Psychological Reports, 1987, 60, 743-746.
@ Psychological Reports 1987
THE COMPANION ANIMAL BONDING SCALE:
INTERNAL RELIABILITY A N D CONSTRUCT VALIDITY
ROBERT H. PORESKY,1 CHARLES H E N D R I X ,
Departmend of Human Development and Family Studies
Kansas State University
JACOB E. MOSIER, A N D MARVIN L. SAMUELSON
College of Veterinary Medicine
Kansas State university
Summmy.-The literature o n the relationship of compsnion animals and
children shows only a weak effect of human-animal bonding o n child develop
ment. The use of "pet ownership" or cohabiration rather than the relation-
ship o r inreraction between the child and the animal as a measure of bonding
appears to be a serious and limiting deficiency, which impaired the empirical
evidence concerning the development and effects of human-animal bonding.
The Companion Animal Bonding Scale is an 8-item behavioral scale describing
the extent of child-animal activities. T h e scale was administered by question-
naire with a childhood focus and a contemporary focus to 121 high school and
college srudents. T h e Gonbach alpha estimates of internal reliability were
0.82 and 0.77, respectively. Consuucr validity was indicated by significant
correlations between scores o n the Pet Attitude Scale and the childhood and
contemporary bonding scale of .39 and .40, respectively.
The effects of pets (companion animals) on human development have
received renewed interest recently. One issue of Marriage and Family Review
( 1 ) was dedicated to review articles on effects of pets, but in many recent
papers a need for tighter research on both the bonding process and its effects
across the lifespan was noted.
Broad discussion of presumed effects of companion animal is available
(3, 5 ) . Levinson concluded that "closeness to animals can reduce alienation"
(4, p. 1031), based upon the argument that people and companion animals
evolved together over rhousands of generations. Children's empathy, self-
esteem, self-control, and autonomy could be promoted by raising pets ( 4 ) .
Hyde, Kurdek, and Lason ( 2 ) found pet owners higher on measures of
social sensitivity and interpersonal trust than nonowners and greater effects
were associated with sex of the respondent. A correlation of low magnitude
( Y = .11) between pet ownership and young adolescents' self-esteem was noted
( 1 ) . Comparison of pet histories of teenagers in juvenile institutions with
those of noninstitutionalized youth ( 6 ) showed similar "Merely having a
'Contribution 87-195-J, Kansas Agricultural Experiment Station. Manhattan. K S 66506.
Request reprints from Dr. R. H. Poresky, Department of Human Development and
Family Srudies, Justin Hall, Kansas State University, Manhattan, KS 66506.
744 R. H. PORESKY, AL.
special pet played no part in whether or not a child was eventually institu-
tionalized" (6, p. 7 5 ) . The lack of distinction between "pet ownership" and
companion animal bonding may be why only effects of low magnitude were
found.
Although "pet ownership" has often been the operational definition o f companion-
animal bonding, the key factor should be the quality of the relationship or social inter-
action between the child and pet. The Companion Animal Bonding Scale was developed
to provide a sensitive scale for the assessment of self-reported behavior indicative of the
esrablishmenr of a bond between a person and an animal. Scientists at Kansas State
University i n the fields of child development and veterinary medicine collaborated in
the development of the following items o n the retrospective childhood Companion Animal
Bonding Scale to provide a betcer operational definition of the human-animal bond:
1. H o w often were you responsible for your companion animal's care?
-Always -Generally -Often -Rarely -Never
2. H o w often did you clean u p after your companion animal?
-Always --Generally -Often -Rarely -Never
3. H o w often did you hold, stroke, or pet your companion animal?
-Always -Generally -Often -Rarely -Never
4. H o w often did your companion animal sleep in your room?
-Always -Generally -Often -Rarely -Never
5. H o w often did you feel that your companion animal nras responsive to you?
-Always -Generally -Often -Rarely -Never
6. How often did you feel that you had a close relationship with your companion animal?
-Always -Generally --Often -Rarely -Never
7. H o w often did you travel with your companion animal?
-Always -Generally -Often -Rarely -Never
8. How often did you sleep near your companion animal?
-Always -Generally -Often -Rarely -Never
The contemporary scale used rhe above items in the present tense and the childhood
scale used the past tense with regard to the animal each respondent personally identified
as mosc important during his childhood. T h e scale total score is the sum of the item
responses with always = 5, generally = 4 , often = 3, rarely = 2, and never = 1.
These items were administered to a group of students with the Pet Attitude Scale ( 7 ) .
One hundred twenty-one students ranging in age from 14 to 47 yr. (mean age
= 26.4 yr., SD = 4.5) responded to the questionnaire. They included high school
age srudencs (2.8% ) and college undergraduate (77.3 % ) and graduate ( 19.8% )
students. T h e college students were enrolled in 22 majors at Kansas State University.
There were 5 3 women and 68 men.
RESULTSAND DISCUSSION
An SPSS-X Reliability analysis of the internal reliability of the scale yielded
a Cronbach alpha of 0.77 for the eight-item childhood scale and 0.82 for the
contemporary scale. The corrected item-total correlations are given in Table
1 for boch forms of the scale.
Scale Characteristics
There were 113 sets of responses on the childhood scale. The childhood
COMPANION ANIMAL BONDING SCALE 745
TABLE 1
VALUESOF CRONBACH'SALPHA OF THE COMPANION ANIMALSCALES
IF SPECIFIC
ITEMS WEREDELETED AND M A N D SD
Item Childhood Scale Contemporary Scale
a M SD a M SD
1. Responsible for care .75 3.70 1.13
2. Clean u p .75 3.36 1.31
3. Hold, stroke, or pet 7 3.92 0.84
4. Sleep in your room .72 2.35 1.48
5. Animal was responsive .75 4.25 0.86
6. Close relacionship .75 4.34 0.88
7. Travel .78 2.39 1.08
8. Sleep near .71 2.36 1.36
No items deleted .77
mean score was 26.7 (SD = 5 . 7 ) , with a range from 9 through 45. The mean
contemporary score was 28.6 ( N = 99, SD = 6.3). The Pearson correlation
between the two arrays was fro = .35 ( p .001) and rhe difference between the
childhood total score and the contemporary toral score was statistically signifi-
cant ( t o 5 = 2.26, p <0.05). This difference appears t o relate to the be-
havioral nature of the questions. T h e first two questions concern the amount
of responsibility the respondent actually rakes for care of the companion animal.
For younger children parents would provide a greater share of the animal care,
so the children would provide less care than older adolescents and adults. The
difference in age-related pet care responsibilities would be expected to yield a
higher score for the respondents when they were older, as reflected by their
contemporary vs childhood scores.
The high Cronbach alpha coefficients of the total instrument (.77 and
3 2 ) demonstrate the strength of the scale with regard to both childhood and
contemporary human-animal relationships. T h e eight items represent diverse
behavioral aspects of the human-animal bonding process. T h e intercorrelations
for childhood bonding i t e n s ranged from .92 to -.01 (median r .24, p <
TABLE 2
PRINCIPAL
AXISFACTORANALYSIS CONTEMPORARY
BONDING
SCALE
Factor 1 Factor 2 Faccor 3
1. Responsible for care .74 .49 -.32
2. Clean up .76 .31 -.31
3. Hold, stroke, or pet .54 .I8 .22
4. Sleep in your room .68 -.65 -.08
5. Animal was responsive .59 .07 .55
6. Close relationship .59 .19 .37
7. Travel .46 -.02 -.I9
8. Sleep near .70 -.56 -.07
746 R. H. PORESKY, rn AL.
.05). For contemporary bonding, the range was .85 to .12 (median r .34, p
< .01). The corrected item-total correlations suggest that all items should
be retained, since removing any of the items reduces the internal reliability of
the bonding instrument; see Table 1.
The internal structure of the scale is shown by the principal axis factor
loadings (Table 2 ) . This analysis yielded three factors for the contemporary
use of the scale which accounted for 41.1%, 14.5% and 9.1% of the total
variance, respectively. The first factor appears to be a bonding or involve-
ment factor; the second appears to be related to animal size (inferred from
sleeping arrangements); and, the third appears to relate to the companion
animal's responsiveness and autonomy.
Constract Validity
Pet ownership has been associated with measures of attitudes toward pets
(Templer, Salter, Dickey, Baldwin, & Veleber, 1981). The scale should show
even stronger relationships with attitudes toward pets and pet ownership. This
expectation is supported by the correlation between this scale and the Pet Atti-
tude Scale for both the childhood and contemporary bonding scales (.42 and
.38, respectively, 9 < .001).
The Companion Animal Bonding Scale shows promise as a reliable and valid meas-
ure of an individual's bond with a pet. Through utilization of factors relating to inter-
action and relationship with a companion animal, this scale provides a basis for further
study of the human-animal bond. In this study, this instrument yielded internal re-
liabiliry coefficients of 0.77 for the childhood scale and 0.82 for the contemporary scale.
Both the childhood and contemporary uses showed construct validity through their sig-
nificant correlations with the Pet Attitude Scale. The Companion Animal Bonding Scale
provides by self-report a direct operational definition of companion animal bonding,
which can be used in research to focus o n the human-anirnal bond, rather than animal
ownership, and consequent effects o n children's emotional and social development.
REFERENCES
1. COVERT, A. M., WHIREN,A. P., KEITH,J., & NELSON,C. (1985) Pets, early
adolescents. and families. Marriage and Family Review, 8, 95-108.
2. HYDE,K. R.. KURDEK,L., & LARSON. P. (1983) Relationship between pet owner-
ship and self-esteem, social sensitivity. and interpersonal trust. Psychological
Repom, 42, 110.
3. KELLERT, S. R., & WESTERVELT, M. 0. (1983) Historical trends in American
animal use and perception. Internu~donalJournal for the Study of Animal Prob-
l e m ~ 42.
, 133-146.
4. LEVINSON, B. M. (1978) Pets and personality development. P~ychological Re-
ports, 42, 1031-1038.
5. LEVINSON, B. M. (1982) The future of research into relationships beriveen people
and their animal companions. International Journal for the Strrdy of Animal
Problems, 3, 283-294.
6 . ROBIN,M., & TEN BENSEL.R. (1985) Pets and the socialization of children. Mar-
riage and Family Review. 8. 63-78.
7. TEMPLER, D., SALTER,C., DICKEY,S., BALDWIN, R.. & VELEBER,D. (1981) The
construction of a pet attitude scale. Psychological Record, 31, 343-348.
Accepted February 24, 1987.
