibit
the content of the test. Content validity research uses both more of a particular trait from subjects who exhibit less

23
 of that trait. On a mathematics test, for example, a dis- magnitude. The .35 minimum is large enough to ens ...
criminating item would be one that students who earned that each item is making a meaningful contribution tc, : -
high scores on the test answered correctly and students subtest.
with low scores answered incorrectly. That is, the good A confirmatory item analysis was done on 935 ca.-'
students tended to get the item right, and the poor stu- from the normative sample (see Table 4.6), These ca:.,
dents tended to get it wrong. Obviously, an item does not were selected because they had complete data; that ,
discriminate if all the students get it correct or all of them all 36 items of the ADHDT were completed. In most case,
miss it, because both good students and poor students item analyses are performed for each age interr..
scored the same on that item. Test items that have good Because little relationship exists between age and scc,t.=.
item-to-total correlations discriminate, and, conversely, on the ADHDT subtests" item analyses were not necr:-
items that have Iow item-to-total correlations do not dis- sary at each age. The following median coefficients r,' er=
criminate. obtained: Hyperactivity, .Tl; Impulsivity, .72; Inattent1,,.
On tests like the ADHDT, however, items are not .69. The median coefficients for each subtest w€r€ .:
scored as correct or incorrect. Instead the ADHDT is con- statistically significant (p < .01). In addition, they rter.
cerned with how well the item correlates with the total well beyond the minimum criteria for magnitude. One , ;..
amount (domain) measured by the subtest. Discriminat- conclude from these data that the item d.iscriminat -
ing items describe a trait that is observed (present) in per- coefficients of the ADHDT are acceptable
sons whose total scores are large for that domain and
small for persons low in that domain. For example, in the
activity domain, persons who are hyperactive should Criterion-Related Va lidity
receive high scores on items 1 through 13. Conversely,
persons who are normally active should receive lower Criterion-related validity is concerned with the relat. -'
scores on these items. ship of test scores to some criterion measure, such .,
Two item discrimination criteria were used to select a test, diagnostic classification, or some type of :."'
items for the ADHDT. Using the criteria established by formance. Two types of criterion-related validitl' .'=
Hammill, Brown, and Bryant in A Consumer's Guide to described in the measurement literature: predictive ,' -
Tests in Print, Second Edition (1992), the item discrimina- concurrent. In the validation of the ADHDT, eri.--
tion coefficients should be statistically significant at or sive studies were conducted to establish the CoflCUrr:-'
beyond the .05 level and should reach or exceed .35 in criterion-related validity. In the following text, evide:. .

TABLE 4.6
Item-to-Total Correlations for the ADHDT from the Normative Sample'

ADHDT Subtests

Hyperactivity lmpulsivity lnattention

Item No. Item No. Item No.

1 .51 14 .61 24 .66

2 .75 15 .55 25 .69
3 .71 16 .77 26 .61
4 .77 17 .81 27 .67
5 .69 18 .71 28 .57
5 .67 19 .71 29 .73
7 .62 20 .78 30 .67
8 .78 21 .77 3'l .75
9 .59 22 .72 32 .70
10 .70 23 .55 33 .78
11 .78 34 .79
't2 .76 35 .76
13 .77 36 .56

Median .71 .59

an
= 935.

24

E
 about concurrent validity is provided by establishing ADHDT to correlate significantly and strongly with the
the relationship of the ADHDT to tests that are fre- subtests from these various instruments.
quently used in the assessment of persons with ADHD.
Other studies of the concurrent criterion-related validity Correlation of the ADHDT with the CTRS. The stan-
of the ADHDT were done by discriminative analyses dard scores on the ADHDT were correlated with scores
of ADHDT scores with subjects from varying diagnos- on the CTRS-2S and CTRS-39. The CTRS-28 and the
tic groups. CTRS-39 are similar in purpose to the ADHDT. They are
Studies of concurrent validity examine the relation- used for evaluating subjects with behavioral disorders for
ship of test scores to some criterion measure obtained at screening purposes. The CTRS-28 contains 28 items, and
the same time. The criterion measure is usually another the CTRS-39 contains 39 items. Both use a 1 (not all) to
test related to the domain being testcd. During the nor- 4 {uery much) scale to rate the severity of problem
t matization process, data were collected on a variety of behavior. The CTRS*28 produces four subscores: Conduct
rating scales along with the ADHDT. Seven tests were Problem, Hyperactivity, Inattentive-Passive, and Hyper-
chosen for correlating with the ADHDT: the Conners' activity Index.
Teacher Roting Scales-29 (CTRS-28) and the Conners' The correlations between the ADHDT and the
Teocher Ratinq Scoles-3g (CTRS-39) (Conners. 1990); the CTRS-28 were computed for two groups of students with
Attention Deficit Disorders Eualuation Scale-School Ver- ADHD; 10 of the students came from Alabama, and the
sion (ADDES-SV)(McCarney, 1989); the ADD-H Compre- other group of 20 students was from Arizona. The total
hensiue Teacher's Rating Scale-Second Edition (ACTeRS) sample inclucted 24 males and 6 females, ranging in age
(Ullmann, Sleator, & Sprague, 1991); the Behauior Eual- from 5 to 13. All but one of the correlations were signifi-
uation Scale-Second Edition (BES-2) (McCarner. & Leigh, cant and exceeded the .35 criterion. The results of these
1990); the Behauior Rating Profile-Second Editron (BRP-2) correlations demonstrate that there is a moderate-
(Brown & Hammill, 1990); and the Reursed Behauior Prob- to-strong relationship between the ADHDT and the
lems Checklrsr (RBPC) (Quay & Peterson. 1987) These CTRS-28. These data are reported in Table 4.7.
instruments are frequently used for screening and assess- A sample of 65 students from Minnesota (n = 20),
ment of students with behavioral problems. and thev are Colorado (n = 25), and Maryland (n = 20) were given
commonly used for evaluating students suspected of hav- both the ADHDT and the CTRS-39. Forty-six of these sub-
ing ADHD. In all the studies correlating the .\DHDT to jects were male and 19 were female, their ages ranging
these criterion measures, standard scores were correlated from 3 to 23. The correlations obtained were significant
to control for age, and the coefficients were corrected for and strong. The only low correlations are those related
attenuation in the criterion variable only. to the Anxious subtest of the CTRS-39, but one would
One would hypothesize that these tests should corre- not expect much of a relationship between the charac-
late strongly with the ADHDT. Because some of the sub- teristics measured by this subtest and the characteris-
tests of these criterion instruments do not assess in the tics measured on the ADHDT. Of particular note is the
domains tested by the ADHDT, not all of the subtests oi exceptionally strong ratings of the Hyperactivity subtest
any one instrument can be expected to correlate rvith the of the CTRS-39 with the three subtests and ADHD Quo-
ADHDT. For example, one cannot reasonabl-v expect the tient of the ADHDT. Overall, of the 28 pairings, 19 of them
ADHDT to correlate with the subtest Anxious-Passive on are above .60 and only 5 are below .40. These data are
the CTRS-39. This subtest is measuring characteristics reported in Table 4.8.
from domains different from those measured b-"-' the The relationship of the ADHDT to the Conners' Scales
ADHDT. However, in general, one should expect the is encouraging. The correlation coefficients of the ADHDT

TABTE 4.7
Correlation of ADHDT with Conners'Teacher Rating Scales-28 (n : 30)
I
ADHDT Values

CTRS-28 Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

Conduct Problem .49** .52** "42* .53**

Hyperactivity .78** .56** .60** .72**
lnattentive-Passive .42* .30 .88** .59**
Hyperactivity lndex .70** .49* .63** -67r,*

*p < .05. **p < .01.

25
 TABLE 4.8
Correlation of ADHDT with Conners' Teacher Rating Scales-39 (n : 65)

ADHDT Values

CTRS-39 Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

Hyperactivity .77* .91* .61* .83*

Conduct Problem .69* .80* .40* .69*
Emotional-Overind u I gent .62* .7'.t* .36* .60*
,Anxious-Passive - .10 .'t0 .23 .09
Asocial .49* .75* .46* .6.l*
Daydream-Attention Problem .53* .71* .88* .77*
Hyperactivity lndex .76* .87* .60* .81*

*p .01.

with the CTRS-28 and CTRS-39 are at least moderate and Oppositional Behavior. Two of these subtests n-
overall and strong in the areas that assess ADHD These sure the same domain as ADHDT subtests. Low sc
correlations provide evidence of the ADHDT's validity. on the ACTeRS are indicative of behavior problenr-
Data ,.vere collected from teachers in California.
Correlation of the ADHDT with the ADDES-SV. The nois, Louisiana, Michiqan, Missouri, and Texas. T:
ADDES-SV is one of the newer rating scales for evaluat- teachers rated a total of 115 students (72 males airc
ing ADHD. It contains three subscales and a sum of sub- females; age range from 3 to 23) on both the ACTeRS
scale scores. The names of the ADDES-SV subtests are the ADHDT. Table 4.10 provides the results of the c,
similar to those in the ADHDT: Inattentive, Impulsive, and lations of the two tests.
Hyperactive. More importantly, they purport to measure The correlations between the subtests of the ACI
the same constructs. Low scores on the ADDES-SV are and the ADHDT demonstrate a moderately neqative :
indicative of problems in the construct being assessed. tionship. All but one exceed the .35 criteriorr. The me.
To evaluate the relationship between the two instru- of the correlations is - .52. The strongest correlation.
ments, 66 students with ADHD were rated on the ADHDT expected, were between Attention and Inattention
and the ADDES-SV. Twenty ol the students were from Hyperactivity and Hyperactivity. The correlations of ,
North Carolina, 20 from New Mexico, and 26 from Ohio. tests Attention and Hyperactivity are also strongly reial
In the sample were 52 males and 14 females. Their ages to the ADHD Quotient (r = -.71).
ranged from 3 to 23. Correlations of test scores revealed
a significant and strong relationship between the ADHDT Correlation of the ADHDT with the BES-2. The
and the ADDES-SV. Because the two tests use opposite BES-2 is a behavior ratinq scale used for assessment oi
formats for scoring, the relationship o{ the ADDES-SV students with behavioral problems. Comprised of 76 items
to the ADHDT is negative. High scores on the ADHDT that group into five subscales, the BES-2 assesses for the
correlate with low scores on the ADDES-SV. Results of characteristics of behavioral disorders,/emotional distur'
the correlations can be found in Table 4.9. bance. Its five subscales are Learning Problems; Interper'
All of the correlations reported are strong; the median sonal Difficulties; Inappropriate Behaviors; Unhappiness,
of the 16 correlations is -.81. More importantly, the rela- Depression; and Physical Symptoms/Fears. The sum o1
tionship between similar subtests is exceptionally strong: the subscale standard scores generates a Behavior Quo.
Hyperactivity and Hyperactive. *.82; Impulsivity and tient. Low scores are indicative of more serious behavior
Impulsive, -.81; Inattention and Inattentive. -.86; and problems. The BES-2 scoring format is the opposite o1
the ADHD Quotient and the Sum of Subscale Standard the ADHDT. Low scores on the BES-2 correlate with high
Scores, -.88. These results are substantial and provide scores on the ADHDT.
further evidence of the validity of the ADHDT relative Thirty-five subjects with ADHD were rated on the
to another instrument that measures ADHD. BES-2 and the ADHDT. The subjects were from five
states: Iowa (n = 5), Ohio (n = 9), Louisiana (n = ,

Correlation of the ADHDT with the ACTeRS. The Missouri (n = 7), and Texas (n = 7).Twenty-two of th=
ACTeRS is a short behavior checklist that includes 24 subjects were male, and 13 were female, their ages rar=.
items relevant to classroom behavior. The items cluster ing from 3 to 21. Results of this study are reported '
into four factors: Attention, Hyperactivity, Social Skills, Table 4.1 1.

26
 TABTE 4.9
Correlation of ADHDT with Attention Deficit Disorders Evaluation Scale-School Version (n : 65)

ADHDT Values

ADDES-SV Subtest Hyperactivity- lmpulsivity lnattention ADHD Quotient

lnattentive - .57* - .63* -.85* - .77*

lmpulsive -.78* - .81* -.71* - .83*
Hyperactive -.82* - .81* - .74* -.96*
Sum of Subscale Standard Scores _ .79* - .81* r .8'l * -.88*
*p < .01.

TABLE 4.10
Correlation of ADHDT with the ADD-H Comprehensive Teacher's Rating Scale (n = 115)

ADHDT Values

ACTeRS Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

,Attention _.59* -.57* -.78* - .71*

Hyperactivity - .71* -.52* *.45* -.64*
Social Skills - .41* -.47* -.54* -.50*
Oppositional Behavior -.37* -.50* -.27* -.41*
*p < .01.

TABLE 4.11
Correlation of ADHDT with the Behavior Evaluation Scale-2 (n : 35)

ADHDT Values

BES-2 Subtest Hyperactivity lmpulsivity lnattention ADHD Quotient

Learning Problems -.35* -.31 -.78** -.54**

lnterpersonal Difficulties -.55** - .72** - .53** - .67*r'
lnappropriate Behaviors -.63** - .70** - .67** _ .73**
Unhappi ness/Depression -.53** -.59** - .77** - .70**
Physical Symptoms/Fears -.55** -.71** -,55** - .57**
**
Behavior Quotient -.47** -.55** - .61** - .6'l

*p < .05. **p < .01 .

The correlations achieved are relatively strong. Of the students'behaviors at home, in school, and in interper'
24 correlations, 23 are equal to or above .35. The median sonal relationships. For correlating the BRP-2 with the
correlation coefficient was .61. The strongest correlation ADHDT, only the Teacher Rating Scale was used. Com-
was between Learning Problems on the BES-2 and Inat- posed of 30 items, each item of the Teacher Rating Scale
tention on the ADHDT (r = -.78), followed by the strong is a sentence describing behavior that may be observed
correlation of Interpersonal Difficulties with Impulsivity at school. Low scores are indicative of behavior problems.
(r = -.72). Overall, these results indicate that there is a Thus, one would predict a negative relationship with
strong but negative relationship between the scores on scores on the ADHDT.
the ADHDT and the scores on the BES-2. Teachers from four states (Michigan, Connecticut,
Illinois, and Maryland) completed the BRP-2 and the
Correlation of the ADHDT with the BRP-2. The ADHDT on 67 students with ADHD. Forty-three of the
BRP-2 is a battery of six instruments used to evaluate students were male, and24 were female, their ages rang-

27

l
ing from 3 to 16. The correlations of test results are pre- The median correlation coefficient was .34. More
sented in Table 4.12. The correlations, as hypothesized, importantly, the relationships between scores that shoulC
are negative and substantial. Averaging -.85, these corre- be related are significant and substantial. Very strong rela-
lations demonstrate the strong relationship between the tionships are seen in the correlations of Attention Prob-
problems of ADHD and teachers' overall ratings of behav- lems on the RBPC to Inattention on the ADHDT (.95), Motor
ior. These results are further evidence of the validity of Excess to Hyperactivity (.83), and the ADHD Quotient to
the ADHDT in assessing problem behaviors of students Attention Problems (.88) and Motor Excess (.81). These
in school. findings support the validity of the ADHDT as an effective
assessment instrument for evaluating behavioral problems.
Correlation of the ADHDTwith the RBPC. The RBPC
is a behavior checklist that is frequently used in research
concerning children and adolescents with emotional and Construct Validity
behavioral disorders. The RBPC is comprised of six sub-
scales: Conduct Disorder, Socialized Aggression, Attention Construct validity addresses the theoretical framework
Problems-lmmaturity, Anxiety-Withdrawal, Psychotic on which a test is built by examining the relationship of
Behavior, and Motor Excess. Most of the items on the test performance to the hypothetical constructs that
RBPC are strongly related to interpersonal behavior underlie or explain the test performance. To demonstrate
problems. It uses an identical scoring format to the the construct validity of a test, one must delineate as fully
ADHDT. Therefore, one would hypothesize a positive as possible the variable (construct) that the test purports
relationship between the two tests. to measure. This is done by setting up hypotheses about
Teachers from Indiana, Iowa, Michigan, and Ohio scores on the tests in light of all that is known about the
rated their students (n = 35) on the ADHDT and the variable. The hypotheses are subjected to scientific inves-
RBPC. Thirty-three of the students were male and 2 tigation, and they are accepted or rejected on the basis
were female. Their ages ranged from 6 to 17. As hypothe- of the results. The following hypotheses were tested:
sized, there was a positive relationship between scores
on the two tests. Table 4.13 presents the results of the 1. The various subtests of the ADHDT should be posi-
correlations. tively related to each other.

TABLE 4.12
Correlation of ADHDT with the Behavior Rating Profile-2 Teacher Rating Scale (n : 35)

ADHDT Values

Hyperactivity lmpulsivity lnattention ADHD Quotient

BRP-2 Teacher Rating Scale - .84* -.83* -.87*

"p < .01.
all r'sp < .01

TABLE 4.13
correlation of ADHDT with the Revised Behavior problem checklist (n = gs)

RBPC Subtest Hyperactivity lmpulsivity lnattention ADHD euotient

Conduct Disorder .40* .60** .'13 .37*

Socialized Aggression .04 .09 -.23 -.07
Attention Problems .82** .69** .95** .88**
Anxiety-Withdrawal .23 .24 .39* .31
Psychotic Behavior .26 .24 .12 .25
Motor Excess .93** .77** .69** .8'l **

*p
'05.
**p 01

28

I
2. Subtest items should strongly relate to the subtest total information provided by this type of validation is most
score. relevant to tests used in the selection and classification
of subjects. According to Anastasi (1988), "Concurrent
3. Scores on the ADHDT should discriminate persons validation is relevant to tests employed lor diagnosis of
with ADHD from subjects who do not have ADHD. existing status, rather than prediction of future outcomes"
4. Scores should discriminate subiects with ADHD from (p. 1a6). One of the purposes of the ADHDT is to distin-
subjects with other types of behavioral problems. guish persons with ADHD from persons with behavioral
problems that are not a result of ADHD.
Interrelationship Among ADHDT Subtests. A com- The ability of the ADHDT to differentiate between
mon way to study a test's validity is tc inspect the inter- subjects from various diagnostic groups was established
correlations among its subtests. Because all of the ADHDT from analysis of data collected during the standardiza-
subtests are intended to measure ADHD behaviors, one tion procedure. A sample of 551 subjects was randomly
would expect to find strong intercorrelations between drawn from subjects in the normative study. Of the sub-
them. To examine the relationships of the ADHDT sub- jects in this sample, 352 were previously diagnosed by
tests, standard scores of each subtest and quotient were their school district as having ADHD, and 178 were
correlated. not diagnosed as having ADHD but had a variety of other
Table 4.14 displays the results of the correlations of diagnoses (e.g., mental retardation, emotional distur-
the ADHDT subtests. All of the correlations are signifi- bance, learnirtg disabilities). Standard scores and the
cant (p < .01) and quite large in magnitude. Analysis of ADHD Quotients of these subjects were subjected to a dis-
these data makes apparent the strong relationship the sub- criminant analysis to determine how well the ADHDT
tests have to each other and to the overall composite. discriminated the students diagnosed with ADHD from
Clearly, the items of each subtest are measuring the same those who were not diagnosed with ADHD. Table 4.15
construct (i.e., behavioral characteristics of ADHD). presents the results of this analysis.
Inspection ol the data in Table 4.15 show statistically
Item Vatidity of ADHDT Subtests. Evidence of the
significant differences between the means of the ADHD
item validities associated with the ADHDT may be found
sample and the non-ADHD sample. On every subtest,
in Table 4.6. The discriminating power of an item. com-
the ADHD sample was significantly higher than the non-
puted by the point-biserial method of item-total correla-
ADHD sample. Although each subtest was effective in
tion. is sometimes referred to as item validitl' because
correctly classifying the subjects with ADHD, the ADHD
these coefficients reflect the degrce to which the items
of a subtest or test are measuring the same constructs, Quotient was the most accurate. Using the ADHD Quo-
tient alone, the computer was able to correctly classify
These data can be cited as evidence of a test's reliabilitv
the subjects in terms of their assigned diagnostic group
because strong item discrimination can only result from
al a 92'% accuracy rate. The results of the discrimination
strong construct validity.
analysis were all significant (p < .01).These results verify
Table 4.6 depicts the point-biserial correlations of
the ability of the ADHDT to discriminate the ADHD from
each item to its respective subtest. These items are all
the non-ADHD sample. This finding provides the exam-
significant (p < .01) and related to their respective sub-
iner with confidence in the use of the ADHDT as a diag-
test, indicating that these items are making strong con-
nostic instrument for identifying persons suspected of
tributions to the construct being measured.
having ADHD.
Discrimination of Diagnostic Groups. The ability' to
identify subjects who belong to different diagnostic groups Validation Through Contrasted Groups. Another
also requires concurrent criterion-related validitv. The method used to establish a test's validity is to contrast

TABLE 4.14
lntercorrelation of the ADHDT Subtests

ADHDT Subtest Hyperactivity lmpulsivity lnattention

Hyperactivity
lmpulsivity .86*
lnattention .70* .75*
ADHD Quotient .93* .94* .89"

*p < .01.

29
 TABLE 4.15 dard score data from 1,948 subjects were used ir
Classification Results of the analysis. The subjects ranged in age from 3 to 23
ADHD Quotient on the ADHDT age : 11-11) and included 1,336 males (68.6%)and -
females (31.4%). The subjects were previously diagn,-,..
Diagnostic Predicted Predicted by school district officials as mentally retarded (n = i : -
Group ADHD Non-ADHD emotionally disturbed (n = 154), and learning disa:, =
(n = 405). To control for confounding effects of coexi:
ADHD 311 (88.4%) 41 (11.60/o\ ing handicaps, subjects were selected for this study if the
(n = 3s2) had only one diagnosis. Therefore, the handicaps of th
subjects in the sample were singular and represent t
Non-ADHD 2 (1.10/o) 176 (98.9o/o) "pure" a sample as possible. A group of nonhandicappe
(n = 178)
subjects (n = 541) was used as a control group.
To test the hypothesis that persons from differer
diagnostic groups will differ from persons with ADHD o
groups who are expected to differ on the attributes being the ADHDT, the author performed a one-way analysis r
measured on the test. In terms of the ADHDT, subjects who variance procedure, contrasting subtest standard score
will be tested with this instrument will be persons with and the ADHD Quotient for the ADHD and non-ADH
behavioral problems. Therefore, to establish the validity subjects by diagnostic category. The results from th
of the ADHDT, subjects with ADHD were compared to study are reported in Tables 4.16 and 4.17.
non-ADHD subjects. These subjects were drawn from the Table 4. 16 reports the mean standard scores and star
subjects used in the standardization study. ADHDT stan- dard deviations for all of the diagnostic groups. On eac

TABLE 4.16
Performance of Diagnostic Groups on ADHDT Subtests and the ADHD Quotient

Diagnostic Group
Emotional
Learning Disability Disturbance Mental Retardation Nonhandicapped

ADHDT Values M 5D 5D

Hyperactivity 7 3 9 3 7 3 7 3
lmpulsivity 7 3 10 3 8 3 7 3
lnattention 7 3 9 3 8 3 6 3

ADHD Quotient 79 16 95 18 16 19

TABLE 4.17
Significant Differences* Between Diagnostic Groups on ADHDT Subtests and ADHD Quotient

Diagnostic Group
Learning Emotional Mental
Disability Disturbance Retardation Nonhandicapped

ADHD "t, 2, 3, 4 1,3,4 1,2,3,4 1,2:,4

Learning Disability
Emotional Disturbance "t, 2, 3, 4
Mental Retardation 1,2,3,4 t, {2, a
Nonhandicapped 3,4 1,2,3,4 t, Jt, q

Note. 1=HyperactivitySubtest;2=lmpulsivitySubtest; 3=lnattentionSubtest; 4=ADHDQuotient.

*p < .05.

30
 ADHDT subtest and the ADHD Quotient, the ADHD ing disability groups are another validation of the effec-
group received significantly higher scores (p < .05) tiveness of the ADHDT in differentiating between persons
than the other diagnostic groups. These results validate with ADHD and persons without any known impairments.
that the ADHDT scores can be used in the identifica-
tion of subjects with ADHD from subjects with other
diagnoses. These results also validate that the attributes
of persons in specific diagnostic groups are reflected on Summary
the scores of the ADHDT subtests. These differences are
illustrated in Table 4.17 and are discussed in the follow- The Attention-Deficit/Hyperactiuity Disorder lesr (ADHDT)
ing text. is a highly standardized, norm-referenced instrument
In comparing the standard scores oi the different diag- designed for use by teachers, parents, and others ior the
I
nostic groups, one can see that, with one exception, the purpose of assessing students who are suspected of hav-
ADHD group is rated significantly greater than the other ing attention-deficit disorders. It was developed through
diagnostic groups on every ADHDT subtest. The excep- empirical and logical techniques and was normed on a
tion is found in the emotionally disturbed group's mean sufficiently large sample of persons who have the diag-
rating of l0 on Impulsivity (see Table 4.16). This finding nostic characteristics for whom the test will be utilized
is not remarkable, however. Because of behaviors charac- in the future. The quality of the ADHDT was confirmed
terized by lack of impulse control, many students are iden- through studies of the test's reliability and validity.
tified as emotionally disturbed. Thus one would expect The reliability of the ADHDT is well within accept-
to see this diagnostic group score high on this subtest. able ranges. The internal consistency and reliability of
Otherwise, the scores of the different diagnostic groups the subtests were determined to be in the .80s and .90s.
on the subtests and ADHD Quotient are what one might Studies of both test-retest and interrater reliability con-
predict. Tables 4.18 through 4.21 graphically illustrate the firm the utility ol the ADHDT as a diagnostic instrument.
significant differences between the diagnostic groups The ADHDT is one of the few tests for ADHD, if not the
included in the validation study. only, that was normed entirely on persons with ADHD.
On all the subtests and the ADHD Quotient, the non- The validity of the ADHDT was demonstrated through
handicapped group and the learning disabilitl' group several studies. These studies confirm that (a) the items
received significantly lower ratings than the other diag- of the subtests are representative of the characteristics
nostic groups. This is to be expected because persons in of ADHD; (b) the scores are strongly related to each other
these groups are not known to demonstrate the intensity and to performance on other tests that screen for ADHD;
of behavior problems captured on the ADHDT. The low and (c)the ADHDT can discriminate persons with ADHD
scores of the subjects in the nonhandicapped and learn- from subjects with other behavior disorders.

TABLE 4-18
Significant Differences Between Diagnostic Groups on the Hyperactivity Subtest of the ADHDT

Diagnostic Learning Mental Emotionally

M Group Disability Nonhandicapped Retardation Disturbed

6.51 LD
6.76 Nonhand.
7.98 MR
9.74 ED
't0.'t0 ADHD

*Significant difference between these diagnostic p< .05.

aroups,

31
 TABLE 4.19
Significant Differences Between Diagnostic Groups on the lmpulsivity Subtest of the ADHDT

Diagnostic Learning Mental Emotionally

M Group Disability Nonhandicapped Retardation Disturbed

6.74 LD
6.76 Nonhand.
7.98 MR *
9.74 ED *
10.10 ADHD :rt

*Significant difference between these diagnostic groups. p< .05.

TABLE 4.20
Significant Differences Between Diagnostic Groups on the lnattention Subtest of the ADHDT

Diagnostic Learning Mental Emotionally

M Group Disability Nonhandicapped Retardation Disturbed

5.54 Nonhand.
7.05 LD
7.96 MR
8.73 ED
9.98 ADHD

*Significant difference between these diagnostic groups, p< .05.

TABLE 4.21
Significant Differences Betureen Diagnostic Groups on the ADHD Quotient of the ADHDT

Diagnostic Learning Mental Emotionally

M Group Disability Nonhandicapped Retardation Disturbed

76.46 Nonhand.
79.07 LD
8s.56 MR
94.91 ED
100.31 ADHD

*Significant difference between these diagnostic groups, p< .05.

DO
.)L
 References

American Psychiatric Association. (1980). Dragnostic and sta- McCarney, S. B. (1989). Attention Deficit Disorders Eualuation
tistical manual of mental disorders (3rd ed.). \\'ashington. Scale-School Version. Columbia, MO: Hawthorne.
DC: Author. McCarney, S. 8., & Leigh, J. E. (1990). Behottior Eoaluation Scale
American Psychiatric Association. (i991). Dlognostic and sta- (2nd ed.). Columbia, MO: Hawthorne.
tistical manuol of mental disorders (4th ed). \\'ashington. Nussbaum, N., & Bigler, E. (1990). Identification and treatment
DC: Author. of attention deficit disorder. Austin, TX: PRO-ED.
Anastasi, A. (1988). Psychological lesfrng (6th ed.) \eu'\'ork: Quay, H. C., & Peterson, D. R. (1987). Reuised Behauior Problem
Macmillan. Checklist. Coral Gables, FL: Quay.
Bain, L. J. (1991). A parent's guide to attention deficit disorders Safer, D. J., & Krager, J. M. (1988). A survey of medication treat-
New York: Dell. ment for hyperactive,/inattentive students. Journal of the
Barkley, R. A. (1981). Hyperactiue children: A handbook of diag- Amer ican Medical Association, 2 60, 2256-2258.
nosis and treatment. New York: Guilford.
Sha-"-witz. S. E., & Shaywitz, B. A. (Eds). (1992). Attention deficit
Brown, L., & Hammill, D. D. (1990). Behauiar Rating Prr,tfiLe
disorder comes of age: Touard the twenty-first century.
(2nd ed.). Austin, TX: PRO-ED.
Austin, TX: PRO-ED.
Conners, C. K. (1990). Conners' Rating Scales Manual. )iorth
Task Force on DSM-IV. (1993\. DSM-IV Droft Criteria, 3-1-93.
Tonawanda, NY: Multi-Health Systems.
Washington, DC: American Psychiatric Association.
Cronbach, L. J. (1951). Coefficient alpha and the internal struc-
ture of tests. Psyclrometriha, 16, 297-334. Ullmann, R. K., Sleator, E. K., & Sprague, R. L. (1991). ADD-fi
DeMeyer, M. K., Hingtgen, J. N., & Jackson, R. K. 11981). Infan- Comprehensiue Teacher's Rating Scale (2nd ed.). Cham-
tile autism reviewed: A decade of research. Schize-tphrenio paign, IL: MetriTech.
Bulletin, Z(3), 388-451. Wender, P. H. (1987). The hyperactiue child, adolescent, and
Hammill, D. D., Brown, L., & Bryant, B. R. (1992). .A consumer's adult: Attention deficit disorder through the lifespan. New
guide to tests in print (2nd ed.). Austin. 1X: PRO-ED. York: Oxford University Press.
Ingersoll, B. D., & Goldstein, S. (1993). Attention deficit disorder Zentall, S. S. (1993). Research on the educational implications
and learning disabilities: ReaLities, myths, and controDer' of attention deficit hyperactivity disorder. Exceptional
sial treatments. New York: Doubledav. Children, 60(2), 143-t53.

,,
.).)
 Appendix
Normative Tables

Table A. Converting Raw scores to Standard scores and Percentiles (Males)

(Females)
Table B. Converting Raw Scores to Standard Scores and Percentiles
and Percentile Rank
Table C. Converting Sum of Standard Scores to the ADHD Quotient

35

I
 TABLE A
Converting Raw Scores to Standard Scores and Percentiles
(Males)

Hyperactivity lmpulsivity lnattention

Standard
Score Ages 3-7 Ages 8-23 Ages 3-23 Ages 3-23 Percentile

1
2 0 <x
3 0
:
4 1 tJ 0 {-
{ 5
5 2*3 1

6 4-7 2-3 2-4 I

il-i 5-7 11-12 ta
7 8-1 0
8 11-13 E-r 0 8,9 13-15 r:
s 14*16 11*13 10-11: 37
't0 't7-2A 14-"t6 12=13 50
11
tl la t a )f1 14-15 ai
12 f;1 21- 22 16-17 il
13 24 23 18 84
14 z> v4 19 91
15 26 )r 20 *i,

16 2b $8
17 99
18 >99
19
20

TABLE B
Converting Raw Scores to Standard Scores and Percentiles

Hyperactivity lmpulsivity lnattention

Standard
Score Ages 3-7 Ages 8-23 Ages 3-23 Ages 3-23 Percentile

2 0 <,1
3 0 1-2
4 1 3-4
5 2-4 I 1 5*6 5
6 5*7 1A
2*4 7-9 9
7 B-9 5-7 10-11 :&
-] ii
8 10-12 8-9 12-13
I 't 3-1s 10-13 14*16 37
IO 16*18 13-15 17-19 50
11 't9-20 16-18 't3-15 20-21 ?is 1

12 21-22 19-20 16-'.t7 22-23 li

13 23*24 21*22 18 24 84
14 25 23-24 19 25 91
15 26 25 26 t'.
16 26 :r8
17 99
I8 >99
'19
20

36

I
 TABTE C
Converting Sum of Standard Scores to ADHD Quotient and Percentile Rank

Sum of
ADHD Quotient Subtest Standard Scores Percentile Rank

ffi
153
't62
,, , 'l''I
59
'
$.ffi
ffi
>99
>99
ffifuri":t'lrllitr .::, . #.ffi
*ffi*+
159
58 1
a,wi
>99
1s8 ; >99
ffi"lf;l#- l] lr,r.. #
ffi+ffitffi '56 ' i$$,is
155 >99

ffi
1s4 >99
'j ,*9
-:- iHffi
15'l 54 >99
150 >99
rffii&ffi
li?;l$g,
53 +€s
199.
't47 52 >99
146 >99
iw
,,i{$
5I r.$n.$SJ

i*ffi
,
143 50 >99
142 >99
i,4t
rf.$
49 ffi.
ffi
139 48 >99
138 >99
rr!.,'*,?: ,sig&ii
!t& :
Sq*
.-99-
135
134 45 99
i$,*il iiffi
.li$*, 11]ffii
131 : 99
130 44 98
{i,e$, .*-:. ii:
,ffi.,'
't27
43 ffi
97
126 o, 95
:1Zx
.*.2*
123
' '.,:'+,..

1',,
ffi
#*
94
122 93
#.*,i , 4.0 ffi
:$,S6, '1..,. ; .
iffi
119 39 iio
118 89
*,1rfl 38 ,ffi
l*{ -:- ffi
115 37 84
114 82
rtil& 36 #s1

37

I
TABLE C. Continued

Percentile Rank

7t)
6C
i:
*.

58
{i

:
5G
4?

4i
39
3;

3G
27

l:
1t
:9
l::

14

't,l
1f

2
2

<1
<1

38
 TABLE C. Continued

Sum of
ADHD Quotient Subtest Standard Scores Percentile Rank

60 o:j
59 <!
58
57
56 {':
55 "::
34
53
52 <3
.- 1
51
50
49
48 -- -1

47 ":1
46
45
44 ii
43 _/1

42
41
40 <i
39 1"a

38
37
36 <,!
35 <i

39

I
 ADHDT
Section l. Information

Subject's Name
Address

Rater's Name
Attention-Deficit/ Relationship to Subject
Hyperactivity Disorder Test Examiner's Name and Title
A Method lor ldentifying Date of ADHDT Rating
lndividuals with ADHD Year

SUMMARY/RESPONSE Subject's Date of Birth

Year Month
FORM Subject's Age
Year Month

Section ll. Score Section lV. Profile of Scores

Raw Other Measures
Subtests Score SS ADHDT ADHDT of lntelligence,
Subtests Composite Achievement,
or Behavior
Hyperactivity

lmpulsivity

lnattention o
o
o
o
<1,

t, I 9o
o>'E a> oo0)oo
aoaao
Sum of Standard Scores o(E
o!, !E'6e 8..E o.9 lffff
o=.Y I6
:s
o.h
CLCL=
I}:
tso
of
()o o,
<o
6AAAA
oooo0)
FFFFF
ADHD Ouotient
l::i*rtg;:t*f
:3s$.ff
:i:ffi:i:' Section lll. Interpretation Guide
Subtest
Standard ADHD Degree of Probability
Scores Quotient Severity of ADHD
17-19 131+ Very High
15-16 121-130 High
13-14 111-120 Above Average
8-12 90-1 1 0 Average

6-7 80-89 Below Average

4-5 70-79 Low
1-3 <69 o Very Low
J=

O 1995 by PRO-ED, lnc. Additional copies of this form (#6882) are available from PRO-ED,
5 6 7 BI 10 04 03 02 01 8700 Shoal Creek Blvd., Austin, TX 78757, 5121451-3246.
 'l

DIRECTIONS: Please indicate which of the following behaviors/characteristics are a problem for this individ-
ual. Mark or circle 0 if the behavior is not a problem (the subject rarely demonstrates this problem, and it does
not impair his or her functioning) or if you have not had the opportunity to observe the behavior. Mark or circle
1 if the item refers to a behavior that is a mild problem (the subject sometimes demonstrates this behavior, and
it occasionally causes problems and impairs his or her functioning.) Mark or circle 2 if the item refers to a be-
haviorthat is a severe problem forthis individual (the subject frequently demonstrates this behavior, and it usu-
ally causes problems and impairs his or her functioning.) Do not skip any items.

Hyperactivity Subtest
t-q)
ob;
EOJE
orl.: E
o-Fo)
<urh
9>
o=dJ
z=a

1. Loud 0 12
2. Constantly "on-the-go" 0 12
3. Excessive running, jumping, climbing 0 12
4. Twisting and wiggling in seat 0 12
5. Easily excited 0 12
6. Grabs objects 0 12
7. Excessive taiking 0 12
8. Difficulty remaining seated 0 12
9. Constantly manipulating objects 0 12
10. lnability to play quietly 0 12
11. Fidgets 0 12
12. Restless 0 12
13. Squirms 0 12
Hyperactivity Sum
 tr
lmpulsivity Subtest L-0)
-L
ot;
EOE
9*d
o-Yo
rErb 9>
2 = E

14. Acts before thinking 012

15. Shifts from one activity to the next 012
16. Fails to wait for one's turn 012
17. Difficulty waiting turn 012
18. Blurts out answers 012
19. lmpulsive 012
20. lnterrupts conversations 012
21. lntrudes on others 012
22. Does not wait for directions 012
23. Fails to follow rules of games 012
lmpulsivity Sum

lnattention Subtest tr
E.E
:E6 E
o-E,.
te-o
-rL
9>
2=E
24. Poor concentration 012
25. Fails to finish projects 012
26. Disorganized 012
27. Poor planning ability 012
28. Absentminded 012
29. lnattentive 012
30. Difficulty following directions 012
31^ Short attention span 012
32. Easily distracted 012
33. Difficulty sustaining attention 012
34. Difficulty staying on task 012
35. Difficulty completing tasks 012
36. Frequently loses things 012
lnattention Sum
3
 Section Vl. Key Questions

1. Does the person demonstrate six or more symptoms of inattention, or six or more symptoms of hyperactivity,
or impulsivity listed in each subtest?

2. Does the person exhibit the behavioral problems in a variety of environments?

3. Does the person demonstrate the behaviors considerably more frequently than do most people of the same
mental age?

4. Has the person demonstrated the behaviors for at least 6 months?

5. Did the person first demonstrate the behaviors before age 7?

6. ls the person's functioning (at school, home, and work) significantly impaired?

7. Are there other conditions that could possibly be causing the behavioral problems? lf yes, what are the
conditions?

8. Who has previously evaluated this person and what were the results?

9. What specific interventions have been attempted to treat the person's problems?

'10. What additional information needs to be collected?

Section Vll. Recommendations and Comments