Document 14469609

advertisement
hychology and Aging
1986, Vol. l,No. 2, 127-13:
Copyright 1986 by the Ammo I Piychohviol Association, Inc.
0882-7974/86/$00.75
Adulthood Age Differences in Causal Attributions for Cognitive,
Physical, and Social Performance
Margie E. Lachman and Leslie Z. McArthur
Brandeis University
Past research has revealed an unflattering pattern of attributions for the performance of the elderly.
More specifically, poor performance by the elderly is attributed to internal and stable factors such as
inability, whereas poor performance by the young is attributed to external and unstable factors such
as bad luck. In the present study, 42 young (M age = 19.18 years) and 39 elderly (M age = 74.90
years) men and women made causal attributions for their own or for another person's hypothetical
performance in the cognitive, physical, and social domains. When attributions for the same performance
by young and elderly adults were compared, the results presented an unflattering view of the elderly,
similar to the pattern in previous research. In contrast, when attributions for good versus poor performance by the elderly were compared, a more favorable picture emerged: The elderly were more
likely to be given credit for their good performance than to be blamed for their poor performance.
These findings give reason to question the pervasiveness of the negative view of the elderly that has
been presented in previous studies.
In past research comparing performance attributions for young
and elderly adults, a clear pattern of age differences has emerged.
The failures of the elderly are more often attributed to internal
and stable causes, such as age or inability, than are those of the
young (Banziger&Drevenstedt, 1982; Blank, 1982; Reno, 1979;
Rodin & Langer, 1980). Although age differences in attributions
are more consistently found for failures than for successes, there
is some evidence that the successes of the young are more often
attributed to internal and stable causes than are those of the
elderly (Banziger & Drevenstedt, 1982;Gekoski&Knox, 1983).
This unflattering attributional pattern reflects a stereotypic view
of the elderly as being less competent than the young. Whether
these views are held by the young or by the elderly themselves,
they can have an impact on behavior by and toward the elderly.
In studies with young adults, this attributional pattern has been
associated with learned helplessness (Abramson, Seligman, &
Teasdale. 1978). Thus, it is important to establish the conditions
under which this attributional pattern exists for elderly adults,
so that potentially debilitating consequences such as depression,
low motivation, lack of effort, or poor subsequent performance
can be either prevented or modified.
but the generalizability of the findings is uncertain because these
investigations also have been limited in a number of ways. First,
with few exceptions (e.g., Blank, 1982), the studies have used an
observer paradigm, in which subjects make attributions for the
performance of another person rather than for then- own performance. It is known that attributions about others (observer
attributions) differ from self- (actor) attributions in a variety of
ways (Watson, 1982). Second, those making the attributions for
the elderly have typically been young adults rather than the elderly
themselves. Although it is important to consider how the young
perceive the elderly, it is also important to consider how the elderly
perceive themselves (Blank, 1984). Third, the behavioral events
and performances for which attributions are made have been
restricted to the cognitive or physical domains. Both of these
domains are associated with negative stereotypes about agingrelated decrements in functioning. One question addressed in
the present study was whether the attributional results from the
cognitive and physical domains would be generalizable to functioning in the social domain, an area in which the elderly are
likely to be seen in a more favorable light.
To begin to address some of the limitations of past research,
in the present study we examined the causal attributions made
by both young and elderly adults for performance in the cognitive,
physical, and social domains. Subjects made attributions about
hypothetical situations involving themselves, another person of
the same age, or another person of a different age. Based on
previous research, we predicted that more internal stable (inability) attributions would be made for the elderly than for the
young for poor performance. It was also predicted that ability
attributions would be less likely for the elderly than for the young
for good performance.
Whereas previous research has focused almost exclusively on
between-age-group comparisons, in the present study we also
compared attributions for good versus poor performance by the
elderly. Although attributions for performance by the elderly may
be unflattering relative to those for the young, it is also important
As we have summarized here, past studies on performance
attributions in the elderly have yielded fairly consistent results,
This research was supported in part by National Institutes of Health
Biomedical Research Support Grant SO7 RR07044. Portions of this article were presented at the 37th annual meeting of the Gerontological
Society of America, San Antonio, Texas, November 1984.
We express our appreciation to Bart Moss, Shira Sanders, and Naomi
Yadin for their assistance in developing the attribution questionnaire; to
Barry Paul and Margaret Sail for their assistance with data collection; to
Susan Trotter for her assistance with data analysis; and to two anonymous
reviewers for their constructive comments on the manuscript.
Correspondence concerning this article should be addressed to Margie
E. Lachman, Department of Psychology, Brandeis University, Waltham,
Massachusetts 02254.
127
128
MARGIE E. LACHMAN AND LESLIE Z. McARTHUR
to compare attributions for success and failure by the elderly.
Are the elderly more likely to receive blame for failure than to
receive credit for success? In keeping with the unflattering patterns found in past age-comparative work, it was predicted that
internal attributions would be more likely for the elderly for poor
performance than for good performance. The predicted outcomes
were expected for attributions for the self and for attributions
for others in all domains, although the most favorable views of
the elderly were expected for self-attributions and in the social
domain.
Table 1
Method
3.
Subjects
Ninety-six young and elderly adults were randomly assigned to one of
Internal Stable (Ability) Attributions for Hypothetical
Events in Three Domains
Event
no.
Cognitive domain
1.
2.
4.
5.
three yoked attribution conditions: attributions for the self, attributions
for another person of the same age, and attributions for another person
of a different age. The yoked subjects were grouped into 32 triads, each
6.
composed of the three attribution conditions. The person in each triad
who made self-attributions served as the target for the two members of
the triad who made other attributions. The three subjects in each triad
were always the same sex. Five triads (15 subjects) were dropped from
Event and attribution
7.
8.
the analyses because of five unusable questionnaires from elderly subjects.
The final sample included 81 subjects with the following age and sex
John wins a prize in a radio station contest because he is
good at remembering numbers. (Good memory)
John was asked to give out name tags at a reunion because he
is good at remembering names. (Good memory)
John's absentmindedness led him to purchase two
subscriptions to a magazine. (Poor memory)
John's forgetfulness led him to leave soup cooking too long.
(Poor memory)
John's solution to a brainteaser puzzle was printed in a
newspaper because he is good at solving puzzles. (Good
problem solving)
John voted against a fund-raising activity because his good
mathematical skills revealed that it would not be profitable.
(Good problem solving)
John's bank account interest was more than expected because
John is poor at calculating interest. (Poor problem solving)
There's an error in John's tax return and he owes money,
because John is bad at figuring income taxes. (Poor
problem solving)
distribution in 27 triads: 8 young male triads, 7 young female triads, 6
Physical domain
elderly male triads, and 6 elderly female triads.
TheyoungaduIts,whohadameanageofl9.18years(SD = 1.5 years),
were students in an introductory psychology class and were given research
credit for their participation. The elderly adults, who had a mean age of
9.
10.
74.89 years (SD = 6.42 years) and a mean educational level of 12.83
years (SD = 2.53 years), were community-residing members of senior
11.
citi2en organizations. A nominal donation was made to the senior organization for each member who participated in the study.
12.
Measures
The attribution questionnaire, which was modeled after Duval and
13.
14.
Wicklund's (1973) questionnaire, included 24 hypothetical events, equally
distributed across the cognitive, physical, and social domains. Within
each domain there were two behavioral categories, each represented by
4 events. In the cognitive domain, behavioral categories were memory
!5.
16.
John completed a long distance bicycle trip because he has a
lot of strength. (Strength)
A pane of glass fell out when John shut the door because he
shuts doors very forcefully. (Strength)
John scored higher than other players at shuffleboard because
John is too weak to overshoot the mark (Weakness)
John had internal bleeding after an operation because he was
in poor physical shape. (Weakness)
John is not hit when a baseball comes out of nowhere
because he is able to move quickly. (Speed)
A car ran into John when he stepped on the brakes to avoid
hitting a child chasing a ball, because John's braking
reflexes are very fast. (Speed)
John typed accurately because he types slowly. (Slowness)
John runs into another car because he has slow reflexes.
(Slowness)
and problem-solving; in the physical domain, strength and speed; and in
the social domain, independence and nurturance. In each behavioral cat-
Social domain
egory there were two items for each performance outcome (poor, good).
The performance led to negative consequences for one item in each outcome pair, and to positive consequences for the second item. Thus, performance outcome was crossed with consequences. This design feature
was included to examine whether the elderly would be consistently blamed
17.
18:
when things go wrong (negative consequences) and would not be given
credit when things go well (positive consequences), regardless of whether
19.
their performance was poor or good. The internal stable attributions for
the 24 items are summarized and presented by domain, behavioral cat-
20.
egory, performance outcome, and consequences, in Table 1.
All of the subjects completed a demographic questionnaire about their
21.
education (years in college for young adults, or number of years of
schooling and degrees received for the elderly), occupation (job aspirations
for the young, or current or previous job for the elderly), marital status,
22.
23.
age, race, and sex. Although the only variables used for analyses in the
present study were age and sex, the other information was included so
these variables would not be made salient to the subjects.
24.
John had a good time at a social gathering where he knew no
one, because he is very self-reliant in making new friends.
(Independence)
John had a terrible time on a vacation group tour because he
is so independent that he dislikes group tours.
(Independence)
A relative does nice things for John because John needs help
and companionship. (Dependence)
A relative is abrupt with John because John is too
emotionally dependent. (Dependence)
A child listens to John's stories for hours because John is
good at amusing children. (Nurturance)
Whenever John goes out his puppy cries, because John is too
nurturant and spoiled the puppy. (Nurturance)
John refuses to visit a depressed friend because he has little
sympathy for depressed people, and the friend recovered
anyway. (Non-nurturance)
A child gets hurt when John is babysitting because John is
bad at caring for children. (Non-nurturance)
Procedure
Subjects were tested in groups of from 5 to 10 persons. All of the
subjects completed the demographic questionnaire before the attribution
Note. Odd-numbered items have positive consequences, and even-numbered items have negative consequences. This version was for men who
made attributions for others.
129
AGE DIFFERENCES IN ATTRIBUTIONS
questionnaire. Those making self-attributions were instructed to imagine
that the 24 hypothetical events described in the attribution questionnaire
had happened to them, and to rate each cause for their own performance.
Those making attributions for others were instructed to imagine that the
event had happened to the person described on a biographical sheet that
was attached to the questionnaire, and to rate each cause for that person.
The biographical sheet contained accurate information (race, age, sex,
marital status, education, and occupation), taken from the demographic
questionnaire, about the target in the same triad. Those making attributions for another person were told that the information pertained to
a real person who had participated in the same study, but that a fictitious
name was being used to conceal the person's identity. For those making
self-attributions, the questionnaire used the pronoun you; for men making
attributions for others, it referred to John; and for women making attributions for others, it referred to Jane. For each event, subjects rated three
causes—internal stable (ability), internal unstable (effort), and external—
on a 7-point scale, very unlikely cause (1) to very likely cause (7). An
additional forced-choice measure was created by asking subjects to circle
what they considered to be the best explanation or the most important
of the three causes for each event. The analyses focused on the ability
attributions because these have the most debilitating consequences if they
are made for failure, or not made for success. The effort and external
variables were included for the forced-choice analysis, to determine if
ability was the most frequent variable chosen for the elderly's failure and
the least likely to be chosen for their success.
Results
A 2 X 2 X 2 X 2 (Age Group [young, elderly] X Sex [male,
female] X Performance Outcome [poor, good] X Consequences
[positive, negative]) analysis of variance was performed for each
behavioral category, to obtain error terms for subsequent comparisons of means. All a priori predictions were examined with
two-tailed t tests. Analyses were done separately for each attribution condition and for positive and negative consequences,
collapsed across sex.1 As we mentioned earlier, the analyses were
performed on the ability attributions and on the forced-choice
variable.
Between-Subjects Analyses
Berween-subjects analyses compared the ability and forcedchoice attributions made for each event when the target was young
versus elderly.
Of the 24 events for each attribution condition, significant age
differences were found in ability attributions for the self for three
events, in ability attributions for different-age others for four
events, and in ability attributions for same-age others for four
events. All but one of these outcomes reflect the predicted negative
view of the elderly. These results are summarized in Table 2.
Attributions for the self. The results for self-attributions supported the predictions that the elderly would be more likely than
the young to rate inability as a cause for poor memory (Event
4: M young = 3.60, M elderly = 5.33), t(25) = 2.47, p < .001;
poor problem solving (Event 7: M young = 2.87, M elderly =
4.67), f(25) = 2.31, p < .05; and weakness (Event 12: M
young = 2.20, M elderly = 4.92), f(25) = 3.49, p < .01.
Attributions for different-age others. Also consistent with
predictions, those making attributions for a target of a different
age were less likely to rate ability as a cause for the good memory
of elderly than for the young targets' good memory (Event 1: M
Table 2
Age Differences in Ability Attributions That Reflect a
Flattering (+) or Unflattering (-) View of the Elderly
Attribution condition
Domain
Self
Different-age
other
Same-age
other
Cognitive
Good memory
Poor memory
Good problem solving
Poor problem solving
Physical
Strength
Weakness
Speed
Slowness
Social
Independence
Dependence
, Nurturance
Non-nurturance
Note. An empty cell indicates no significant age difference. Entries are
collapsed across positive and negative consequences.
• Indicates an unflattering view of the elderly for both positive and negative
consequences.
young = 5.53, M elderly = 3.58), t(25) = 2,79, p < .01; strength
(Event 9: M young = 5.93, M elderly = 4.25), t(25) = 2A5,p<
.05; and speed (Event 14: M young = 5.80, M elderly = 2.75),
t(25) = 3.86, p < .001. These effects are consistent with the
predictions that ability attributions would be less likely for the
elderly than for the young for good performance, and they reflect
a tendency to make more unflattering attributions for the elderly.
One final significant finding for those making attributions for
different-age others was that they were more likely to rate inability
as a cause for non-nurturance by the elderly than by the young,
*(25) = 2.54, p < .05 (Event 23: M young = 2.20, M elderly =
4.08), which is also consistent with the general tendency to make
unflattering attributions for the elderly.
Attributions for same-age others. As predicted, those who
were the same age as the target were more likely to rate inability
as a cause for the elderly than for the young for weakness (Event
12: M young = 3.20, M elderly = 4.83), t(25) = 2.08, p < .05,
and for the two (positive and negative) non-nurturance events
(Event 23: M young = 3.00, M elderly = 5.17), t(25) = 2.93,
p < .01, and (Event 24: M young = 2.33, M elderly = 4.75),
t(25) = 3.21, p < .01, respectively. Those making attributions
for same-age others also were more likely to rate ability as a
cause for speediness by the elderly than by the young, t(25) =
2.34, p < .05, which was an unexpected outcome (Event 14: M
young = 3.07, M elderly = 4.92). In contrast to all of the other
findings presented thus far, this latter attribution represents a
positive view of the elderly.
1
Only one significant sex difference was found: The elderly women
were seen as more independent than the elderly men. To increase cell
size and for ease of reporting the results, sex was ignored in the remaining
analyses.
130
MARGIE E. LACHMAN AND LESLIE Z. McARTHUR
Choice. Subjects were asked to choose the most salient of
the three (ability, effort, external) causes for each event. The
proportion of subjects who selected the ability attribution as the
most likely cause was computed for each event. An arcsine transformation was performed on these proportions prior to conducting the (-test analyses.
For those making self-attributions, there were four significant
differences. A greater percentage of elderly (42%) than young
(07%) chose the ability attribution for good memory (Event 2),
t(ao) = 2.24, p < .05, which reflects a positive view of the elderly.
In contrast, a negative view of the elderly was observed for three
events. A greater percentage of elderly than young chose ability
as the cause for poor problem solving (Event 7: young = 07%;
elderly = 40%), f(oo) = 2.03, p < .05, and strength (Event 12:
young = 07%; elderly = 50%), /(oo) - 2.53, p < .05. A greater
percentage of the young (80%) than the elderly (36%) chose ability
for nurturance (Event 22), I(oo) = 2.16, p < .05.
For those making attributions for another of the same age,
there was one significant difference that reflected a positive view
of the elderly. A greater percentage chose the ability attribution
for independent behavior (Event 17), t(ao) = 2.65, p < .01, for
the elderly (50%) than for the young (07%).
For those making attributions for another of a different age,
there were two significant effects. A greater percentage chose
ability as the cause of nurturance (Event 21), /(oo) = 2.72, p <
.01, for the elderly (92%) than for the young (47%), which reflects
a positive view of the elderly. Also, a greater percentage chose
ability as the cause of slowness (Event 16), f(oo) = 3.09, p < .01,
for the elderly (50%) than for the young (0%), which reflects a
negative view of the elderly.
Within-Subjects Analysis
Within-subjects analyses compared attributions for poor performance versus good performance for event pairs in the same
behavioral category with the same consequences, using the dependent t test. Only the results found with the elderly are reported
here.3
Based on results of previous studies examining between-subjects differences, it was predicted that ability attributions would
be more likely for the poor performance of elderly targets than
for their good performance. Overall, 3 out of the 17 significant
effects were in the predicted direction, and 14 were contrary to
predictions, reflecting a positive view of the elderly. These results
are summarized in Table 3.
Attributions for the self. There were significant performance
outcome effects for self-attributions on 6 of the possible 12 pairs.
One of these revealed the predicted tendency to make greater
ability attributions for poor performance, whereas 5 revealed a
reversal of the prediction. Contrary to predictions, the elderly
made more internal stable attributions for their own nurturance
than for non-nurturance (Event 21, M = 5.58, vs. Event 23,
M = 2.00), /(!!) = 8.95, p < .001, and (Event 22, M = 4.50, vs.
Event 24, M = 3.25), r ( l l ) = 3.13, p < .01; for speed than for
slowness (Event 13, M= 5.83 vs. Event 15, M = 4.83), J ( l l ) =
2.27, p< .05, and (Event 14, M = 5.00, vs. Event 16, M = 3.50),
t(l 1) = 3.41, p < .01; and for strength (Event 9, M = 4.08) than
for weakness (Event 11, M = 3.00), 1(11) = 2.57, p < .05. Consistent with predictions, the elderly were more likely to make
Table 3
Ability Attributions for the Elderly That Reflect a Flattering
(+) or Unflattering (-) View of the Elderly
Attribution condition
Domain
Self
Different-age
other
Same-age
other
Cognitive
Good vs. poor memory
Good vs. poor problem solving
Physical
Strength vs. weakness
Speed vs. slowness
Social
Independence vs. dependence
Nurturance vs. nonnurturance
Note. An empty cell indicates no significant difference in attributions for
good versus poor performance. Entries are collapsed across positive and
negative consequences.
' Indicates a flattering view of the elderly for positive consequences and
an unflattering view of the elderly for negative consequences.
ability attributions for dependence (Event 20, M = 3.92) than
for independence (Event 18, M = 2.75), «11) = 2.54, p < .05.
Attributions for different-age others. For attributions made
for different-age others, 7 out of 12 effects were significant. Contrary to predictions, 6 of these revealed greater ability attributions
when they were flattering to the elderly (i.e., for good performance) than when they were unflattering. More specifically,
ability was more likely to be selected as a cause for nurturance
than for non-nurturance (Event 21, M = 6.0, vs. Event 23,
M = 4.08), t( 11) = 4.8, p < .001, and (Event 22, M = 4.75, vs.
Event 24, M = 2.83), 1(11) = 4.8, p < .001; for strength (Event
9, M = 4.25) than for weakness (Event 11, M = 3.08), «(!!) =
2.79, p < .05; for good problem solving (Event 5, M = 5.25)
than for poor problem solving (Event 7, M= 3.33), /(!!) = 4.17,
p<.0l; for independence (Event 17,Af = 5.25) than for dependence (Event 19, M = 4.17), t( 11) = 2.35, p < .05; and for speed
(Event 13, M = 4.75) than for slowness (Event 15, M = 3.42),
f(l 1) = 3.02, p < .05. The latter effect held true for positive
consequences only. The one unflattering attribution was that
ability was seen as a more likely cause for slowness (Event 16,
M = 5.17) than for speed (Event 14, M = 2.75), when the event
had negative consequences, t(\ 1) = 5.5, p < .001.
Attributions for same-age others. For attributions for sameage others, there were four significant effects, all of which reflected
greater ability attributions when they were flattering to the elderly
than when they were unflattering. Ability was seen as a more
1
The primary focus of the within-subjects analyses was on attributions
for the elderly, because this had not been examined in past research. For
comparison purposes, the same analyses were conducted for the young
adults. A similar pattern was found: The young were more likely to take
or to be given credit for good performance than to be blamed for poor
performance. For ability attributions, there were 24 significant effects,
and 22 of these reflected a flattering view. For the forced-choice variable,
there were 16 significant effects, and 14 of these showed a flattering view
of the young.
AGE DIFFERENCES IN ATTRIBUTIONS
likely cause for nurturance (Event 22, M = 5.75) than for nonnurturance (Event 24,M = 4.75), t(\ 1) = 2.5,p< .05; for strength
(Event 9, M = 5.42) than for weakness (Event 11, M = 3.25),
((11) = 5.17, p < .001; and for a good memory (Event 1, M =
5.50) than for a bad memory (Event 3, M = 3.42), ((11) = 3.92,
p < .01. The latter effect held true for positive consequences,
but was reversed for negative consequences (Event 4, M poor
memory = 6.25) and (Event 2, M good memory = 4.83),
J ( l l ) = 2.68,p<.05.
Choice. The choice variable again was analyzed with t tests
after arcsine transformations were performed on the proportions
of those selecting the ability attribution. For attributions for the
self there were no significant effects. There was one significant
effect for attributions for others of the same age. A greater percentage chose ability as the cause of good problem solving (Event
5 = 64%) than of poor problem solving (Event 7 = 18%),
((oo) = 2.28, p < .05, which reflects a positive view of the elderly.
Six effects were significant for attributions for different-age others.
Two effects represent negative views of the elderly: A greater
percentage chose ability as the cause of poor memory (Event
4 = 50%) than of good memory (Event 2 = 08%), r(oo) = 2.43,
p < .05, and as the cause of slowness (Event 16 = 50%) than of
speed (Event 14 = 0%), /(oo) = 2.82, p < .01. In contrast, a
positive view of the elderly was reflected in the greater tendency
by those making attributions for another of a different age to
choose ability as the cause of good problem solving (Event 5 =
75%) than of poor problem solving (Event 7 = 17%), t(oo) =
3.04, p < .01; as the cause of nurturance than of non-nurturance
(Event 21 = 92% vs. Event 23 = 42%), ((oo) = 2.82, p < .01
(Event 22 = 42% vs. Event 24 = 0%), f(oo) = 2.43, p < .05; and
as the cause of independence (Event 17 = 75%) than of dependence (Event 19 = 17%), f(oo) = 3.04, p < .01.
Discussion
The findings were consistent with earlier research, which had
revealed more unflattering attributions for performance by the
elderly when compared to the young (Banziger & Drevenstedt,
1982; Blank, 1982), whereas there was also evidence for adaptive
attributions for performance by the elderly. As in earlier research,
poor cognitive or physical performance by the elderly was attributed more to inability than was identical performance by the
young, whereas good cognitive or physical performance by the
elderly was attributed less to ability than it was for the young.
Although the attributions in all three conditions were equally
unflattering to the elderly, an interesting pattern emerged when
the three groups were compared. Those making self-attributions
and attributions for another of the same age were more likely to
rate inability as the cause for poor performance by the elderly
than by the young. In contrast, the negativity of those making
attributions for another of a different age was manifested in their
attributions for good performance. They were less likely to rate
ability as a cause for good performance by the elderly than by
the young. In short, the young discounted good performance in
the elderly, whereas the elderly confirmed expectancies for poor
performance in themselves and in other elderly.
In several ways the present results suggest a more favorable
view of the elderly than has been found in previous research.
First, only 11 out of the 72 possible differences in attributions
131
for the performance of elderly versus young adults were significant. Although this is beyond what would be expected by chance,
the number is still small. Second, the negative view of the elderly
occurred predominantly in the cognitive and physical domains;
there were fewer negative outcomes in the social domain.
An examination of attributions for good versus poor performance by the elderly also revealed a more favorable view of the
elderly than had been expected on the basis of past research
focusing on between-age-group comparisons. Here 14 out of the
17 significant effects showed a positive view of the elderly. Ability
was seen as a more likely cause for good performance than inability was for poor performance. Good performance for the
elderly was more likely to be attributed to ability, perhaps because
it was generally unexpected and, therefore, distinctive.' These
findings are similar to the results of another study by Lachman
and Jelallian (1984), in which the elderly took credit for their
successes but did not accept the blame for their failures. Rather
than using a self-deprecating or helpless attributional style, the
elderly, like the young, used a more adaptive, egotistical attributional style. Although the present study involved imagining
one's successes and failures, the results were remarkably similar
to the results of previous studies in which the elderly made attributions for their own actual performance (Lachman & Jelallian, 1984; Rutman & Freedman, 1984).
Although valence of consequences did not appear to have an
effect on the results, there was one noteworthy pattern. For the
within-subjects comparisons, when the consequences of the events
were positive the causal attributions for the performance of the
elderly were always flattering. Thus, the elderly were always given
credit for positive outcomes. In contrast, when the outcomes
were negative, both unflattering (3 event pairs) and flattering (4
event pairs) attributions were made.
What might account for the more favorable views of those
making attributions about the elderly in the present study? One
possible explanation is that the attribution targets were presented
in a more realistic way. In the literature on attitudes toward the
elderly, more favorable views are found when more realistic and
detailed descriptions of the stimulus persons are provided (Lutsky,
1980). In the present study, more information about the targets
was given than in previous research, and age was not made as
salient. Moreover, the subjects were told that the targets were
real people, whereas in past research the subjects were given a
vague description of a fictitious person.
In conjunction with the positive views of the elderly, there
were still findings consistent with the negative view of the elderly
found in past research. Although the number of significant effects
was small, all but one was in the same direction as previous
research: Compared to the young, internal attributions were more
likely for the failures of the elderly and less likely for their successes. One possible explanation for this negative view is that in
most cases poor performance is expected for the elderly. Some
3
A pilot study with 24 subjects examined whether the events were
seen as more likely to occur (a) for a young person, (b) for an old person,
or (c) equally likely for both young and old. For all items, 75% or more
of the subjects rated poor performance as more likely to occur for the
elderly and good performance as more likely to occur for the young. The
only exceptions were for the nurturance items, for which the opposite
pattern was found.
132
MARGIE E. LACHMAN AND LESLIE Z. McARTHUR
research has suggested that expected outcomes tend to be attributed internally, and unexpected events are attributed externally (Deaux, 1976; Gekoski & Knox, 1983; Green, 1984; Rutman & Freedman, 1984). If this is so it could explain both past
and present results, because when performance was good it was
also unexpected for the elderly (e.g., solving a difficult problem,
moving quickly). In contrast, when performance was poor it was
also expected for the elderly (e.g., forgetting about a pot on the
stove, driving slowly).
In the present study all of the events depicting poor performances were expected for the elderly with the exception of nonnurturance (see Footnote 3). Even though non-nurturance was
unexpected for the elderly it was still attributed internally, however, suggesting a negative bias rather than an expectancy effect.
One challenge for future studies is to find examples of good
performance other than nurturance that are expected for the
elderly. This will be required in order to conduct a systematic
test of whether attributions are based on expectancies for performance versus negative biases or stereotypes.
Consistent with other studies (Banziger & Drevenstedt, 1984),
the attributions about the elderly in the present study seemed to
vary across tasks. What are the relevant task features that determine the nature of the attributions for performance by the
elderly? The results shed some light on this question, but there
is a need for further research aimed at examining task characteristics in more detail. In the present study, tasks were varied
by domain, and as expected, the attributions were more favorable
for the elderly in the social domain than in the cognitive or physical domains.
In concert, the findings give reason to question the largely
negative view of the elderly that has been presented in previous
attribution studies. The present results suggest that there is indeed
an age difference in attributional patterns that favors the young,
but this seems to be consistent with expectancies about agingrelated performance decrements in the cognitive and physical
domains. Despite this unflattering view, when the elderly are
compared to the young, distinctions are made in attributions for
various behaviors by the elderly, and they are more likely to
receive credit for successes than they are to be blamed for failures.
This distinction may be important for future examination of the
affective and behavioral implications of attributions for young
and elderly adults. Previous research has shown that providing
the elderly with adaptive attributions through experimental manipulations can have a positive impact on performance (Prohaska,
Parham, & Teitelman, 1984; Rodin & Langer, 1980). This suggests that attributions may be an important antecedent of performance changes in later life. Further work is needed to determine the antecedents of attributions and to examine whether
performance changes have an impact on attributions by and
about the elderly.
References
Abramson, L. Y., Setigman, M. E. P., & Teasdale, J. D. (1978). Learned
helplessness: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74.
Banziger, G., & Drevenstedt, J. (1982). Achievement attributions by young
and old judges as a function of perceived age of stimulus person. Journal
of Gerontology, 37, 468-474.
Banziger, G., & Drevenstedt, J. (1984). Age as a variable in achievement
attributions: The Weiner model and beyond. Basic and Applied Social
Psychology, 5, 97-104.
Blank, T. O. (1982). A social psychology of developing adults. New \brk:
Wiley.
Blank, T. O. (1984). Meaning and motivation in adult perceptions of
causality. Basic and Applied Social Psychology, 5, 111 -120.
Deaux, K.. (1976). Sex: A perspective on the attribution process. In J. H.
Harvey, W. J. Ickes, & R. F. Kidd (Eds.), New directions in attribution
research (Vol. 1, pp. 335-352). Hillsdale, NJ: Eribaum.
Duval, S., & Wicklund, R. A. (1973). Effects of objective self-awareness
on attribution of causality. Journal of Experimental Social Psychology,
9, 17-31.
Gekoski, W. L., & Knox, V. J. (November, 1983). Identifying age stereotypes in adolescents: An attributional approach. Paper presented at
meeting of the Gerontological Society of America, San Francisco.
Green, S. K. (1984). Senility vs. wisdom: The meaning of old age as a
cause for behavior. Basic and Applied Social Psychology, S, 105-110.
Lachman, M. E., & Jelallian, E. (1984). Self-efficacy and attributions for
intellectual performance in young and elderly adults. Journal of Gerontology, 39, 577-582.
Lutsky, N. S. (1980). Attitudes toward old age and elderly persons. In C.
Eisdorfer (Ed.), Annual review of gerontology and geriatrics (Vol. 1,
pp. 287-336). New York: Springer.
Prohaska, T. R., Parham, I. A., & Teitelman, J. (1984). Age differences
in attributions to causality: Implications for intellectual assessment.
Experimental Aging Research, 10, 111-117.
Reno, R. (1979). Attribution for success and failure as a function of
perceived age. Journal of Gerontology, 34, 709-715.
Rodin, }., & Langer, E. (1980). Aging labels: The decline of control and
the fall of self-esteem. Journal of Social Issues, 36, 12-29.
Rutman, D. L., & Freedman, J. L. (1984, August). Age differences
in
causal attributions for success and failure experiences. Poster presented
at the meeting of the American Psychological Association, Toronto,
Ontario, Canada.
Watson, D. (1982). The actor and the observer: How are their perceptions
of causality divergent? Psychological Bulletin, 92, 682-700.
Received May 31, 1985
Revision received September 25, 1985
Download