Content uploaded by Hanoch Livneh
Author content
All content in this area was uploaded by Hanoch Livneh on Nov 16, 2015
Content may be subject to copyright.
d i s a b i l i t y a n d r e h a b i l i t a t i o n , 2000 ; vol. 22, n o . 5, 211±224
Measurement of attitudes towards persons
with disabilities
RICHARD F. ANTONAK‹* and HANOCH LIVNEHŒ
‹School of Education, Indiana State University, Terre Haute, IN 47809, USA
ŒDepartment of Special Education and Counsellor Education, Portland State University,
Portland, OR 97207, USA
Accepted for publication : March 1999
Abstract
Purpose : The purpose of this paper is to assist those engaged in
research dealing with attitudes towards personswith disabilities
by presenting a catalogue o f various attitude measurement
methods.
Method : A review of the metho dological and psychological
literatures on t he measurement of attitu des to wards persons
with disabilities.
Results : The review uncovered 10 direct methods to measure
attitudes, in which the respondents are aware that they are
partic ipating in an experi ment an d 14 in direct metho ds in f our
categori es that are not plagued b y attitu de-distorti ngin¯ u ences
becau se t he resp onden ts are not aware t hat the ir attit udes are
bein g measured . A d iscussio n of each metho d with examples i s
provided, fo llowed by implications for reh abilitation prac-
titioners, rehabilitation education and training, and rehabili-
tation researchers.
Conclusions : The investigation of attitudes towards persons
with d isabilit ies requi res i nnovati ve experim ental meth ods and
psycho metrically sou nd instr uments th at are rel iable, valid , and
multidimensional. Without such instruments, it will not be
possible to obtain conclusive answers to important research
questions concerning the relationship between these attitudes
and the acceptance and int egration of persons with disabilities
into society.
Introduction
Profound changes in the provision of educational,
psychosocial and vocational services are increasing the
integration of children, adolescents and adults with
disabilities into our schools, neighbourhoods and work-
places. Full acceptance of persons with disabilities by
persons wi thout disabili ties will n ot occur, ho wever, until
subtle barriers can be eliminated. Most scholars and
researchers agree that one of the factors inherent in the
subtle barrier is the attitudes of health and rehabilitation
professionals, teachers, employers and coworkers,
* Author for correspondence; e-mail: soedean!indstate.edu
educators- and counsellors-in-training, parents, peers
and persons with disabilities themselves."±(Negative
attitudes towards persons with disabilities create real
obstacles to the ful® llment of their roles and the
attainment of their life goals. Knowledge of attitudes of
persons without disabilities towards persons with dis-
abilities helps us to understand the nature of the
interaction between the two groups. Furthermore,
understanding the underlying dimensions of negative
attitudes may suggest diŒerential change procedures and
promote appropriate assessment of the eŒects of these
interventions.
Estimation of the predominant attitudes of various
populations concerning persons with disabilities,
together with study of the interrelations of knowledge,
attitudes and behaviour, is necessary for suggesting
desired ends to policymakers, designing intervention
programmes to modify attitudes towards persons with
disabilities and evaluating professional training pro-
grammes in counselling, rehabilitation and special edu-
cation.),*The usefulness of attitude research for accom-
plishing these goals is dependent upon the data that are
obtained, and that, in turn, is dependent upon the
method selected to measure attitudes. Without assurance
of the psychometric soundness of the method and
freedom of the obtained data from respondent biases,
con® dence in the conclusions of attitude research may be
unjusti® ed." !
Attitudes towards persons with disabilities have
changed over the years. Attitude measurement tech-
niques also have changed, becoming increasingly soph-
isticated both theoretically and technically. In addition,
a collection of ingenious methods has been suggested as
alternatives to the traditional overt and obtrusive
methods to measure attitudes especially when the
targeted attitude referent is socially sensitive and where
conscious or unconscious mechanisms may interfere and
alter the respon dent’ s attit udes." " The purpose of this
paper is to assist those engaged in research dealing with
Disability and Rehabilitation IS SN 0963-8288 print}ISSN 1464-5165 o nline ’2000 Taylor & Francis Ltd
http:} } www.tandf.co.uk}journals}tf}09638288.html
R.F.Antonak and H.Livneh
attitudes towards persons with disabilities by presenting
a catalogue of various attitude measurement methods.
We begin with an overview of attitude measurement and
the measurement of attitudes towards persons with
disabilities. Next, we present a brief description of 10
traditional methods to measure attitudes, together with
examples, in which the respondents are aware that they
are participating in an experiment. In this part, we also
discuss confounding variables that decrease the validity
of data obtained from these measurement methods.
Next, we discuss and illustrate 14 measurement methods
in four categories that are not plagued by attitude-
distorting in¯ uences because the respondents are not
aware that their attitudes are being measured. We
conclude by presenting implications for rehabilitation
practitioners, rehabilitation education and training, and
rehabilitation researchers.
o v e r v i e w o f a t t i t u d e m e a s u r e m e n t
Attitudes are regarded as latent or inferred psycho-
social processes that lie d ormant within one’ s self unless
evoked by speci® c referents." # ," $ Because attitudes are
acquired through experience, predisposing one’ s
responses to sociocultural events and other people, the
study of attitudes may shed light on the socialization
process and the event s contributing to i t. It may, l ikewise,
contribute to our understanding of prejudice formation
as acquired through assimilation of values held by one’ s
parents and peers. Finally, ® ndin g ou t what a p erson’ s
attitudes are towards a referent, in conjunction with
knowledge of various situational and other personality
variables, may aid the researcher in explaining and
ultimately predicting the respondent’ s behaviour towards
the referent. Knowledge of the mechanisms underlying
the development and structure of attitudes towards
persons with disabilities is considered to be necessary for
changing them and thereby increasing the integration of
persons with disabilities into the larger society.*," %
Attitude measurement converts observations of a
respondent’ s b ehaviour tow ards a referent into an index
that represents the presence, strength and direction of the
attitude presumed to underlie the observed behaviour." !
The researcher selects a measurement method re¯ ecting
assumptions ab out t he respond ent’ s in ternal state, the
referent towards which the respondent directs his or her
behaviour, and the relationship between the respon dent’ s
internal state and the external behaviour, as well as the
parameters of the research situation, such as cost, time,
availability of respondents, availability of scales, and the
researcher’ s competence and motivati on. Some
researchers are committed to only one type of measure-
ment method, using it to answer all research questions.
Other researchers may avoid a particular type of
measurement method, perhaps because this method has
not traditionally been a part of their professional
discipline, even if it is the most appropriate method to
obtain the data needed to answer the research question.
Experts generally divide methods to measure attitudes
into direct and indirect methods."&," ’ Direct methods are
those in which the respondents are either informed that
their attitudes are being measured or are made aware of
it by the nature of the attitude measurement technique.
Alternatives to direct measurement methods have been
developed for those attitude measurement situations in
which : (1) the act of measurement itself may create in the
respondents an attitude towards a referent to which they
were previously un aware ; (2) ot her meth ods may l ead
the respondents to a reply that is inconsistent with their
true attitude ; or (3) the referent is so sensitive or so
charged with emotion that the observed response may be
nonpurposefully distorted by unrelated but powerful
personality characteristics.
d i r e c t m e t h o d s t o m e a s u r e a t t i t u d e s
Direct methods are by far the most widely known and
used in measuring attitudes towards persons with
disabilities."(±" * Opinion surveys ask respondents to
express their attitudes by responding to a list of questions
about the referent. A structured (closed) opinion survey
asks the respondents to select one among a small set of
responses, or all of those that they agree with, or those
that they endorse. Unstructured surveys ask that the
respondents provide not only an answer but also a
justi® cation or explanation for the answer. Caruso and
Hodapp# ! used an opinion survey to investigate
diŒerences in respondents’ perceptions of persons with
mental retardation and persons with mental illness.
Comer and Piliavin# " explored the reactions of
respondents with physical handicaps in interactions with
interviewers with and without physical handicaps.
Linkowski et al.# # used an opinion survey that asked
respondents to list those disabilitiesthey considered to be
the most and least severe, and to respond to a series of
questions, suc h as : wh at wou ld a person w ith thi s
disability be able to do ? Not be able to do ?
Interviews require that the researcher interact directly
and verbally with the respondent, although the interview
may take place over the telephone or through the use of
® eld assistants. Structured interviews use a ® xed set of
questions in a ® xed sequence with all respondents,
although branching may allow the researcher to skip
over c ertain questions depend ing o n the respon dent’ s
212
Measurement of attitudes towards persons with disabilities
answer. In an unstructured interview, the researcher can
ask additional questions to ensure that the respondent
understands the questions and to explore the
respondent’ s atti tudes. Interviews have been used to
investigate the attitudes of employers towards the
employment of persons with disabilities,# $ ,# % to inquire
about lan dlord’ s experi ences renting to persons wit h
mental retardation,# & to explore the attitudes of parents
of children with mental retardation# ’ and to study
educators’ views about the e cacy of special education.# (
A comprehensive study of the educational and residential
consequences of community placement for
institutionalized children illustrates several interview
techniques.# )
Ranking methods require the respondent to arrange a
small set of items into an ordered sequence according to
some speci® ed criterion (see table 1, part A). Ranking
methods have been used frequently to compare various
groups of respondents on their attitudes towards persons
with disabilities.# * ,$ ! Abroms and Kodera$ " asked a
group of college students in introductory courses in
special education to rank order disability category labels
according to the acceptability of the disability to them, a
concept that was purposefully not de® ned. Richardson et
al.$ # used a set of drawings of children with various types
and levels of physical disabilities (e.g. a child seated in a
wheelchair, or a child who was obese) to study the
attitudes of young children with and without disabilities
of various races and religions. This technique was
subsequently used by Richardson and by others with a
variety of other respondent groups, including cross-
cultural comparison studies.$ $
Q methodology$ % requires the respondent to sort a set
of phrases or statements about the attitude referent into
piles according to some criterion, such as favourability,
intensity of agreement, or descriptiveness. The sorts of
diŒerent respondents may be analysed to derive clusters
of respondents, and the content of the items in each pile
for diŒerent clusters of respondents is examined in order
to characterize the respondents. The researcher may then
assign a respondent to a group by comparing his or her
orderings with those obtained in previous research. The
clusterings may also be examined before and after some
event or interventio n to discern changes in respondent’ s
attitudes.$ & Barker$ ’ used Q methodology to study
similarities of views of a variety of disabilities, while
Shaver and Scheibe,$ ( in a pre- and post-test design,
investigated t he in¯ uenc e of an intensive camp experie nce
on the attitudes of college students towards persons with
chronic mental illness.
Sociometrics are designed to uncover how a respon-
dent within a group behaves or intends to behave
towards another person (the referent) within the group
when given a choice of behaviours. For example, the
respondent may be presented with a roster of all the
persons in h is or her cl ass and asked : `W hich of your
classmates on the list below do you most like to sit with
in the lu nchroom ? Next most ? ¼ L east ?’ Th e resultant
data can yield a pictorial sociogram illustrating the
number of times a child is nominated or selected in
various situations. Gottlieb$ ) provided an excellent
review of sociometric research techniques and results
concerning persons with disabilities. Altman$ * and
MacMillan and Morrison% ! discussed sociometric re-
search with a variety of groups and areas within
rehabilitation.
The original 300-item Adjective CheckList (ACL) was
constructed by Gough% " for the investigation of per-
sonality and self-concept. Subsequent investigations led
other researchers to derive a variety of scales from the
ACL, presumably measuring various aspects of per-
sonality, such as achievement, autonomy, self-con® dence
and personal adjustment. The ACL has been modi® ed
for the measurement of attitudes towards persons with
disabilities (see table 1, part B). A review of research
using adjective checklists to measure attitudes towards
children with mental retardation was provided by
Gottlieb et al.% # Other examples include investigations to
study the attitudes of children towards their peers with
disabilities,% $ and an investigation of the reactions of
undergraduate students to persons with physical and
speech disabilities.% %
The method of paired comparisons requires that the
researcher present all possible pairs of the referents that
the researcher wishes to scale and ask the respondents to
select the item in each pair that they would rate higher in
terms of some criterion (see table 1, part C). The
resultant data are analysed to yield an ordering of the
items for each respondent or for a group of respondents.
Jones, Gottfried, and Owens% & scaled respondents’
rankings of 11 groups of persons with disabilities plus
persons who are gifted and persons without disabilities in
seven social situations (e.g. as a neighbor, as a visitor).
Wicas and Carluccio% ’ investigated the attitudes of
counsellors towards persons who were culturally
diŒerent, ex-convicts, and ex-mental patients. Janicki% (
used paired-comparison dyads to investigate the attitudes
of hospital employees to 12 diŒerent groups of persons
with disabilities.
In the semantic diŒerential method,% ) a single concept
is presented followed by a set of 7 to 20 scales anchored
at each end by bipolar adjectives connected by a line
marked in intervals. Respondents are asked to mark the
line at a point that represents their rating of the concept
213
R.F.Antonak and H.Livneh
Table 1 Example s of direct methods to measure attitude s towards p ersons with disabilities
A.Ranking
Please rank the following impairments in order according to the extent to which you think they are disabling to persons with the impairment.
Use 1 for the most disabling, 2 for the next most disabling, 3 for the n ext most disabling, and 4 for the least disabling.
jjjjj cancer jjjjjblindness jjjjj epilepsy jjjjj diabetes
B.Adjective checklist
Below is a list of adjecti ves. Read these adjectives quickly and put a -next to each one you consider to be descriptive of persons with spina
bi® da. Wo rk quickly and do no t spend too mu ch time thinking about any o ne adjective.
jjjjj active jjjjj cautiou s jjjjjdistrustful jjjjjexcitable jjjjj gentle
jjjjj impulsive jjjjj lazy jjjjj meek jjjjjnervous jjjjjpleasant
jjjjj quick jjjjjreliab le jjjjjtalkative jjjjj unkind jjjjjweak
C.Paired comparisons
Each ite m below present s two possible impairments. Please read each pair and circl e the impairment that you beli eve is the more disabling to
persons with the impairment. Judge each pair separately.
1. (a) polio 2. (a) epilepsy 3. (a) polio
(b) epilepsy (b) blindness (b) blindness
D.Semantic diŒerential scales
Please ind icate your perception of the concept of mental illne ss by placing a ch eck on one of th e seven spaces on each of the co ntinua presented
below.
Mental illness
good jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj bad
slow jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj fast
weak jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj :jjjjj strong
E.Summated rating scale
The statements below express opinions or ideas about person s with men tal retardation. R ead e ach state ment carefully and circle the one
response that best corresponds with how you feel about the statement. You should work as rapidly as you can.
Key
SD : I Strongly Disagree D: I Disagree A: I Agree SA: I St rongly Agree
1. The child wh o is
mentally retarded should be in tegrated SD D A SA
into regular classes i n school .
2. I would allow my
child to attend a birthday party given SD D A SA
for a child with mental retardation.
3. A Sunday sch ool
teacher has the right to d ecide whe ther SD D A SA
to le t a c hild with mental retardati on ente r his or her class.
4. Persons with
mental retardation sometimes imagine th ey SD D A SA
have been discriminated against even when they have
been treated fairly.
F.Social distance scale
For each of th e impairments listed below, select the one le vel of soc ial contact that expresses your feelings of personal-group di stance and mark
the appropriate space. Please provide your ® rst reaction in every case. React to each impairment as a group. Do no t give your reaction to the
best or worst members of the group that you have kn own or can imagin e.
1234567
Would Would Would Would Would Would Would
marry have as have as work in have as have as debar
into close next same speaking visitors from my
group friend s door o ce acquaint- only to nation
neighbour ances only my nation
Blindness jjj jjj jjj jjj jjj jjj jjj
Amputation jjj jjj jjj jjj jjj jjj jjj
Hemophilia jjj jjj jjj jjj jjj jjj jjj
214
Measurement of attitudes towards persons with disabilities
on each scale (see table 1, part D). Osgood et al.% ) derived
three clusters of scales that they termed evaluative,
potency and activity scales. Other researchers, using
diŒerent scales and diŒerent concepts, have derived other
dimensions in addition to the original three. Kravetz et
al.% * used the semantic diŒerential method to assess the
validity of a model relating responsibility for disability
and attitudes towards persons with disabilities. The
technique has also been frequently used to measure the
change in attitudes of various groups of respondents
following a training intervention.& ! ,& "
Rating scales represent a ubiquitous attitude measure-
ment method. A probabilistic rating scale assumes that
the response to any item on the scale is not determined by
the respondent’ s latent attitude , but rather the
respondent’ s attitude is assumed to increase the prob-
ability that a particular response will be selected. The
most widely known (and abused) probabilistic rating
scale is the summated rating scale method developed by
Likert& # (see table 1, part E). The use of probabilistic
rating scales for measuring attitudes towards persons
with disabilities is discussed in detail elsewhere." ! A
deterministic rating scale assumes that the response to
any item on the scale is completely determined by the
latent attitude of the respondent. The most widely
known deterministic scaling method is scalogram analy-
sis,& $ but few examples of this method to measure
attitudes towards persons with disabilities are avail-
able.& % ±& ’ Bart and Krus& ( proposed a deterministic
measurement method they called ordering-theoretic data
analysis to generate multidimensional and nonlinear
scales. Examples of the use of ordering theory in attitude
research include a study of attitudes towards person with
disabilities,& ) an investigation of attitudes towards the
school integration and community integration of persons
with various exceptionalities& * and a study to determine
whether components of society contribute to the dis-
ablement of persons with chronic health, emotional,
mental, physical, and sensory impairments.’ !
Thurstone’ " proposed consensual location scaling as a
method in which an index value could be assigned to
each item in a set of items characterizing the attitude
referent. The researcher then constructs a scale by
selecting a small number of these items of suitable index
values covering the continuum of agreement. Bogardus’ #
used a modi® ed version of the method to develop what
is now referred to as a social distance scale. The scale
consists of seven equal-appearing intervals that the
respondent is asked to consider (see table 1, part F).
Tringo’ $ created the : Disability Social Distance Scale and
Horne’ % created the : Perceptions of Social Closeness
Scale, each of which have been used in a number of
studies of attitudes towards diŒerent groups of persons
with disabilities.’&±’ )
Direct methods of measuring attitudes are subject to a
number of threats to the validity of the obtained data." !
The mere process of responding may transform a
nonexistent attitude to an existing one, or create a
transient attitude in the respondent that the researcher,
quite erroneously, interprets as a meaningful, typical, or
stable attitude, a validity threat known as respondent
sensitization. Respondent reactivity concerns the
respondent’ s realization that his or her attitudes,
opinions, or values are being measured and the resultant
attempt to modify or distort privately held attitudes
when respondi ng to an att itude instrumen t. One may : (1)
attempt to please the researcher by providing a response
that one thinks will con® rm the researcher’ s hypothesis
(the experimenter demand eŒec t); (2) wish to give a good
impression of oneself as open-minded, sophisticated, or
enlightened (the evaluation apprehension eŒect, faking
good) ; (3) grant the attitude referent the bene® t of the
doubt when asked to make evaluative judgments (the
generosity eŒect) ; (4) deny socially undesirabl e t raits by
endorsing only those statements that one believes
represent the socially appropriate response (the social
desirabili ty bias) ; (5) try to sabotage the stu dy by
purposefull y disclosing inaccurate attitude s (the sabotage
eŒect, faki ng bad) ; (6) fail to gi ve discerning responses
because of a lack of interest in the measurement task (the
thoroughness eŒect) ; or (7) refuse to provide responses
for fear of revealing nontypical and controversial views
(the refusal bias).
Another source of invalidity of direct measurement
methods, known as response styles, arises from the
nonpurposeful attitude distorting in¯ uences of an
unrelated personality attribute of the respondent. One
may demonstrate an acquiescence response style in
which all, or almost all, of the items are answered
a rmatively. The halo eŒect refers to the tendency to
rate similarly items that one perceives to be related,
although the objective data provided in the experimental
task do not support such a correlated response. One may
select response alternatives acco rding to their location on
the response continuum. For example, a midpoint
response style represents o ne’ s at tempt to ® nd a `pl ace to
hide ’ on the response scale by selecting only the middle
or neutral value; a deviation response style is suggested
when one tends to give only unusual or uncommon
responses.
Means are available by which researchers can reduce
these threats and increase the validity of the data
obtained from direct methods.’ * Occasionally, however,
the attitude referent is so sensitive, controversial, or
215
R.F.Antonak and H.Livneh
personal that even these procedures cannot assure
con® dence in the validity of the obtained data as a true
measure of attitudes. In these cases on ingenious
measurement method known as the randomized response
technique (RRT) should be given careful consideration.
Warner( ! proposed that, rather than ask the sensitive
question directly, the researcher could provide the
respondent with a randomizing device (e.g. a spinner
common to many childhood board games) that directs
him or her either to answer the question `Do you endorse
attitude A ? ’ or the questio n ` Do you no t e ndorse
attitude A ? ’ Warner argued th at one w ould be more
likely to cooperate and provide a truthful answer if the
researcher does not know which question one answered.
It is possible to obtain an aggregate estimate of the
proportion of a population that endorses a particular
attitude on a sensitive issue using RRT sample data and
straightforward probability calculations. Antonak and
Livneh( " have provided detailed information on eight
diŒerent RRT models, as well as extensions and
modi® cations of the RRT for special purpose research
designs, inc luding the implementation of t he techniqu e in
telephone interviews and with mailed questionnaires.
i n d i r e c t m e t h o d s t o m e a s u r e a t t i t u d e s
To obviate the threats to the validity of attitude data,
measurement experts have suggested the use of one of the
indirect attitude measurement methods.( # The
respondent’ s performance on a seemin gly straightfor-
ward objective task is thought to unconsciously reveal
latent psychosocial constructs that are interpreted as
attitude. Indirect measurement methods can be
organized into four classes" " : namely, those in which the
respondents : (1) are unaware that they are being
observed or measured (nonobtrusive behavioural
observations) ; (2) are aware that they are being observed
or measured, but are unaware of or are unclear about the
purpose of the measurement situation (projective tech-
niques) ; (3) are purposefully deceived as to the true
purpose of the measurement situation (disguised tech-
niques) ; and (4) are aware of being measured but are
inactive parti cipants in the measurement process (physio-
logical methods).
Several studies have used nonobtrusive behavioural
observations in natural settings. For example, Samerotte
and Harris( $ had unaware adult shoppers approached by
an adult male confederate carrying a stack of envelopes.
The confederate either brushed against the shopper,
dropping the envelopes or merely dropped the envelopes.
The confederate assumed the roles of a person with a
dis® guring disability (an eye patch and extended facial
scars), a nondis® guring disability (a bandage wrapped
around the forearm), and with no visible disability
(neither eyepatch and scar, nor bandage around the
forearm). Pancer et al.( % used a woman who appeared to
be crossing an uncontrolled intersection in one of four
conditions. She either had a disability (i.e. used a
wheelchair) or not and either carried a bag of groceries or
not. McCauley et al.( & compared interactive behaviours
of children with hearing impairments and children
without hearing impairments in a typical classroom.
Dunlop et al.( ’ examined the social interactions between
children with and without disabilities in an integrated
preschool classroom. A recent innovative study( (
measured attitudes towards the community integration
of persons with mental retardation using a mock petition
drive concerning a proposed group home in a neigh-
bourhood. Other researchers have used nonobtrusive
behavioural observations in simulated settings. For
example, Kleck and associates( ) directed respondents
into an interviewing room where they were greeted by
either a confederate with a disability (using a wheelchair)
or without a disability. Similar studies involved the
interaction of respondents with a confederate perceived
to have epilepsy,( * a hearing impairment) ! and a mental
illness.) "
A number of di culties limit the routine use of
behavioural observations in attitude measurement. Ex-
pensive video and audio recording devices are necessary
to permit careful analyses of respondent behaviours in
experimental situations. Behavioural observations in
natural settings will require hidden recording techniques
such as those used in popular te levision shows. Obtain ing
permission to use a ® eld setting for data collection (e.g.
public school classroom, post o ce) may be di cult and
time consuming. Using and training confederates may
add to the expense of the research. If only one respondent
can be studied at a time, then considerable time may be
necessary to complete the investigation. Questions have
also been raised about the criteria for assessing the
reliability of observational data. Finally, speci® c and
local characteristics of the ® eld and laboratory settings
for the research may preclude generalizability of the
® ndings.
Projective techniques present an ambiguous stimulus
or a task for which only dim clues have been provided.
Because the likelihood of conscious response distortion
is considerably lessened, the respondent is expected to
project his or her true attitudes onto the measurement
task. This projection is then interpreted and scored by a
clinically trained professional. Rabin) # suggested that
216
Measurement of attitudes towards persons with disabilities
three main ingredients are common to all projective
techniques: (1) the essence of the stimulus is its ambiguity
coupled with the freedom of response aŒorded to the
respondent ; (2) th e responden t’ s response is noted b y its
richness and by the wide range of diverse possibilities,
thus limiting awareness of the purpose and implications
of the testing material ; and (3) the interpreter’s task is
highly complex, requiring a holistic-ideographic ap-
proach and a multidimensional analytical perspective.
Projective techniques are often classi® ed into ® ve major
categories according to the type of task or stimuli
presented or, alternatively, by the type of response that
is elicited.
Association techniques require the individual to
respond with the ® rst word or image that comes to mind
at the presentation of a stimulus. Review of the literature
on attitudes towards persons with disabilities disclosed
few studies using ambiguous image association tech-
niques) $ ,) % and no studies using word association
techniques. Construction techniques focus on a
respondent’ s out put when requ ested to construct a story
or a picture in response to a test stimulus. Despite the
potential insights that may accrue from using con-
struction techniques to study attitudes, few studies were
found that used this technique to investigate attitudes
towards persons with disabilities.)&±) ( One of the earliest
attempts to use a completion technique for the measure-
ment of attitudes towards persons with disabilities was
reported by Thurston.) ) Ford et al.) * also used a
sentence completion technique in their study of the
attitudes held by medical personnel towards persons with
a chronic illness. Choice techniques require the re-
spondent to select among several alternative items (e.g.
pictures, photographs, colours, designs, body parts)
those that appear most relevant or correct when applied
to a referent. No studies were found that investigated the
relationships of these measurement constructs to
attitudes towards persons with disabilities. Expressive
techniques permit the researcher to study the projection
of personal wishes, fears, and con¯ icts by having the
respondent form a product out of raw material. The
emphasis in expressive techniques is on the manner or
style in which the respondent forms the product and not
on the product formed. The respondent may be asked to
manipulate or interact with toys, pencils, crayons, paints,
dolls, and even other individuals (e.g. role-play or
psychodrama techniques). The use of expressive tech-
niques that allow interaction with dolls or other human-
like ® gures (some of which can be made to have various
types of visible disabilities) would seem to be particularly
valuable as a measure of attitudes of young respondents,
but no published research was found in which they were
selected to study attitudes towards persons with dis-
abilities.
The infrequent use of projective techniques for the
measurement of attitudes towards persons with dis-
abilities may be due in part to: (1) the specialized clinical
training that is required to administer and score the
instrument, and to interpret the respondent’ s projection;
(2) the additional time and expense associated with these
techniques; and (3) the lack of well-formulated hy-
potheses accounting for the relationship between per-
formance on a projective technique and speci® c mani-
festations of attitudes towards persons with disabilities.
Respondents’ reactions, products and manners of re-
sponse are vulnerable to interpersonal variables (i.e.
experimenter characteristics such as age, sex, race and
appearance), variables inherent in the experimental
situation (e.g. temperature, noise, room arrangement
and decoration, methods of instruction and adminis-
tration) and variables rel ated to the respondent’ s
psychological (e.g. stress, motivation, response set,
expectations) and physiological (e.g. medications,
hunger, fatigue) attributes. The scoring of projective
responses, even by the most highly trained and ex-
perienced clinician or psychometrician, is likely to re¯ ect
perspectives idiosyncratic to the scorer.
Disguised procedures diŒer from the projective tech-
niques in several respects. First, there is an inherent
structure to the task to be performed by the respondent
in the disguised approach. Second, a deliberate attempt
is made to direct the respondent’ s attenti on away from
the attitude for which a measurement is being sought.
Finally, whereas projective techniques assume the op-
eration of certain psychodynamic processes as the basis
for the respondent’ s reaction, the disguised procedures
make no such claim. These disguised procedures can be
classi® ed int o three catego ries : namely, those in which
the respondent is : (1) unclear about the real purpose of
the investigation; (2) led to believe that no control can be
exerted over his or her respon ses ; or (3) du ped into
believing that the purpose of the investigation is other
than what it actually is.
The researcher may provide a set of photographs of or
statements about individuals diŒering in sex, age, race,
disability and other visible characteristics that belong
together without stating what characteristic de® nes each
group. The sorting of the photographs is thought to
re¯ ect the salience of the various characteristics
portrayed in the photos. Another approach requires that
respondents select from among individuals described in
brief case vignettes those with whom they would prefer
to a liate in various social and work situations. For
example, in a study of social work students’ willingness
217
R.F.Antonak and H.Livneh
to treat alcoholics, Peyton et al.* ! had the students select
one case vignette in each of six pairs of vignettes varying
by sex, age, race and presenting complaint (depression,
job problem, marital problem, welfare problem). One of
the vignettes in each pair included an indicator of an
alcoholism problem. The assessment of humour prefer-
ence is another creative line of investigation using a
task with an unclear purpose. For example, Losco and
Epstein* " asked male and female respondents to rate on
a 15-point scale the funniness of each of a set of cartoons
depicting acts of hostility by members of one sex to
members of the other sex. The lost-letter technique,* # ,* $
another task with an unclear purpose, involves the
researcher leaving a large number of stamped envelopes
in conspicuous locations in a city. The ® rst line of the
address is a ® ctitious organization that represents the
attitude referent (e.g. Society for the Support of Persons
with AIDS). The remainder of the address is a post o ce
box in another city rented by the researcher. The
percentage of envelopes that are returned, adjusted to
account for lost, damaged, or destroyed envelopes, is
taken to be a measure of global attitudes of citizens of the
city towards the referent.
Lack-of-control procedures include the bogus pipeline
technique* % in which the respondent is led to believe that
he or she is connected to an ominous looking machine
resembling an electromyograph. The respondent is told
that the machine measures attitudes through the analysis
of autonomic indices such as involuntary muscle move-
ments or blood pressure. In fact, the respondent is not
connected to the machine and the machine measures
nothing at all. Through rigged demonstrations manipu-
lated by a confederate, the researcher validates the
machine for the respondent as an accurate measure of
attitude direction and intensity. The respondent is then
presented with questions about the attitude referent and
asked to estimate the readout of the machine. The
hypothesis is that the respondent’ s answ ers to the
questions will be a more accurate self-report of their
attitudes in these deceptive circumstances because the
respondent does not want to be second-guessed by the
machine. Only a single study* & was located that used the
bogus pipeline procedure to investigate feelings of
respondents towards a ® ctitious male interviewee who
was described as a person with a disability. Duping
procedures include the memory distortion task in which
the respondent is ® rst shown a picture or asked to read
a vignette and then requested to answer questions, some
of which concern persons or events that were not
originally presented. For example, a picture taken of a
classroom of young children during free play may be
shown to the respondent for a brief period and then
hidden. A series of questions is then asked concerning
details of the picture, among which might be the
question : `What was the child with epilepsy doing ?’ We
were unable to locate any research that used this
procedure to measure attitudes towards persons with
disabili ties. Another duping proced ure attempt s to d ivert
the participants’ attention through the presentation of a
task with ® ctitious instructions. Comer and Piliavin,* ’
for example, asked respondents to give their ® rst
impressions of two men whose photographs were
provided Ð one with a disability and one without a
disability. The cover story to the questionnaire concen-
trated on techniques used by television commercial
advertisers to form attitudes and sell their products, thus
distracting the participants from what was actually being
measured.
A unique duping procedure known as the error-choice
test is attributed to a study by Newcomb* ( of social
attitudes towards the Spanish Civil War. In addition to
a rating scale, Newcomb administered an information
test. The signi® cant correlation between scores on the
information test and scores on the attitude measure lead
Newcomb to reason that, because correct responses to
the information test required knowledge of minutiae, the
respondents guessed frequently and the direction of their
guesses was not purely random. `When in doubt,
respondents tend to answer in the direction consistent
with their own attitudes Ð and the more intense the
attitude the more intense this tendency ’ (p. 295). A
comprehensive survey of the social and behavioural
science research literature since 1948 revealed only 11
studies using the error-choice test method,* ) none of
which measured attitudes towards persons with dis-
abilities. Antonak and col leagues have created two error-
choice tests, one to measure attitudes towards persons
with epilepsy* * and the other to measure attitudes
towards persons with mental retardation."!!
Disguised procedures have limitations that are more
di cult to counteract than those of the other types of
indirect measurement methods. Tasks with no clear
purpose require consistency of presentation or else the
administration itself may become a confounding vari-
able. The additional cost and time required for im-
plementation of the bogus pipeline technique may not be
justi® ed, especially if the respondents have su cient
con® d ence in the researcher’ s clai m for the anonymity
and con® dentiality of their self-reports."!" Nederhof"!#
pointed out other limitations of the bogus pipeline
technique, including that the equipment is not trans-
portable and requires both a laboratory setting and a
confederate for its use.
There is considerable diŒerence of opinion about the
218
Measurement of attitudes towards persons with disabilities
ethics of disguised procedures. Some social psychologists
consider disguised procedures, particularly duping pro-
cedures, to be ethically dubious and decry their use.
Weschler" ! $ used an error-choice test as one of several
measures to evaluate labour mediators, using peer ratings
as the criterion for eŒectiveness. The results of his
research subsequently appeared in an industrial trade
publication and were misinterpreted as proof that
mediators were not objective in their resolution of labour
disputes. Weschler,"!% although convinced of the value of
the error-choice and other indirect methods for
measuring attitudes, argued against their use, appre-
hensive that the technique could not be kept out of the
hands of unscrupulous individuals. For example,
employers might use it to discriminate against job
applicants with unacceptable attitudes revealed on a so-
called objective information test administered as part of
an application procedure. Kidder and Campbell"!&
reiterated some of Weschler’ s con cerns but concl uded
that, in a well-designed study adhering to ethical
guidelines, disguised procedures should be considered as
one part of a multimethod approach to the measurement
of attitudes.
Physiological methods are often regarded as being the
most straightforward measures of attitudes because they
purport to measure reactions over which the respondent
has no conscious or voluntary control. Physiological
methods assume that the magnitude of the physiological
reaction is directly and positively associated with the
extent of the autonomic arousal or the intensity of the
underlying attitude."!’ The direction of the correspond-
ing attitude (e.g. pleasurable or unpleasurable, favour-
able or unfavourable), on the other hand, can not be
assumed with clarity, although it is often inferred that
the greater the magnitude of the emotional arousal, the
more unfavorable the attitude. The most widely used
procedure to measure autonomic activation had been the
electrical conductiveness (or resistance) of the skin. It is
assumed that when intense aŒective reactions are
experienced (i.e. strongly held negative attitudes are
aroused in the respondent) they are accompanied by a
physiological activation lowering the level of skin
resistance. Kleck et al.( ) reported that respondents who
interacted with a confederate who used a wheelchair
showed lowered skin resistance than did respondents
who interacted with a confederate who did not use a
wheelchair. Zych and Bolton" ! ( studied respondents’
reactions to a variety of photographs using the method.
Wesolowski and Deichman"!) monitored skin conduc-
tance while respondents viewed three diŒerent scenes:
namely, persons with a vi sible physical disability, persons
without a physical disability, and neutral scenery.
Hess" ! * considered pupil dilation elicited by emotional
arousal to an observed referent as a sign of favourable,
pleasant, or positive reaction, while pupil constriction
signaled an unfavourable, aversive, or negative response.
Although these researchers have studied a wide range of
reactions to aversive and abhorrent stimuli (e.g. corpses),
they have not attempted to study pupillary responses to
persons with visible disabilities. Measures of accelerated
or decelerated heart rate, pulse and blood pressure are
thought to be indicators of one’ s level of state anxiety.
Other researchers have hypothesized that respondents
showing more negative attitudes towards persons with
disabilities would manifest more state anxiety, as
measured by heart rate, in interactions with them.""!,"""
Gargiulo and Yonker" " # found a statistically signi® cant
diŒerence between novice and experienced regular and
special educators in the magnitude of change of pulse
rate after viewing slides of persons with various disabling
conditions. Despite their popularity as physiological
signals that are associated with psychological processes,
the electroencephalogram (record of intrinsic electrical
activity associated with the brain) and the electro-
myograph (record of intrinsic electrical activity related to
muscle contraction) apparently have not been used as
measures of attitudes towards persons with disabilities.
VanderKolk" " $ u sed an electronic device that purportedly
measured the respondent’ s l evel of stress by chartin g his
or her vocal modulations from an audiotape recording.
Respondents demonstrated signi® cantly higher levels of
stress when discussing persons with physical, mental and
emotional disabilities as compared to a normal situation
stress level. More recently, Gouvier et al." " % demonstrated
the utility of analysing verbal interaction patterns (e.g.
word counts, frequency of interrogatives, use of locator
words) to diŒerentiate interactions with individuals with
and without physical disabilities (persons using wheel-
chairs).
Although physiological methods represent a unique
and creative line of measurement, they may not always
be feasible in many attitude research situations. Physio-
logical methods require a laboratory setting, costly
equipment, careful calibration and control of the
measuring instrument, and technical expertise to obtain
and interpret the response data. There is also the
di culty of determining with clarity the direction of the
attitude corresponding to a physiological response.
Shapiro and Crider""& have noted, moreover, that
characteristics of the experimenter have been shown to
be confounding sources of variation in psycho-
physiological measurement. Cognitive factors (e.g. the
instructions provided to the respondent) may also
in¯ uence physiological outcomes. Finally, unrelated
219
R.F.Antonak and H.Livneh
aŒective factors (e.g. anxiety associated with the lab-
oratory setting itself, stress induced by the testing
situation) may also contaminate the results obtained.
For these reasons, physiological measures of attitudes
should be adopted cautiously and considered as ex-
perimental methods until additional data on their
reliability and validity can be accumulated and analysed.
Implications and recommendations for rehabilitation
practise
Our goal in writing this paper was to assist those
seeking to understand attitudes towards persons with
disabilities by providing information on the measure-
ment of these attitudes. Understanding the formation,
nature, structure and correlates of these attitudes must
precede the development and implementation of in-
tervention strategies designed to improve attitudes and
to ultimately remove barriers to serving persons with
disabilities. The ® nal part of this paper begins with
implications of the measurement of attitudes towards
persons with disabilities for rehabilitation researchers.
Global impl ications and suggestio ns±where we permitted
ourselves a certain degree of inferential freedom±are then
directed at rehabilitation educators and practitioners.
The concept of attitudes towards persons with dis-
abilities is complex, and the measurement of these
attitudes is not a simple task. Far too often researchers
investigating attitudes have been responsible for per-
petuating the myth that this area of research is simple
and, therefore, simplistic. It is easy to create a useless
instrument and to collect useless data. The investigation
of attitudes towards persons with disabilities requires
innovative experimental methods and psychometrically
sound instruments that are reliable, valid and multi-
dimensional. Without such instruments, it will not be
possible to obtain conclusive answers to important
research questions concerning the relationship between
these attitudes and the acceptance and integration of
persons with disabilities into society. Listed below are
recommendations for the design of innovative research
investigations and the development of sound attitude
measurement instruments :
(1) Because of the excess number of scales purporting to
be valid measures of attitudes towards persons with
disabilities in general (presently more than 40), and
towards grou ps of persons with speci® c disabil ities in
particular, researchers may wish to concentrate their
eŒorts on re® ning, revising, updating, and
revalidating older scales rather than on developing
new scales ;
(2) When selecting an existing direct attitude measure,
researchers are advised to ® rst consider multi-
dimensional scales since the preponderance of em-
pirical evidence supports the multidimensional
nature of attitudes towards persons with disabilities;
(3) Researchers should report, whenever possible, in-
dices concerning the psychometric properties of the
scales used (reliability, validity, item characteristics,
scale characteristics), including the values reported
for the original scales and the values calculated from
the data in their investigation;
(4) When using existing scales or adapting scales for
speci® c research purposes, researchers should specify
the attitude referent (e.g. a person with a hearing
impairment, persons who are mentally retarded), as
well as be clear about whether they are focusing on
attitudes towards a speci® c group of persons with
disabilities (e.g. people who are visually impaired) or
the disability per se (e.g. blindness, visual impair-
ment) ;
(5) Because of their relative immunity from the del-
eterious eŒects of confounding and respondent
sensitization, indirect attitude measurement methods
are well-suited for examining attitudes towards
persons with disabilities and should be given serious
consideration for use by attitude researchers ;
(6) Because of the inconsistency in research ® ndings on
the relationship between verbal or written expressed
attitudes and overt (e.g. observed) behaviours,
researchers should investigate the nature of these
relationships more carefully; and
(7) To further clarify the nature, structure and correlates
of attitudes towards persons with disabilities,
researchers should continue to investigate the
relationships between attitude components (e.g.
aŒective, cognitive, behavioural) and sets of socio-
demographic (e.g. age, gender, educational level,
profession), personality (e.g. anxiety, hostility, stress
level, locus of control) and situational variables (e.g.
social context, family dynamics).
Implications for rehabilitation education and training,
derived from the measurement of attitudes towards
persons with disabilities and from the more generic
literature on attitudes towards persons with disabilities,
focus mainly on assisting students and trainees to gain a
deeper awareness of their attitudes towards groups of
persons with disabling conditions and modi® cation of
these attitudes when necessary. Many students, upon
entering the ® eld of rehabilitation, possess only minimal
awareness of their attitudes towards and actual
behaviours in the presence of persons with disabilities.
220
Measurement of attitudes towards persons with disabilities
These early attitudes often convey stereotypical thinking,
generalization of certain characteristics and attributes to
persons with similar conditions and even to all persons
with disabilities, and periodically attribute unrelated
special negative or posit ive characteristics and abil ities to
people who are disabled (the spread phenomenon). Early
training geared towards enhancing awareness of
students’ and trainees’ attitudes towards persons with
disabilities could dispel such notions. In a similar vein,
academic institutions and related training programmes
should engage in direct eŒorts to consciously modify
students’ and trainees’ attitudes towards persons with
disabilities. Because research generally supports the
notion of attitudinal-behavioural consistency, it may be
argued that fostering positive attitudes towards persons
with disabilities in rehabilitation trainees cou ld be linked
to more vigorous eŒorts on their part (e.g. increased
optimism, higher levels of motivation, more time spent
on planning and providing rehabilitation services) to
serve persons with disabilities.
Rehabilitation educators may ® nd incorporating the
topic of measurement of attitudes towards persons with
disabilities into the core curriculum of their programme
a w orthwhile and useful en deavour. In courses con cerned
with psychosocial aspects of disability, rehabilitation
research methods and even human growth and de-
velopment, knowledge of attitude measurement could
provide students with the practical know-how of various
psychometric considerations, measurement methods,
and speci® c scales in the ® eld of disability and attitude
studies. Due to the plethora of scales purporting to
measure attitudes towards persons with disabilities and
of data suggesting diŒerentially reported attitudes across
the lifespan, students enrolled in coursework focusing on
human development issues may ® nd this topic par-
ticularly useful.
Implicati ons for practici ng rehabilitati on professionals
derived from the studies on the nature and measurement
of attitudes towards persons with disabilities focus
primarily on awaren ess of one’ s own at titudes tow ards
persons with disabilities (e.g. preferences of and
prejudices towards speci® c groups, issues of counter-
transferrence) and recognition of the broad range of
correlates that are linked to attitude manifestation. The
literature on the social position (i.e. preference ranking)
of persons with disabilities indicates that a rather
constant trans-situational eŒect exists for social rank
ordering of speci® c groups of persons with disabilities.
This suggests that when working with diŒerent groups of
persons with physical, psychiatric, intellectual and
behavioural disabilities, the practitioner typically faces
diŒerent issues of stigma and prejudice as manifested in
one’ s social envi ronment. Rel atedl y, i t has b een often
suggested that practitioners must be aware of their own
attitudes towards persons with disabilities. It has been
argued that the unconscious over-identi® cation with a
client with a particular disability may elicit powerful
aŒective and defensive reactions (i.e. counter-
transferrence) in the practitioner that very likely will
negatively in¯ uence the process of counselling." " ’
Research has consistently indicated that a variety of
sociodemographic, experiential and personality variables
of the observer (i.e. the person without a disability), in
addition to interactive contexts (i.e. social settings) and
disability-speci® c factors (e.g. visibility, degree of
severity) strongly eŒect interaction between persons with
and without disabilities. It is, therefore, incumbent upon
rehabilitation practitioners to acknowledge these con-
tributin g factors when d esigning programmes fo r attitude
modi® cation in any of a variety of community, school, or
work settings. Researchers and practitioners must be
cognizant not only of the need to secure participants’
informed consent, respecting participants’ rights to
con® dentiality and privacy, but also appreciate the need
to continue the investigation of attitudes towards persons
with disabilities±their origins, developmental aspects,
structure, and measurement±in order to ensure their
ultimate change.
References
1 Antonak RF, Fiedler CR, M ulick JA. Misconceptions related to
mental ret ardation. Mental Re tardation 1989 ; 27 : 91±97.
2 Beattie J, Anderson RA, Antonak RF. Modifying attitudes of
preservice educ ators toward persons with disabilities and the
integration of students with disabilities into regular cl assrooms.
Journal of Psychology 1997 ; 131 : 1±17.
3 Furnham A, Th ompson R. Actual and perceived attitudes of
wheelc hair users. Counselling Psychology Quarterly 1994 ; 7: 35±53.
4 Gilbride DD. Parental attitu des toward their c hild wit h a disabil ity:
Implications for rehabilitation counsellors. Rehabilitation
Counselling Bulletin 1993; 36 : 140±150.
5 Holaday M, Wolfson A. A ttitude s toward c hildren with severe
burns. Rehabilitation Counselling Bulle tin 1997; 41 : 54±69.
6 McCarthy H. Attitudes that aŒect empl oyment op portunitie s for
persons with disabilitie s. In : Y uke r HE (ed.) Attitudes Toward
Persons with Disabilities. New York, NY: Sp ringer, 1988: 246±261.
7 Siperstein GN, Bak JJ, O’ Keefe P. Relatio nship betwee n child ren’s
attitudes toward and their social acceptance of mentally re tarded
peers. American Journal of M ental Retardation 1988 ; 93: 24±27.
8 Towner A G. Modifying attitudes toward the handicapped: A
review of th e literature and meth odol ogy. In: Jone s RL (ed.)
Attitude and att itude c hange i n special educati on:Theory and
practice. Reston, VA: Cou ncil for Exceptio nal Children, 1984 :
223±257.
9 Yuker H E. Prefac e: Attitudes toward persons with d isabilities :
progress an d prospec ts. In : Yuk er HE (ed.) Att itudes toward
persons with disabiliti es. New York, NY : Springer, 1988 : xi±xiv.
10 A ntonak R F, Livneh H. The Measurement of Attitudes Toward
People Wit h Disabilities :Methods,Psychometrics,and Scales.
Spring® eld , IL : C C Thomas, 1988.
11 Livne h H, Antonak RF. Ind irect met hods to measure at titudes
221
R.F.Antonak and H.Livneh
toward pe rsons with disabilities. Rehabilitation Education 1994 ; 8:
103±137.
12 O skamp S. Attitudes and Opinions (2nd ed). Engl ewood Cl iŒs, NJ :
Prentice Hall, 1991.
13 Triand is HC. Attitude and atti tude change. New York, NY : John
Wiley & Sons, 1971.
14 Jo nes RL, Guskin SL. Attitu des and attitude change in special
educat ion.In : J ones RL (e d.) Attitude and Att itude Change i n Spec ial
Education :Theory and Practice. R eston, VA : Council for Ex-
cepti onal Child ren, 1984 : 1±20.
15 Mu eller DJ. Measuring social attitudes:A handbook f or researchers
and practitioners. New York, NY : Teachers Co llege Press, 1986.
16 Do vidio JF, Faz io RH. Ne w tech nologies for th e direct and ind irect
measurement of attit udes. In: Tan ur J M (Ed.) Questions About
Questions :Inquiries into the Cognitive Bases of Surveys. New Y ork,
NY: Russell Sage Found ation, 1992 : 204±237.
17 Dawes RM, S mith TL. Attitud e and opini on measurement. In :
Lindzey G and Aronson E (eds) The Handbook of Social
Psychology :Vol.I.Theory and Method (3rd ed .). New Yo rk, NY :
Rando m House, 1985 : 509 ±566.
18 Horne MD. How Attitudes are Measured : A Review of Investi-
gations of Professional, Peer, and Parent Attitudes Toward t he
Handicapped. Princeto n, NJ : Clearingh ouse on Tests, Measure-
ment, and Evaluation, 1980.
19 R ajecki DW. Attitudes (2nd ed). Su nderland , MA : Sinaue r, 1990.
20 Caruso DR, Hodapp RM. Perceptions of mental retardation and
mental illness. American Journal of Mental Retardation 1988 ; 93 :
118±124.
21 Com er RJ, Pil iavin JA. Th e eŒects o f physical d eviance upo n face
to face i nteraction : the o ther side . Journal of Personality and Social
Psychology 1972 ; 23 : 33±39.
22 Lin kowski DC, Jaq ues ME, Gaier EL. Re action s to disabilities: a
thematic analysis. Journal of Social Psychology 1969 ; 77 : 201±214.
23 Lyth M. Emp loyers ’ attitudes to the empl oyment of t he disable d.
Occupational Psychology 1973 ; 47: 67±70.
24 Philips GB. An exploration of employer attitudes concerning
employme nt opp ortunit iesfor d eaf p eople .Journal of Rehabili tation
of the De af 1975 ; 9: 1±9.
25 S alend SJ, Giek K A. Indepe ndent livin g arrangemen ts for indi-
viduals with mental retardation : the land lord’s perspe ctive. Mental
Retardation 1988 ; 26 : 89±92.
26 Me rcer JR. Patterns of family crisis related to reacceptance of the
retardate. American Journal of Me ntal De® ciency 1966; 71: 19±32.
27 Barngrover E. A study o f edu cator’s pref erences in spe cial educ ation
programs. Exceptional Children 1971 ; 37 : 754±755.
28 Mallory BL, Herrick SC. Investigati ng the impact of co m-
muniti zation : issues in research me thod ology. In : Anton ak RF an d
Mulick JA (eds.) Transitions i n Mental Retardation :Vol. 3. The
Community Imperative Revisited. Norwood, NJ: Ablex Publishing
Corporation ., 1987 : 43 ±71.
29 Barsch R H. Th e h andicappe d ranking scale among parents of
handicappedchildren.American Journal of Public Health 1964 ; 54 :
1560±1567.
30 Orlansky MD. Ac tive le arning an d student attitudes toward
exceptional children. Exceptional Children 1979; 46 : 49±52.
31 A broms KI, Kodera. TL. A cceptanc e hierarchy of hand icaps:
validation of Kirk’ s state ment ` Special edu cation o ften begins
where medicine stops ’. Journal of Learning Disabilities 1979 ; 12 :
15±20.
32 R ichardson SA , Goodman N, Hastorf A, Dornbusch S. Cultural
unifo rmity in react ion to physical d isabilitie s.Am erican Sociological
Review 1961 ; 26: 241±247.
33 Chigier E, Chigier M. Attitudes to disability of children in the
multi-cultural society of Israel. Journal of Health and Social
Behaviour 1968 ; 9: 310±317.
34 Stephenson W. The Study of Behaviour :Q-Technique and its
Methodology. Chic ago, IL : University of Chicago Press, 1953.
35 Mc Keown B, Thomas D. Q Methodology. Beverly Hil ls, CA: S age,
1988.
36 Barker DG. Concepts of disabilities. Personnel and Guidance
Journal 1964 ; 43 : 371±374.
37 S haver PR , S cheibe E. Transfo rmation o f soci al id entit y: a stud y o f
chronic me ntal patients and college vo lunteers in a summe r camp
setting. Journal of Psychology 1967 ; 66 : 19±37.
38 Go ttlieb J. Publi c, peer, an d profe ssional attitu des toward mentally
retarded pe rsons. In : Begab MJ and Rich ardson SA (eds.) The
Mentally Retarded in Socie ty:A Social Science Perspective.
Baltimore, MD : University Park Press, 1975 : 99±125.
39 A ltman BM. S tudies of attit udes to ward t he h andicapp ed : the need
for a new direction. Social Problems 1981 ; 28: 321±337.
40 MacM illan DL, Morrison GM. Soc iometric research in special
educat ion.In : Jo nesR L (ed.) Attitude and Attitude Change i n Special
Education :Theory and Practice. R eston, VA : Council for Ex-
cepti onal Child ren, 1984 : 70±92.
41 Go ugh HG. The Adjective Check List as a personality assessment
research scale. Psychological Reports 1960 ; 6: 107±122.
42 Go ttlieb J , Corman L, Curci R. Attit udes toward mentally re tarded
children. In: Jon es RL(ed.) Attitude and Attitude Change in Special
Education :Theory and Practice. R eston, VA : Council for Ex-
cepti onal Child ren, 1984 : 143±156.
43 Gra S, Minnes PM. Att itudes o f p rimary school chil dren to ward
the physic al appearance and label s associate d with Do wn syndrome.
American Journal on Me ntal Retardation 1988 ; 93: 28±35.
44 A nderson R J, An tonak RF. Th e in¯ uenc e of at titudes and con tact
on reactions to persons wi th physical and speech disabilities.
Rehabilitation Counselling Bulletin 1992 ; 34: 240±247.
45 Jo nes RL, G ottfried NW, Owens A. The social distance o f the
exceptional: a study at the high school level. Exceptional Children
1966 ; 32 : 551±556.
46 Wic as EA, Carluccio LW. Attit udes of co unsellors toward three
handic app edclien tgroups. Rehabilitati on Counselling Bulle tin 1971 ;
15 : 25±34.
47 Janicki MP. At titudes of health professionals toward twelve
disabilities. Perceptual and Motor Skills 1970 ; 30 : 77±78.
48 O sgood C E, S uci G J, Tannenbaum P H. The Measurement of
Meaning. Urbana, IL : University of Illinois Press, 1957.
49 Kravetz S, Katz S , A lbez D. Att itudes t oward Israeli war veterans
with disabilit ies: Combat versus noncombat mil itary service an d
responsibility for disability. Rehabilitation Counselling Bulletin
1994 ; 37 : 371±379.
50 R ees LM, Spreem D, Harnade k M. Do attitudes towards persons
with handicaps really shift over time? Comparison between 1975
and 1988. Mental Retardation 1991 ; 29 : 81±86.
51 S adlick M, Penta FB. Changin g nu rses ’ attit udes toward q uadri-
plegics t hrough t he use of t elevision . Rehabilitation Literature 1975 ;
36 : 274±278.
52 Lik ert R. A t echniqu efo r the me asurement o f attitude s. Archives of
Psychology 1932 ; 140.
53 Guttman L. A basis f or scaling qualitative data. American
Sociological Review 1944 ; 9: 139±150.
54 Cohe n JS. Empl oyer attitu des toward hiring mentally retarded
individuals. American Journal of Mental De® ciency 1963 ; 67 :
705±713.
55 Y amamoto K, Dizney HF. Rejec tion of the mentall y ill : a study of
attitudes of student teachers. Journal of Counselling Psychology
1967 ; 14 : 264±268.
56 Y amamoto K, Wie rsma J. Rejection of self and of d eviant ot hers
among stude nt teachers. The Journal of Special Education 1967 ; 1:
401±408.
57 Bart WM, Krus D. A n ordering-theoretic method to determine
hierarchies among items. Educational and Psychological Measure-
ment 1973 ; 33 : 291±300.
58 A ntonak R F. An ordering-theoretic analysis of attitudes toward
disabled persons. Rehabilitation Psychology 1979 ; 26 : 136±144.
59 Antonak RF. A hierarchy of attitud es toward exceptionality.
Journal of Special Education 1980 ; 14 : 231±241.
60 Antonak R F. Societal factors in disablement. Rehabilitation
Counselling Bulletin 1985; 28 : 188±201.
222
Measurement of attitudes towards persons with disabilities
61 Thu rstone LL. Attitudes can be measured. American Journal of
Sociology 1928 ; 33 : 529±554.
62 Bogardu s ES. A social di stanc e scale. Sociology and Social Research
1933 ; 17 : 265±271.
63 Tringo JL. The hierarchy of p reference toward disabilit y groups.
Journal of Special Education 1970 ; 4: 295±306.
64 Ho rne MD. Cultu ral eŒect on attit udes toward label s. Psychological
Reports 1978 ; 43 : 1051±1058.
65 Harasymiw SJ, Horne MD, Lewis SC. A longitudinal study of
disability group accep tance. Rehabilitati on Literature 1976; 37 :
98±102.
66 Leyser Y , Ab rams PD. Teacher attit udes toward normal and
exceptional groups. Journal of Psychology 1982 ; 110 : 227±238.
67 Sigelman CK. Social distance from stigmatized groups : False
consen susan d false un iquene sseŒec ts on respon ding. Rehabilitation
Psychology 1991 ; 36 : 139±151.
68 Jo nes GE, Stone DL. Perceived discomfort associated with working
with persons with varying disabilities. Perceptual and Motor Skills
1995 ; 81 : 911±919.
69 Wainer H, Braun HI. Test Validity. Hillsdale, NJ: Lawrence
Erlbaum Associ ates, 1988.
70 Warner SL. Ran domized respon se: A survey tech nique for
eliminating evasive answer bias. Journal of the American Statistical
Association 1965 ; 60 : 63±69.
71 A ntonak RF, Li vneh H. Rando mized respo nse techniq ue : a review
and p roposed extension to d isability attit ude studies. Genetic,
Social,and General Psychology Monographs 1995 ; 121 : 97±145.
72 Himme lfarb B. Th e measurement of attitud es. In : Eagley AH an d
Chaiken S (eds) The Psychology of Attitudes. O rlando, FL :
Harcourt Brace Jovano vich, 1993 : 23±87.
73 S amerotte GC, Harris MB. Some factors in ¯ uencing h elping: the
eŒects of hand icap, responsibility, and re questing h elp. Journal of
Social Psychology 1976 ; 98 : 39±45.
74 Pancer SM, Adams DA, Mollard D, Solsberg D, Tammen L.
Perceived distinc tiveness of the hand icapped . Journal of Social
Psychology 1979 ; 108 : 275±276.
75 Mc Cauley RW , Bruinink s RH, Ke nnedy P. Behavioral inte ractions
of h earing impaired chi ldren in regular classrooms. Journal of
Special Education 1976 ; 10 : 277±284.
76 Dunlop KH, St oneman Z, Cantrell ML. S ocial interaction of
exceptional and other children in a mainstreamed preschool
classroom. Exceptional Children 1980 ; 47 : 132±141.
77 Zsambok J, Hammer D, Rojahn J. Put your money where your
mouth is : d irect and indirect measures of attitude toward
community integration. American Journal on Mental Retardation
1999 ; 104 : 88±92.
78 Kle ck R, On o H, Hastorf A H. The eŒec ts of physical deviance
upon face-to -face interacti on. Human Relations 1966; 19 : 425±436.
79 Kle ck R, Bu ck PL, Go ller WL, London RS, PfeiŒer JR, Vukcevic
D. EŒe ct o f stigmatizi ng c ondition s on t he use of personal space .
Psychological Reports 1968 ; 23 : 111±118.
80 Wil son ED. A comparison of the eŒects of deafne ss simulation and
observation upo n attitude s, anxiety, and behavior manifested
toward the deaf. Journal of Special Education 1971 ; 5: 343±349.
81 Farina A , Sherman M, Al len JG. R ole of ph ysical abnormaliti es in
interpersonal perception and behavior. Journal of Abnormal
Psychology 1968 ; 73 : 590±593.
82 R abin AI. Assessment with Projective Techniques. New Yo rk, NY :
Springer, 1981.
83 Fishe r S, Cleve land SE. Body Image and Personality. Prin ceton , NJ :
Van Nostrand, 1958.
84 Fishe r S, Cleveland SE. Body Image and Personality (2nd ed) . New
York, NY : Dover, 1968.
85 Bill ings HK. An exploratory study of the attitudes of noncrippled
children toward crippled children in three selected elementary
schools. Journal of Experimental Education 1963 ; 31 : 381±387.
86 Mo ed G, Wright B, Feshbach S, Sandry M. A pi cture story test for
use in physical disability. Perceptual and Motor Skills 1963 ; 17 :
483±497.
87 N oonan J R, Barry JR, Davis HC. Person ality d eterminants i n
attitu des toward visible disability. Journal of Personality 1970 ; 39 :
1±15.
88 Th urston JR . A proce dure for evaluatin g parental atti tudes to ward
the h andicapped. Americ an Journal of Mental De® c iency 1959 ; 64 :
148±155.
89 Fo rd A B, Liske RE, O rt RS . Reacti ons of physicians and me dical
students to ch ronic illness. Journal of Chronic Diseases 1962 ; 15 :
785±794.
90 Pe yton S, Chadd ick J, Go rsuch R. Willin gness to treat alc oholics :
A study of graduate social work students. Journal of Studies on
Alcohol 1980 ; 41 : 935±940.
91 Lo sco J, Epstein S. Humor preference as a subt le measure of
attitudes toward the same and the opposite sex. Journal of
Personality 1975 ; 43 : 321±334.
92 M ilgram S, Mann L, Harter S. The lo st-letter technique of social
research. Public Opinion Quarterly 1965 ; 29 : 437±438.
93 S immons CH, Zu mph C. The lost let tert echniq uerevisited. Journal
of Applied Social Psychology 1983 ; 13 : 510±514.
94 J ones EE, Sigall H. The b ogus pip eline te chnique : A n ew paradigm
for m easuring aŒect an d atti tude. Psycholo gical Bulleti n 1971 ; 76 :
349±364.
95 Carver CS , Glass DC, Katz I. Favorable evaluati ons of Black s and
the hand icapped : p ositive prejud ice, un consci ous den ial, or social
desirability? Journal of Applied Social Psychology 1978 ; 8: 97±106.
96 Co mer RJ, Piliavin JA. As othe rs see us: attitudes of physically
handicapped and normals toward own and oth er groups.
Rehabilitation Literature 1975 ; 36 : 206±221.
97 N ewcomb TM. The in¯ uence of attit ude c limate upon some
determinants of information. Journal of Abnormal and Social
Psychology 1946; 41 : 291±302.
98 A ntonak RF, Livneh H. Direct an d indirect methods to measure
attitudes toward p ersons with disabilities, with an exegesis of the
error-choice method . Rehabilitation Psychology 1995 ; 40 : 3±24.
99 A nton ak RF, Livne h H. Develo pment, psych ometric analysis, and
validation of an error-choice test to measure att itudes t oward
persons with epilepsy. Rehabilitation Psychology 1995 ; 40 : 25±38.
100 Ant onak RF. Development and p sychometric analysis of an
indirect measure of attitudes toward persons with men tal re-
tardation using the error-c hoice method. Mental Retardation
1994 ; 32 : 347±355.
101 Hansen W B, M alotte CR, Fiel ding JE. The bogus pipeline
tech nique revisited : The use of t he th reat of detectio n as a means
of i ncreasing self-rep orts of tobacc o use. Journal of Applied
Psychology 1985; 70 : 789±792.
102 Nederhof A J. Methods of coping with social desirability bi as: A
review. European Journal of Social Psychology 1985 ; 15 : 263±280.
103 Weschler IR. The personal factor in l abor mediati on. Personnel
Psychology 1950; 3: 113±133.
104 Weschler IR. Problems in the use of indirect methods of at titude
measurements. Public Opinion Quarterly 1951; 15 : 133±138.
105 K idder L H, Camp bell D T. The in direct t esting of social at titude s.
In : S ummers GF (ed.) Attitude me asurement. Chicago, IL: R and
McNally, 1970; 333±385.
106 Cacioppo JT, Tassinary LG. Inferring psycholo gical signi® cance
from ph ysiological sign als. Am erican Psychologist 1990 ; 45 : 16±28.
107 Zyc h K, Bolton B. Galvanic skin respon ses and c ognitive attit udes
toward disabled persons. Rehabilitation Psychology 1972 ; 19 :
172±173.
108 W esolowski MD, Deich man J. Physiologic al activity an d attitude s
toward d isabled persons. Re habilitation Counseling Bulle tin 1980 ;
23 : 218±226.
109 H ess EH. A ttitude an d pupil size . Scienti® c American 1965; 212 :
46±54.
110 Marinelli RP, Kelz JW. Anxie ty and atti tudes toward visibly
disabled persons. Rehabilitation Counseling Bulletin 1973 ; 16 :
198±205.
111 Marinelli R P. State anxiet y in interactions with visibly disabled
persons. Rehabilitation Counselling Bulletin 1974; 18 : 72±77.
223
R.F.Antonak and H.Livneh
112 G argiulo RM , Y onker RJ. Assessing teachers ’ at titudes to ward
the handi capped : a methodological investigation. Psychology in
the Schools 1983 ; 20 : 229±233.
113 VanderKolk CJ. Physiological and self-reported reactio ns to the
disabled and deviant. Rehabilitation Psychology 1976 ; 23 : 77±83.
114 G ouvier WD, Coon R C, Tod d ME, Fu ller KH. Ve rbal in teractions
with individuals presenting with and without physical disability.
Rehabilitation Psychology 1994 ; 39 : 263±268.
115 Shapiro D, Crider A. Psycho physiological app roaches in social
psychol ogy. In : Lind zey G and Aronson E (eds) The Handbook of
Social Psychology :Vol. 3. The Individual i n a Social Contex t (2nd
ed.) Read ing, MA: Addison-Wesley, 1968: 1±49.
116 S iller J. Attitud es toward the physic ally di sabled. In: Jo nes RL
(ed.) Attitude and Attitude Change in Special Education :Theory
and Practice. Re ston, VA : Cou ncil fo r Excep tional Children,
1984 : 184±205.
224
CopyrightofDisability&RehabilitationisthepropertyofTaylor&FrancisLtdandits
contentmaynotbecopiedoremailedtomultiplesitesorpostedtoalistservwithoutthe
copyrightholder'sexpresswrittenpermission.However,usersmayprint,download,oremail
articlesforindividualuse.
