A FURTHER CROSS-CULTURAL COMPAEEISON OF CHILDREN'S
JUDGMENT OF PARENT-CHILD CONFLICT IN GERMANY.
FINLAND, ENGLAND AND THE UNITED STATES

Thesis ‘0'.th chreo c! M. A.
MICHIGAN STATE UNIVERSITY
Barbara Mac Phail Robinson
1955

'lthm‘,

ABSTRACT

This study was an analysis of Story No. 4, The Ink Spots, Series A,
of the Anderson In: mplete Stories. Story No. 2, The Lost Meat,, had
been analyzed in a previous study, and it was possible to make com-
parisons between the two analyses and further evaluate the Anderson
Incomplete Stories as a measuring instrument of intercultural
differences.

Samples used were from data collected in Firmingham, Eng‘Lnd;
in Helsinki, Finland; in Karim-r5 and Munich, German? and in Fnoxville,
Tennessee, U;S.A. Subjects were 1,227 children in the seventh school
year in these cities.

Children‘s stories were coded in two ways: by four defined levels

I‘

of outcome (integrat£"', Ecnirrijor~resistance, ambiguous, or domination—
submission); and by categories representing the psychological thenas

in the stories. In Story No. 2 there vere 33 categories for thought,
feeling and action by the child-character and 46 for the mother-
character. In Story No. 4 there were 37 catejories for thought,

feeling, and action by the child—character and 53 categories for the
mother-character.

Aims were to ana yze Story No. b and make sex and socio-economic
compariscrs in categorr clrsters and outcome levels, and these same
comparisons in or): ~cr tnrul samples, and to make C?”:?”1“CFS of
Story No. b with Story No. 2.

Reliability of coding showed percentages of agreement ransing

from 93.5% to 100%.

The Birmingh° sample was highest in the two outcomes of spontan-
eity of the child, integrative outcome and dggigation—resistan£_
It was lowest in ambiguous outcomes. That is, human relations in
Birmingham were more clearly defined by these children in their stories.
Helsinki was significantly different from all other samples in ambiguous,
unstructured outcomes of the stories, indicating a low order of communi-
cation, and significantly lower in domination—submission. Hamburg,
Munich and Knoxville were alike in all four outcomes and showed no
significant differences among themselves.

More boys showed ambiguous, unstructured outcomes than girls
in all five samplings, the differences being significant in four.
More children in the high socio-economic groups showed integrative
outcomes in both Story No. 2 and Story Ho. H, the differences for

Story No. # bein. significant in Hamburg and hunich. More children

in the low socio-economic groups had ambiguous outcomes than in the

 

high groups in all samples, the differences being significant in Helsinki,
Hamburg and Munich.

Analyses of the data showed that in completion of the stories
children were sensitive to the st-icture of the stimulus material.
There were two structural differences in the two stories that showed
effects in the analysis of the data. Story No. 4 had a question after
it: ”How do they both feel about it?" Calling attention to both the
mother and the child seemed to account for higher numbers of integrative

outcomes, higher percentages of constructive behavior by both the

mother and the child and higher frequencies of anxiety than were found

in Story No. 2, while themes of domination and submission were lower

 

 

in frequency. Second, in Story No. 4 the mother enters Just as Susan
finds herself in a predicament. There was less opportunity or necessity
to tell truth or lie, and there were much lower frequencies of both

tells truth and tells lie in Story No. 4.

 

The girls wrote longer stories with wider variety of themes than
boys. As Heter found in Story No. 2, girls gave higher frequencies
in category clusters of constructive behavior Ly mother, submission, and
anxiety. Consistently in all five samples girls were high in constructive
mother category clusters; with the exception of Helsinki, constructive

behavior by_the child; and, with the exception of Knoxville, child tells

 

igg, For all five samples girls had higher frequencies of a rigty and
submiggion, the differences being generally in the same direction as
for Story No. 2, and six of the ten differences for these two categories
being significant.

Analysis of data for Socio-economic levels showed that children
from the high groups generally wrote longer stories. They had consistent-
1y higher frequencies of integrative outcomes (as in Story No. 2), the
differences in Hamburg and Munich being significant. The high groups
had consistently higher frequencies in the category cluster of construgt—
izg behavior by the mother and, with the exception of Helsinki, construct-
i1; behavigr by the ghild. Children in the high group had higher fre-

quencies of tells lie in Birmingham and Helsinki. Children in the

 

low group had higher frequencies of tells lie in Hamburg and Munich,

 

the difference in Munich being significant. Socio-economic levels in
the four cities were exactly reversed for child tells truth. Children

in the high socio-economic group had high percentages of tells truth

 

in Hamburg and.Munich, while tells truth was found more frequently in
Birmingham and Helsinki among children at the low socio-economic level.
Children in the low groups tend to write shorter stories.

Hamburg and Munich were significantly different from each other

in category clusters of constructive behavior hy the mother, truth, lie,

 

and gggiety, supporting the findings for Story Ho. 2, which leaves open
the question of a German national character.

Knoxville stories reveal more initiative and personal integritv
in the child and more respect for the child for both Story N0. 2 and
Story No. 4. Knoxville children showed the highest percentage of t§;;§_
tgggh and the lowest percentage of tgl;§_li_. They showed high attention
to the child in having highest percentages of constructive behavior by
§g§§g, and lower attention to the mother in having next to lowest
percentages of children with themas in the constructive :ehavior by.
mgthg; category cluster. Munich and Hamburg were lowest in frequency
for constructive behavior Ly_§pggp, and Hamburg children, who gave little
attention to constructive behavior EK.&22.Ehlli. were highest in fre-
quency of constructive behavior hy_the gather. Evidence shows that
according to the children's stories highest frequencies of tggth and

lower frequencies of tells lie were accompanied by milder punishment

 

and milder forms of parental domination in Knoxville. Adnonishment
about future behavior ranked first in frequency and admonishment about

past behavior ranked third. In all the other samples scolding ranked

 

first among the types of domination. Admonishment did not rank high

in any of the other samples and scolding did not rank high in Knoxville.
It was pointed out that admonishment represents chiefly disapproval of
some kind of behavior. Scolding and physical punishment represent an

attack on the child himself.

A JURTHHR CROSS—CULTURAL COMPARISON OF CHILDRhN'S JUDGMEFT
OF RAEEHT-CHILD COKFLICT IN GERMANY, FINLAHD,

EEGLAND AHD THE UNITED STATES

13y

Barhsra Mathail Robinson

Submitted to the College cf Science and.Arts of iichigan
State University of.Agriculture and.Applied Science
in partial fulfillment of the requirements
for the degree of

MASTER OF.ARTS

Department of Psychology

1955

(In says

Acknowledgments

The author wishes to extend her sincere appreciation for the
continuous guidance and assistance she received from Dr.
Harold H. Anderson and Dr. Gladys L..Anderson in preparing
this thesis.

The author also wishes to express her gratitude to Mr. Franz
Geicrhaas for his coOperation in coding the German stories, to
Mr. Geierhaas and Mr. Richard Haber for working out the
problems of reliability and for other assistance, and to

Mrs. Bonita Edgar and Hrs. Ruth Vogt for the typing of the
thesis. '

The author deeply appreciates the financial support of the
scholarship provided by Michigan State University which made
it possible to continue with her work.

TABLE OF CONTENTS

LI ST OF TABIJ-ES . O O 0 O O O O O O O O O O

I.

II.

INTRODU TION. .

RELIABILITY OF CODING . . . .

III. COMPARISON OF CODING CATEGORIES FOR STORIES KO. 4,

mm INK SPOTS, AND no. 2, THE LOST MEAT

IV. OUTCOMES OF STORIES. . . . . . . . . . .
Results . . . . . . . . . . -. . . .
Discussion . . . . . . . . . . . . .
V. THE BIRMINGHAM, ENGLAND, SAMPLE. . . . . . .
Sex Comparisons in Story No. 4 and Story No. .
Discussion of Sex Comparisons . . . . . .
Socio—economic Comparisons. . . . . . . .
..Story No. b - Story Ho. 2 Comparisons . . . .
Discussion' . . . . . . . . . . . .

VI. CROSS-CULTURAL COMPARISONS . . . . . . . .

Page

16
17
19
22
22
26
29
32
32
3»

Category Cluster: Constructive thinking, feeling,

”1rd action 32 mother . . . . . . . . .

 

Category Cluster: Constructive thinking, feeling,
_;nd action 3:_§u§§n . . . . . . . . .
Category Cluster: Susan tells $3312. . . . .
Category Cluster: Susan tells lie . . . . .
Category Cluster: Resistance. . . . . . .

 

 

 

Category Cluster: Anxiety. . . . .
Category Cluster: Submission. . . .
Category Cluster: Domination. . . .
Discussion of Cross-cultural Comparisons
Birmingham . . . . . . . . .
Helsinki . . . . . . . .
Hamburg and Munich. . . . . . .
Knoxville. . . . . . . . . .
length of Stories . . . . . . .
VII. CROSS-CULTURAL SEX COMPARISTHS . . . .
VIII.CRCSS—CULTURAL SOCIO—ECOHOMIC COMPARISONS
IX. TYPES OF DCMINATIOH. . . . . . . .
Discussion of Types of Domination . .
K. SIMI-‘ARY . . . . . . . . . . . .
sameness . . . . . . . . . I. . .
.APPEHDIX A. . . . . . . . . . . . .

APEIDIX B O O 0 O O O O O O O O O

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 60

Table 7.

Table 80

Table 9.

LIST OF TABLES

Distribution of Numbers of Boys and Girls
by Socio-economic Level . . . . . . . . .

Outcomes of Story No. 4. Frequencies and Percentages

for Children in Each of Five Cities in the Outcomes

of Their StorieS. . . . . . . . . . . .

Birmingham, England: Numbers and Percentages of
Boys and of Girls with Themas Coded in Eight Cate—
gory Clusters, and According to Classification

of Outcomes of Stories No. 4 and No. 2 . . . .

Birmingham, England: Numbers and Percentages of
Children from High and Low Socio-economic Levels
with Thomas Coded in the several Category Clusters,
and.According to Outcomes for Stories No. 4 and

NO. 2 o o o o o o o o o o o o o o 0

Birmingham, England: Numbers and Percentages of
Children.With Thenas Coded in Eight Category
Clusters, and Integrative Outcomes, of Stories

NO. 1+ and N0. 2 o o o o o o o o o o o 0

Numbers and Percentages of Children in.Each of the
Samplings with Thomas Coded in the Category
Cluster, Constructive Thinking, Feeling, and
Action by Mother for Stories No. 3 and No. 2 . .

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster, Constructive Thinking, Feeling, and
Action by the Child for Stories No. 4 and No. 2 .

Numbers and Percentages of Children in Each of the
Samplings with Thomas Coded in the Category
Cluster, Child Tells Truth, for Stories No. 4

and NO. 2 o o o o o o o o o o o o o 0

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster, Child Tells Lie, for Stories No. 4
and NO. 2 o o o o o o o o o o o o o o

Page
6

18

25

30

35

36

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

10.

11.

12.

13.

14.

17.

18.

19.

20.

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster, Resistance, for stories No. 4 and No. 2.

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster, Anxiety, for Stories No. 4 and No. 2. .

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster,Submission, for Stories No. 4 and No. 2 .

Numbers and Percentages of Children in Each of the
Samplings with Themas Coded in the Category
Cluster, Domination, for Stories No. 4 and No. 2.

Average Number of Thema Tallies per Story for
Story No. 4 in Each of the Samples According to
Sex Groups and Socio-economic Groups; for the
Munich and Hamburg Samples, Sex Comparisons of
tories NO. 4 and N0. 2 e e e e e e e e e
Cross-cultural Sex Comparisons. Recapitulation
of Differences Between Boys and Girls from
Tables 3, 19, 20, 21, 22 . . . . . . . .

rose-cultural Socio—ecanomic Comparisons.
Recapitulation of Differences Between High,
Middle and Low Socio-economic Levels from
Tables 4, 23, 24, 25 . . . . . . . . . .

Story No. 4. Percentages of the Total Numbers of
Children in the Five Samplings Giving Themas of
Domination of the Child, and Percentages of

Total Numbers of Children in Each Sampling Giving
Themas in Twelve Categories of Domination . . .

Rank Order of Types of Domination within Each of
the Five Samples, Taken from Table 17 . . . .

Helsinki, Finland: Numbers and Percentages of
Boys and Girls with Themas Coded in the Eight
Category Clusters, and.According to Outcome . .

Hamburg, Germany: Numbers and Percentages of

Boys and Girls with Themas Coded in the Eight
Category Clusters,and According to Outcome. . .

ii

Page

. 38

41

42

54a

56a

60

63

72

73

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Munich, Germany: Numbers and Percentages of Boys
and Girls with Themas Coded in the Eight Category
Clvsters, and According to Outcome . . . . .

Knoxville, Tennessee: Numbers and Percentages of
Boys and Girls with Themas Coded in the Eight
Category Clusters, and.According to Outcome . .

Helsinki, Finland: Numbers and Percentages of
Children in the High, Middle and Low Socio—
economic Groups with Themas Coded in the Eight
Category Clusters, and.According to Outcomesfor
Stories No. 4 and No. 2 . . . . . . . . .

Hamburg, Germany: Numbers and Percentages of
Children in the High, Niddle and Low Socio-
economic Groups with Themas Coded in the Eight
Category Clusters, and According to Outcomes for
Stories No. 4 and No. 2 . . . . . . . . .

Munich, Germany: Numbers and Percentages of
Children in the High and Low Socio-economic Groups
with Themas Coded in the Eight Category Clusters,
and According to Outcomes for Stories No. 4 and

1:0 0 2 O O I O O I O O O O O O O O 0

iii

Page

74

75

76

77

78

I. INTRODUCTION

This study is part of a research program being carried on by
H. H..Anderson and G. L. Anderson (2, 3). The program is cross-
cultural in nature and is designed to investigate, through analysis
of children's written responses, the kinds of interaction that
take place between children and their peers, their parents, and their
teachers. The approach has been to present children with a projective
device, the Anderson Incomplete Stories (see Appendix B), and ask
them to write down their own conclusions for the stories. This pro-
cedure has been carried out in classrooms of selected public schools
in the German cities of Karlsruhe, Braunschweig, Hamburg, and Munich:
in Helsinki, Finland; in.Drammen, Norway: in Stockholm, Sweden: in
Birmingham, England; in Mexico City and some rural areas in.Mexico;
and in Knoxville, Tennessee.

Analyses have been made of data on Story No. 2, The Lost Meat,
Series.A, from Karlsruhe, Germany, and from Mexico City by‘Anderson
and.Anderson (3). Anderson (2) and Geierhaas (6, 7). Geierhaas
has dealt with problems of reliability of the children, the trans-
lator, and the coder. Heber (8) has also analyzed data from Story
No. 2, The Lost Meat, and has made sex comparisons, socio-economic
comparisons, and cross-cultural comparisons of reSponses of children

from German, English, Finnish, United States, and Mexican cities.

The hypotheses being investigated by.Anderson and his co-workers
in the research program of which this study is a part are:

1. Children brought up in an extremely authoritarian and dom—
inating culture (Germany) are different in their interper—
sonal relations from children in less dominating (more
democratic cultures.

~2. As compared with children in a more democratic culture,
children in the more authoritarian culture will show sig-
nificantly higher frequencies of responses of: anxiety,
lying, cheating, deception, punishment, daydreaming (fantasy
unrelated to reality), escape from reality, ambiguous un—
structured relating, conformity, guilt, and submission in
fantasy conflict situations.

‘9)
0

Children in the more authoritarian culture will also show
significantly fewer responses of honesty, sense of fair play,
social problem solving, communication, action, spontaneity,
cooperation and integrative behavior (working together).

4. When outcomes of interpersonal conflict are classified
according to levels of increasing authoritarianism or domp
ination, the distribution of outcomes for German children
will be skewed at the bottom of the following levels of human
relating:

a) Integrative, problem-solving

b) Domination-resistance

c) Ambiguous, unstructured, indecisive

d) Domination-submission. (2)

Heber found in.his analysis of Et>ry No. 2, The Lost Meat,

that significantly more Children in Knoxville and Birmingham gave
responses portraying the child in the story as telling the truth than
did children in Hamburg and Helsinki. The tells truth theme also
appeared significantly more often in the Hamburg and Helsinki stories
than in the Karlsruhe, Mexico City and Munich stories. For the response
portraying the child as telling a lie, the order of samplings was
reversed, with Munich stories having significantly more of this

response than the other cities. These results were consistent with

the hypotheses.

The greater numbwr of children giving lg; themas which Heber
found in the Munich stories was thought to account for the fact that
the stories from Munich contained significantly fewer responses of
submissiog than any of the other samplings, since stories with themas
of lie or deception did not usually contain themas of gaggiggign.

Domiggtigg themas were found to be significantly more frequentv
in Birmingham than in the other samplings, while Karlsruhe had the
least frequent domination themas. However, the domination in '
Birmingham was accompanied by a high frequency of :gsistange themas
and a low frequency of ggbmissign themas. The Karlsruhe sample con-
tained a high frequency of submission and a low frequency of‘gggiptgppg
themas. Resistance is a higher level of response to domination than
is submission; it is still spontaneous and communicative, whereas
submission is close to psychological atrophy. Thus the pattern in
Birmingham was as a whole at a higher level of response than that in
Karlsruhe.

0n the thema of constructivg behavior by Michael, Birmingham and
Hamburg were significantly higher than Munich, Knoxville, and Helsinki,
which in turn were significantly higher than.Karlsruhe and Mexico City.
On the thema of gopstructive behavio; by;pgphg;,and on stories with
inggggative outcomgg, Hamburg, Helsinki, and Birmingham, in that
order, were significantly higher than Karlsruhe, Munich, and Mexico
City. The results from these three clusters of themas are consistent
with the other clusters of themas which represent high levels of

spontaneous relating (tellg truth, resistance) in that Karlsruhe,

3.

Munich, and Mexico City had the lowest percentages. These results

and consistencies strongly support the hypotheses.

The Present Study

Aimg. The aims of the present study are: (l) to analyze Story
No. Q, The Ink Spots, Series A, of the Anderson Incomplete Stories,
from data collected in Birmingham, Helsinki, Hamburg, Munich, and
Knoxville; (2) to compare outcomes of stories for these samples;
,(3) to make a comparison of results from Story'No. # with results
from Story No. 2 obtained by Heber (8), using the stories from the
Birmingham sample for sex comparisons and socio-economic comparisons;
(u) to find the comparative reSponses of children in each sample to
several category clusters of themas; and (5) to study the types of
domination mentioned by children in each of the samples.

§pbjects. This study is based on data obtained from 1,227 children
in the seventh school year. The distribution of subjects in the five
samples by sex, socio-economic level and location are given in Table l.
The samples are essentially the same as those described by Heber (8)
in his study with the exception that the Karlsruhe and Mexican samples
‘have been omitted and the middle socio-economic group in Hamburg is
smaller.

figphpds and procedure. The methods of collecting and coding
the data have been reported by.Anderson and Anderson (2, 3), Geierhaas
(6, 7), and Heber (8).

For purposes of statistical treatment of the data, each child's
story is coded according to themas having psychological significance.
The thema categories for Story No. 2 have been listed by Geierhaas (6),
along with the description of units used for coding and classifying
of stories as to outcome. The categories used for Story No. 4 are
to be described under Chapter III of this study.

In addition to coding of themas into categories, each story is
Judged as to general outcome and is assigned to one of six mutually
exclusive outcome classifications. These are described under Chapter
IV of this study.

In the tabulating of the coded themas it frequently happens that
in a given story there are several themas for the same category. In
the treatment of data in this study multiple tallies have been dis—
regarded and unless specifically stated otherwise, frequencies refer
to numbers of children. Within category clusters each child is

counted only once.

5.

1114133"1 1

DISTRIBUTION OF EUHBZRS OF BOYS.AHD GIRLS
BY SOCIO—ECOKCEIC IEVE

 

 

 

 

BOYS GIRT
Socio-Economic Level Socio-Economic Level
Location High Middle Low All High Middle low All Total
Birmingham 156 . 64 122 53 395
Hel sinki 36 1+0 27 38 1+0 27 208
Hamburg”; 36 57 30 33 53 37 246
Munich 36 37 #8 M3 164
Knoxville 93 121 214
Total 1227

II. RELIABILITY OF CODING

As reported.above, considerable effort was made to analyze Story
No. 2, The Lost Meat, before analysis was made of Story Be. h, The Ink
Spots. The writer has participated.with the Andersons, Geierhaas and
Heber in revising the coding categories for Story Be. h.

Reliability of coding Story No. h was first established between
the writer and Geierhaas on a sample of Karlsruhe stories, and after
a lapse of 23 days established again on a sample of English stories.
The following percentages of agreement were obtained: 93.5% on cate—
gories and 95.5% on outcomes for the Karlsruhe sample, and 96.6% on
categories and 100% on outcomes for the Birmingham sample.

Since this thesis was to involve a comparison of Story No. 2 and
Story No. b for the Birmingham sample, and since Heber had coded all
the Birmingham children's Stories No. 2, it was desirable to determine
if the writer could code the Birmingham stories reliably with Heber.
Reliability between.the two was established with 97.6% agreement on

categories and 95% agreement on outcomes.

III. COWARISON OF CODING CATEGORIES FOR STORIES NO. 4,

mm INK srors, AND NO. 2, mm LOST mm

In the design of the Anderson.Incomplete Stories it was desired
to have two stories presenting a potential conflict between a child
and his mother. Story No. 2, The Lost Meat, involves a boy and his

mother, and Story No. 4 involves a girl and her mother.
The English texts of the two stories are as follows:

Story No. L The Lost Meat.

Michael's mother sends him to the store to get one pound of
wieners. On the way home he puts the package of meat on the

edge of the sidewalk and plays for a short time with his friends.
A shepherd dog darts forward quickly, claws half of the wieners
out of the package and rushes away with them. Michael wraps up
the remaining wieners and takes them home.

What does Michael say to his mother?
What does his mother do?
How does Michael then feel about it?

Think about these questions, then fipish this story quickly with
a few sentences.

S o No The Ink S ots.
Susan; is sitting in the living roomnworking on her homework.
Suddenly she remembers her mother's new coat. She wonders how
she would look in it. When taking the coat off she notices that
she had got some ink spots on her mother's new coat. Just as
Susan is trying to wipe out the ink spots her mother enters the
room.

 

1The name of the girl used in.Knoxville in 1953 was Elizabeth. The
name was changed to Susan for the Birmingham stories in 1954 because it
would be shorter for the children to write.

What does her mother say? How do they both feel about it?
What does Susan say? What does each one do?

Think about these questions and finish this story quickly with a few
sentences.

‘Abbreviated titles of the coding categories as revised for Story

No. 4 are the following:

Categories for Susan

IV Tells truth spontaneously, the whole truth

ID Tells truth spontaneously, in detail

2“ Tells truth after conflict, anxiety, after fantasy lie (whole truth)
2D Tells truth after conflict, anxiety, after fantasy lie (in detail)

3W Tells truth delayed, after question, scolding, punishment (whole truth)
3D Tells truth delayed, after questions, etc. (in detail)

4W Tells truth spontaneously after lie (whole truth)

4D Tells truth spontaneously after lie (in detail)

5W Tells truth after lie, extorted (whole truth)

5D Tells truth after lie, extorted (in detail)

6. Delays explanation, facing reality

8 Plans fantasy lie or deception

9 Lies after weighing the circumstances or after question, after fantasy lie
10 Lies spontaneously, deception

ll Constructive action and speaking by Susan

11: Constructive action by Susan gpd the mother

12 Makes spontaneous restitution

121 Restitution by Susan and the mother

13 Constructive thinking and feeling by Susan

13x Constructive thinking and feeling by Susan and the mother

14 Hostile, aggressive action, speaking by Susan

14: Hostile, aggressive action, Speaking by Susan and mother
15 Hostile, aggressive feeling, thinking by Susan

15: Hostile, aggressive feeling, thinking by Susan and mother
17 Regrets telling truth

19 Anxiety, emotion, fear

19x.Anxiety, emotion, by both Susan and the mother

21 Tension reduction

21x Tension reduction by both Susan and the mother

22 Begs, requests, leniency, pardon, forgiveness, sympathy

23
24
24a
25
26
27
28

41a
41b
42

45

55

56a
56b
56c
58a
58b
58¢

Submission: action, obeys

Submission: thinking and feeling (non-communication)

Susan regrets telling lie

Voluntary, spontaneous promise to mother (communication)

Promises herself, vows, (non-communication)

Promises after punishment, scolding, admonishing and forgiving

Negative statement pertaining to Susan, concerning some socially
unacceptable act

Categories for the Mother

Mother perceives from Susan: something is wrong, suspects
Mother perceives from non—Susan: ‘something is wrong, suspects
Mother questions: fact-finder, calls Susan for questioning
Mother communicates to Susan her distrust, accuses her
Seeks, or gets outside verification of facts

a) Communication to Susan

b) Non—communication

c) Without seeking, "Deus ex machine"
Mother behaves vaguely
Feeling of disbelief

Mother becomes or feels angry, excited
Miscellaneous, unhappy emotion, cries, feels sad

Mother accepts Susan, action (communication)
Mother's positive feeling, "understands"
Mother believes truth

Mother believes lie or deceptive maneuver

Deprivation of food (after Susan told truth)
Deprivation of food (no evidence of truth or lie)
Deprivation of food (after Susan told lie or deception)
Deprivation of playtime (after truth)

Deprivation of playtime (no evidence)

Deprivation of playtime (after lie)

60a Restitution under duress (after truth)

60b
60c
62a
62b
62c
64a
64b
64c
66a
66b

Restitution under duress (no evidence)
Restitution under duress (after lie)
Physical punishment (after truth)
Physical punishment (no evidence)
Physical punishment (after lie)
Mother tells father (after truth)
Mother tells father (no evidence)
Mother tells father (after lie)
Punishment by father (after truth)
Punishment by father (no evidence)

66c Punishment by father (after lie)

68a Susan sent to bed (after truth)

68b Susan sent to bed (no evidence)

68c Susan sent to bed (after lie)’

70a Miscellaneous punishment (after truth)
70b Miscellaneous punishment (no evidence)
70c Miscellaneous punishment (after lie)
72a Scolding (after truth)

72b Scolding (no evidence)

72c Scolding (after lie)

73 Susan sent to store, or to get material for repair

74 Mother makes restitution

75a Mother admonishes, re past behavior (after truth)

75b Mother admonishes, re past behavior (no evidence)

75c Mother admonishes, re past behavior (after lie)

76a.Mother admonishes, threatens, re future behavior (after truth)

76b Mother admonishes, threatens, re future behavior (no evidence)

76c Mother admonishes, threatens, re future behavior (after lie)

77 Mother shows tension reduction

78 Mother forgives Susan (power relation) _

81 Negative statement of mother's unhappy feeling or act
A.preliminary attempt was made to code Story No. 4, The Ink

Spots, using the same category definitions which had been develOped

for Story No. 2, The Lost Meat, simply substituting for the name

Michael the name of Susan. Psychological differences in the two

stories soon became apparent, however, which called for revision and

addition of categories. For example, one question asked after Story

No. 2 was ”How did Michael then feel about it?" In Story No. 4 there

was the question, "He? do they both feel about it?" This question

resulted in many statements in the children's stories which included

both parent and child behavior in one tally unit. This type of

statement occurred very infrequently in Story No. 2, and when it

did it was coded by giving a tally in the mother category and a

11.

tally in the Michael category, two tallkm for one verb unit. It was
thought that the statistical results would not be invalidated by this
treatment, since its occurrence was so rare.

In Story No. 4, however, the problem had to be dealt with more
directly, as these "both” statements were fairly frequent. If "both"
was given two tallies, one in the mother category and one in the Susan
category, there would be inflation of the data which would be unwarranted
according to the definition of units as one tally for each verb unit.

. Accordingly, separate categories for these statements were set
up. Categories in which it was possible for a "both" statement to
occur were supplemented by a separate category with an "1“ after it
for identification. For example, after category 11, construgtizg
agtion‘by Sus , category 111 was inserted to include the child's
statements of constructive action by Susan gag the mother. Similarly,
"1' categories were added to categories 12, l3, 14, 15, 19, and 21.
In the analysis of data.gggstrugtivg ggtggp,byp§g§an included tallies
in category 11, plus the tallies in category 111. Similarly, ggg:
gtgggtive actiog by thg_mothgg included tallies in category 52,
‘mgthg;,agcepts §p§gg, action (communication) plus tallies in category
111. W acting hr Sag-«an era the art”.

There are differences in reSponse to the given content of Story
Ho. 4 which are brought about by the degree of deve10pment of the
narrative, or the point at which the givens are terminated and the

child must proceed.

12.

In Story No. 4, unlike Story No. 2, the mother has entered the
picture when the child is in the midst of his misdeed so that some
of the categories in Story No. 2 providing for activity before the
child faces the mother are unnecessary. This is the case with category
18, Aggiety, gmotion, before mgggipg_3hg_ggthezg this category has,
therefore, been drapped from the manual for Story No. 4. In those rare
cases where the storywriter alters the givens to create anxiety before
the child faces the mother, it is coded as 19, Anxiety, emotigg, fggg.

‘Another difference in reSponses occurs because in Story No. 4
the storywriter might easily presume that the mother can see the facts
about the ink Spots there before her. The significance of 53333 and
11:3,in this story is therefore somewhat altered from that in The Lost
Meat story. There was at first some confusion in coding when Susan

was represented as telling §_part—truth ouly(i.e., telling her mother

 

that she Just wanted to try the coat on but not saying anything about

the ink spots), for it often seemed that the author may have been

presuming that the mother already knew the truth. It was decided,

then, that except in those cases where it was evident from the context

that Susan was trying to deceive, the child should be given the benefit

of the doubt, and part-truth would be coded the same as truth. Category

7, part-truth‘gnly,has been omitted from the manual for Story'No. 4.
Category 16, Disnlaced emotion_was used in Story No. 2 when

Michael became angry at the dog which had taken the meat. There was

no similar agent in Story No. 4, and this thema was not found in

any of the samples. Category 16 was therefore deleted from the manual

for Story No. 4.

Confusion and unreliability in the Mother categories was encountered

in coding 43a, Mothe; scolds, pgeS'verbal domination or abuse before

 

.tzuthngg.11g, and 43b, Mgthgz,usgs physiga; dgmination before truth
qulig. One coder might infer that the author of the story perceived

the truth as self-evident, and code verbal or physical domination
at the beginning of the author's story in one of the categories for
verbal or physical domination after truth or lie (categories 56 through
72, and categories 75 and 76); while another coder might say that
Elizabeth had not actually said either the truth or a lie and code
it as 43a and b, W mm mm m; Hg. The confusion was
resolved by omitting categories 43a and b and adding a third choice,
to each of the verbal and physical domination categories, namely,
M 1m m gvidmce as, 19, m 9;; L3. The numbering and '
lettering of the domination categories was changed for greater conven-
ience. In Story No. 2, for example, categories 56 and 57 were

56, Deprivation of food (after Michael told t ruth)

57, Deprivation of food (after Michael told lie).
In Story No. 4 this became

56a, Deprivation of food (after Susan told tinth)

56b, Deprivation of food (no evidence of truth or lie)

56c, Deprivation of food (after Susan told lie or deception).

A.source of unreliability in Story No. 2 had been the difficulty

in differentiating between category 49, Miscellaneous, gghgnnx emgtign
(communicatiog) and category 51, flgthgr'g negativg feeling, feel; sad

(gog—communicatiog). These were supposed to separate, among other

14.

things, communicated from uncommunicated behavior. But to draw a ‘
line between communicated and uncommunicated behavior was often very
difficult. For instance, "The mother looked sad" can mean that those
around.her know how she looks and she is communicating her feelings,
~or it can be meant as purely subjective feelings of the mother. It
requires an inference to code the statement as either #9 or 51.
Category 50, fiotheg.ggig§, is closely related to 49 and.5l, and it
seemed that there was little to gain from coding this one particular
thema separately. Hence #9, 50, and 51 were combined into 49,
gisgellaneous, unhappy emotion, which includes both communicated and

uncommunicated behavior.

IV. OUTCOMES OE‘STORIES

Each child's story was classified according to the outcome of the

story as a whole. The outcome classifications have been defined by

Anderson and.Anderson (3) and abstracted by Geierhaas (6) as follows:

HI.

nte ative° P oblem-so vi Outcom

The solving of the problem. There must be evidence of high
spontaneity in both the mother and her child; also of harmony
and working together, active communication and evidence of
mutual confidence. For harmonious cooPeration there must

be active communication in the place of mere thinking and
fantasy.

”D-R, ngination-Resistance Outcome

IA,

.A process of hostility and aggressiveness. There must be
evidence of mutual hostility, active tension, signs of working
against each other; the mother's hostility can be eXpressed

in the form of domination (punishment, scolding, exacting
promises), that she uses, and the child's resistance need

not be expressed orally or by action but simply on the think-
ing and feeling level.

flbiguous , Indecigive Ogtcome

There is lack of closure, low level of communication, story
is essentially unstructured. There is little or no inter—
action between the mother and the child. Problem has not
been faced, and no attempt has been made at its positive or
negative solution. Relation between parent and child little
changed at end from that at beginning. The experience did
not serve as a basis for social learning.

 

'D-S, Domination-Submission Outcome

ID,

NS,

Authoritarian person makes the decisions, has the answers.
Child merely submits, conforms, obeys. These stories show
mother's domination and Michael's subsequent submission.

Child accepts mother‘s verdict, submits to mother's ideas,

punishment, and often eXpresses his submissive acceptance
in the form of a promise, vow, or submissive action.

29mination, with no statement of resistance or submission

Submission, with no statement of domination.“

Subdivisions of submissiqn_and domination were used by Geierhaas
and, although tabulations have been made the data are not treated

by these subdivisions. In this study the outcome defined as domination-

 

submission will include domination outcomes with no evidence in the
story of resistance or submission, and submission outcomes with no

evidence in the story of domination.

Results
The results of these classifications are shown in Table 2 as
total numbers and percentages for the samples from each city.
The mass of the percentages of outcomes fell in the two lower
levels, ambiggous and domination-submission outcomes, where communica—
tion is generally one-way or altogether lacking. For four of the

I five samples, the largest percentages-of outcomes were in the domination-

 

submission classification. By contrast the Helsinki sample showed

significantly fewer figmination-submission outcomes than any of the

other samplings. The highest percentage of children in Helsinki

wrote stories with ambiguous. unstructured outcomes. Furthermore,

the percentages of ambiguous outcomes for the Helsinki sample were

significantly higher than percentages of ambiguous outcome for any

of the other samples. There were significantly fewer ambiguous outcomes

in the sample from.Birmingham.than if any of the other samples,
Without exception, the percentages of gggygtion-resistance out-

comes were lower than the percentages of integrative outcomes. The

Birmingham sample, in addition to having fewer ambiggous outcomes than

the other samples, had significantly more gpmination-resistange_outcomes

and significantly more integratixg outcomes than the other samples.

TABLE 2

OUTCOMES OF STORY #4. RDCUBHCIES 11:11) Pmcmfrgcns FOR 0511133321 111
EACH OF F1113 CITIES 1:1 TEL: OUTCOBES 37‘ 151;; n 3110121113 CLASS 1112:)
ACCORDING TO OUTCOMES OF I (Emma-arms, PROBIEI—L—SOLVING),
D-R (DQ-th’fl‘lON—RfilSISTANGIE), A (gs-BIGUOUS, Il-TDIJCISIVE),
mm Tm; SUM or D (1301-; memos), S SUBEIISSIO‘II),

 

 

 

.1111) D-S (DOE-1117.13IOl~T—SUISI~-1ISSIOH)
Bi mingmam H els ink'i Hamburg Iviuni ch Knoxvi lle
Out come 11 3% N ft N -;a II 5!. N p’
I 90 22.8* 32 15.1} 30 12.2 21 12.8 29 13.6
D-R f 25 16.3* 12 5.8 13 5.3 12 7.3 7 3.3
A 73 18.5* 93 “it-7* 84 34-1 55 33-5 59 27.6

D. S, and 13-3 207 52A 71 3L».1* 119 118.1» 76 116.3 119 55.6

¥

—.

Total 395 208 2% 161+ 211+
*Sié’nificzmtl-y different from the percentages of other population samplings
at the same level of outcomes, by chi-square test, at the .05 level or
better,

18.

Discussion

Birmingham outcomes were significantly different from the out—
comes of the other samples in three of the four outcome classifications.
Though the amount 0f.Q2EiBé£lQE.and submissiun shown in Birmingham
was similar to that of Munich, Hamburg, and Knoxville, the amount
of ambiguity was significantly less than that of the other samples,
while the amount of domination-resistance and Luggggggigu shown were
significantly higher than the other samples. As mentioned in the
introduction, Heber also found that Birmingham stories contained sig-
nificantly more resistance than the other samples.

In a culture tending more toward a democratic than an authoritar-
ian atmosphere one would expect more spontaneity, more social problem-
solving in conflict, more initiative and individual particiption
even in the form of resistance in conflict. The Birmingham children
have put these characteristics into their stories.

Inluuuination—submiguion outcomes the Birmingham children were
second only to the children in Knoxville in percentages, which is
not consistent with the hypotheses concerning findings in a democratic
culture. The percentage of children in Birmingham who wrote stories
with dumination-submission outcomes was not significantly different,
however, from the percentages of Children in.Hamburg and Munich.
Further understanding of this apparent inconsistency is given in the
discussion of cross-cultural comparisons and of types of domination.

Helsinki differed from the other samples in having significantly

lower percentages of domination—submission and significantly higher

19

percentages of ambiguous outcomes, while the percentages of domination—
zesistange and of integration in Helsinki were not very different

from percentages in Munich, Hamburg and Knoxville. Heber, too, found
the Helsinki sample in Story No. 2 lower than the other samples in

the category cluster of domination themas. The history and traditions
of Finland indicate that it is a land of a freedom—loving people,

who have a high sense of personal reaponsibility, honesty, integrity
and respect for the individual person. Their respect for the indiv-
idual may even be confused with a cautiousness in communication which
in this study is expressed in the significantly higher outcomes of
ambiguity and which is furthermore inconsistent with the hypotheses
about high participation in a democracy. In the outcomes of their
stories the Finnish children have revealed human relations between

the mother and child which, though low in domiuution and low in
submission, are also low in communication. The Finnish children are
second only to Birmingham in iuteggativg or problempsolying outcomes
and, though higher in frequency than Knoxville, Munich and Hamburg,

in that order, are not significantly different from the children

in these other samples.

The relative frequencies of outcomes for Munich and Hamburg were
quite similar to each other. Heber, in his cross-cultural comparisons
of clusters of themas, found that in many clusters Munich and Hamburg
were significantly different from each other, leading to the con-
clusion that the concept of a German national character is not neces—
sarily valid. It will be found in a later section of this thesis

that the analysis of categories for Story No. 4 yielded results for

Munich and Hamburg similar to Heber's results with Story No. 2. The
coding by categories is a finer, more detailed analysis and brings
out differences in Specific themas. Coding by outcomes is a more
gross measure and is not sensitive to some of the differences revealed
in the analysis of categories.

On the other hand, the coding of categories is a more static
analysis of the process of relating, with little or no regard for
time, sequence, or evolution of human relations as the story unfolds.
The outcome when coded as domination-submission is probably static
in the manner revealed by coding separate categories. If the story
moves toward a constructive, integrative end, it can be filled with
themas of domination, lying, evasion, or resistance up to the point
of release in the story into an integrative relating.

Knoxville had the greatest number of domination—submission out-

comes, quite contrary to the hypotheses and eXpectations,

V. THE BIRMINGHAM, ENGLAND SAMPLE
Sex Comparisons in Story No. 4 and Story No. 2

Categories by which Story No. # was coded have been grouped

into clusters of themas as described below.

Constructive behavior by mother

11: Constructive action by Susan and the mother

121 Restitution by Susan and the mother

13: Constructive thinking and feeling by Susan and the mother
52 Mother accepts Susan, action (communication)

53 Mother's positive feeling, "understands“

54 Mother believes truth

74 Mother makes restitution

Constructive behavior by Susan

ll Constructive action by Susan

11: Constructive action by Susan and the mother

12 Makes spontaneous restitution

12x Restitution by Susan and the mother

13 Constructive thinking and feeling by Elizabeth

131 Constructive thinking and feeling by Susan and the mother

Susuu tells truth

1W Tells truth spontaneously, the whole truth
1D Tells truth spontaneously, in detail
2W Tells truth after conflict, anxiety, after fantasy lie (whole truth)
2D Tells truth after conflict, anxiety, after fantasy lie (in detail)
3W Tells truth delayed, after question, scolding, punishment (whole truth)
3D Tells truth delayed, after questions, etc. (in detail)
Resistancg

1% Hostile, aggressive action, Speaking by Susan

1&1 Hostile, aggressive action, speaking by Susan and mother
15 Hostile, aggressive feeling, thinking by Susan

15x Hostile, aggressive feeling, thinking by Susan and mother
17 Regrets telling truth

22.

Anxiety

19 Anxiety, emotion, fear

19:.Anxiety, emotion, by both Susan and the mother

21 Tension reduction

21: Tension reduction by both Susan and the mother

22 Begs, requests, leniency, pardon, forgiveness, sympathy

Submission

23 Submission: action, obeys

2“ Submission: thinking and feeling (non-communication)

24a Susan regrets telling lie

25 Voluntary, spontaneous promise to mother (communication)

26 Promises herself, vows, (non-communication) ‘

27 Promises after punishment, scolding, admonishing and forgiving

28 Negative statement pertaining to Susan, concerning some socially
unacceptable act

Domination

56a Deprivation of food (after Susan told truth)

56b Deprivation of food (no evidence of truth or lie)
56c Deprivation of food (after Susan told lie or deception)
58a Deprivation of playtime (after truth)

58b Deprivation of playtime (no evidence)

58c Deprivation of playtime (after lie)

.60a Restitution under duress (after truth)

60b Restitution under duress (no evidence)

60c Restitution under duress (after lie)

62a Physical punishment (after truth)

62b Physical punishment (no evidence)

62c Physical punishment (after lie)

66a Punishment by father (after truth)

66b Punishment by father (no evidence)

66c Punishment by father (after lie)

68a Susan sent to bed (after truth)

68b Susan sent to bed (no evidence)

68c Susan sent to bed (after lie)

70a Miscellaneous punishment (after “truth)

70b Miscellaneous punishment (no evidence)

70c Miscellaneous punishment (after lie)

72a Scolding (after truth)

72b Scolding (no evidence)

72c Scolding (after lie

73 Susan sent to store, or to get material for repair
75a Mother admonishes, re past behavior (after truth)

23.

75b Mother admonishes, re past behavior (no evidence)

75c Mother admonishes, re past behavior (after lie)

76a Mother admonishes, threatens, re future behavior (after truth)

76b Mother admonishes, threatens, re future behavior (no evidence)

76c Mother admonishes, threatens, re future behavior (after lie)

The number of children in each sample who gave one or more
reSponses in each of these clusters was counted; that is, no child
was counted twice in any one cluster. The number of children in each
cluster is then expressed as a percentage of number of children in
the total sample. Table 3 shows the numbers and percentages of boys
as compred to girls who gave responses in each cluster for Story
No. 4 and for Story No. 2 (adapted from Heber's study). Numbers and
percentages of children's stories coded at the four levels of outcomes
are also shown in Table 4. Only integrative outcomes were reported
for Heber's study of Story No. 2.

As seen in Table 3, of the eight category clusters for Story
No. 4, about Susan and her mother, the percentages of girls were
greater in seven clusters and significantly greater in five. For
Story No. 2, The Lost Meat, which concerned a boy and his mother, of
the same eight category clusters, percentages of girls were higher
in seven clusters and significantly higher in one cluster.

In Story No. 4, the thema cluster which showed the greatest
difference was submission, where the girls had significantly more
reSponses than boys at better than the .001 level by chi-Square test.
Other themas in which girls gave significantly more reSponses than
boys, at the .01 level, were resistance, anxiety, constructive

behavig: by Susan, and gpnstructive behavior by mother. Two of the

significant differences between boys and girls in Story Do. 4, in

TABLE 3

BIRMINGHAM, EHGLLND: DUMBERS AND PERCENTAGES OF BOYS ADD OF
GIRLS WITH THEEAS CODED IN EIGHT CATLGORY CLUSTHRS, AND
ACCORDING TO CLASSIFICRTION OF OUTCOMES OF
STORIES HO. 4.AND NO. 2

 

 

 

 

Story Ho. 4 Story Ho. 21

Boys Girls Boys Girls
Category Cluster H s N s N s N 3
Susan tells truth 113 51.4 82 46.9 127 57.4 103 58.2
Susan tells lie 51 23.2 46 26.3 61 27.6 62 35.0
Resistance 28 12.7 43 24.6* 46 20.8 33 18.6
Anxiety 180 77.3 156 89.1* 120 54.3 108 .61.0
Submission 76 34.6 113 58.9* 87 39.4 118 66.7*
Domination 146 66.4 130 74.3 171 77.4 141 79.7
Constructive (child) 64 29.1 76 43.4* 15 6.8 28 15.8‘
Constructive (mother) 120 54.6 122 69.7* 31 14.0 33 18.6

Outcome
Integrative 41 18.6 49 28.0* 10 4.5 8 4.5
Domination-resistance 17 7.7 10 4.6 (Not reported by Heber)
Ambiguous 51 23.2* 2 12.6 " " " “
Domination-submission 111 50.5 94 53.7 " " " "
Total N 220 175 221 177

 

*Significantly higher than the freq. for the other sex by chi-square test
at the .01 level or better.

1
Data for Story Do. 2 adapted from Heber (8)

25.

 

 

the category clusters for submissigp and constructive behavior by
moths; were more frequent in Story No. 2 for girls, though not sig—
nificantly so. The thema cluster of resistance was different in
Story No. 2, where more boys gave resistance reSponses than girls.
The distribution of tgpth themas in the two stories was in

Opposite directions for boys and girls, the boys being higher in
Story No. 4, and the girls higher in Story No. 2. The distribution
of 1;; themas and domination themas was in the same direction for

otb stories, the girls being higher in both, though these differ—
ences were not significant.

In Story No. 4, the girls gave significantly more integrative

 

outcomes than boys, while in Story No. 2 there was no difference
between boys and girls.
Ambigugus outcomes in Story No. 4 were significantly more fre-
quent for boys than for girls. For ggminotion-rosistancg and domination-

submission outcomes boys and girls were not significantly different.

Discussion of Sex Comparisons

2 seven of the eight clusters for Story No. 4, and in seven of
the eight clusters for Story No. 2, frequencies for girls were higher.
This is consistent with well-known observations that girls deVFlOp
verbal facility earlier and are generally superior to boys in verbal
fluency. It will be reported later that girls tended to write longer
stories than boys. In writing more, the chances are greater that girls
will supply a greater varie ty of themas, which seems to have been the

case in Table 3.

In Story No. 4, the five category clusters in which girls were

0‘

significantly higher than boys were: constructive ehavior by mgthgz,
gongtructivc behavior by_§u§gn,_rcsistance, anxiety, gag submission.
The fact of five significant differences related to the story about a
girl and her mother in contrast with only one significant difference
in the story about a boy suggests that girls may have been more
involved in Story No. 4 than in Story No. 2.

There were large differences in several of the category clusters
between the level of p roentages of reaponse for Story No. 4 and
Story'No. 2.

Both boys and girls gave much higher percentages of reaponse for

Story No. 4 to the category clusters constructive behavio; by mgther,

gonstrugtivg behavior bz,the ghild, and anxiety than for Story No. 2.

 

The differences in structure of the two stories accounts in part for
the differences in these category clusters. These are the category
clusters in which many "both" themas were given in the children's
stories. For erample, if the child wrote "The mother and Susan both
felt sorry about the ink spots," tallies were counted for Susan and
for the mother.

p1; and,tgu£h,category clusters were lower in percentages in
Story No. 4 than in Story No. 2. This was probably because the children
who completed the stories often assumed that the mother in Story No. 4
could see Susan wiping the ink spots out of the coat and at once knew
the truth, so there was not as much need for Susan to tell either the

truth or a lie. In Story No. 2, however, there was a time interval

between the loss of the meat and meeting the mother. The children who
wrote the stories had to make up some way for the mother to find out
about the lost meat, and it was usually by having Michael tell her,
either truthfully, or with a lie.

Submission and domination category clusters had lower percentages
in Story No. 4 than in Story No. 2. This, too, may be because the
structure of Story No. 4 and the nature of the questions asked after
it were conducive to more themas of constructive behavior and to fewer
of submissive and dominative behavior.

Integrative outcomes were more frequent for Story No. 4 than for
Story No. 2. This is probably because working together of mother and
child is called forth by the questions, and working together often
results in problem-solving. a part of the definition for integrative
behavior.

In all these changes in nercentages except in the tguth,and
resistance clusters, the relationship of reaponses between boys and
girls remained constant; where the girls were higher in response than
boys in the clusters for Story No. 2, they were also higher in Story
No. 4. This consistency attests to the reliability of the Stories as
a measuring instrument.

In the tggth cluster, boys were higher in.percentage of resPonse
in Story No. 4, and girls were higher in percentage of reSponse in

Story No. 2, but differences were small and not significant.

 

There was an inconsistency in the resistance cluster, where boys'
responses were higher in Story No. 2 and girls' responses were signif—

icantly higher in Story No. 4. Boys tended to show more resistance

 

[é

 

I
“ €
‘ v
v l
I
. I 1 C Q
0
v
.(
O '
S . c
O
a
h ' ' . .
o
I V
f C
. 1
‘
r
. g I
t - ’ ‘
V
I - .
— v
I
o g I

in a story about a boy and girls showed more resistance in a story
about a girl. It is noted, however, in Table No. 3, that in the out-
come of dominationpresistance in this story about a girl the frequencies

were higher for boys, though the difference was not significant.

Socio-economic Comparisons

Table 4 gives the numbers and percentages of children from high
and low socio-economic levels in Birmingham with themas coded in the
several category clusters, and according to outcomes for Stories No. 4
and No. 2. As seen in Table 4, the high socio-economic group gave
higher percentages of response to all he clusters of themas except
the jggfih,cluster. For Story No. 4 in one category cluster, assistance,
the high socio-economic group was significantly higher at the .001
level, by chi-square test. In another category cluster, ghild tells
'lig, the high socio—economic group was significantly higher at the
.10 level. Stories of children in the high socio-economic group were
more frequently given integrative, domination-resistance, and dgminatign-
submission outcomes.

The low secio—economic group was higher in percentage of response
only in the 1;; cluster. More ambigpgus outcomes were given in the
low socio-economic group.

The differences between high and low socio-economic levels found
by Heber in Story No. 2 can be said to be consistent with the differ-
ences reported in Table 4 for Story No. 4. The three significant
differences reported by Heber are confirmation of differences in the

same direction for Story No. 4. The inconsistencies in socio-economic

29.

TABLE 4

BIRMINGHH’Z , HIGH-JD: NUMBERS All D PERCY“1 {PAGES OF CHILDREN FROM
HIGH AND L03 SOCIO—ECOHOMIC LEVELS WITH THEEAS CODED
IN THE SEVERAL CATEGORY CLUSTEP‘, 53D ACCORDIHG
TO OUTCOKES FOR STORIES NO. 4vAED NO. 2

 

 

Story no. 4 Story No. 21

High Low High Low
Category Cluster H % N t N 6 N

l I

o'fl

 

 

Constructive (mother) 173 62.2 69 59.0 54 19.3* 10 8.5
Constructive (child) 104 37.4 36 30.8 36 12.9* 7 5.9
Child tells truth 130 46.8 65 55.6 162 57.9 68 57.6
Child tells lie 75 27.0 22 18.8 84 30.0 39 33.1
Resistance 62 22.3* 9 7.7 67 23.9* 12 10.2
Anxiety 233 83.8 93 79.5 162 ~57.9 66 55.9
Submission 128 46.0 51 43.6 144 51.4 61. 51.7
Domination 196 70.5 80 68.4 217 77.5 95 80.5
Outcome

Integrative 65 23.4 25 21.4 15 5.4 3 2.5
Domination-resistance 21 7.6 4 3.4 (Not reported by Heber)
Ambiguous 45 16.2 28 23.9 ” " " "
Submission, Domination, 14? 52.9 60 51.3 " " " "

and Domination—sub—

mission

Total 278 117 ‘ 280 118

 

*Significantly different from the other socio-economic level at the .05
level or better, by chi-square test.

1Data for Story No. 2 adapted from Heber (8).

30.

BHEDEHGIHG ED:

TABLE 5

IIJLBIRS RID PERCBUTnGQS
CODED IN EIGHT CATQGORY CLUSTERS, ADD INTLSRhTIVE OUTCOELS,

3 0211133131: WITH T5335

0F STORIES no. 4 ill-Tel) NO. 2

Story No. 4

C

Story No. 2

,J

 

 

Category Cluster N p N p
Constructive (mother) 186 47.1 64 16.1
Constructive (child) 14° 35.9 ‘43 [0.8
Susan tells truth 194 49.1 230 57.8
Susan tells lie 97 24.6 123 30.9
Resistance 71 18.0 79 19.8
Anxiety 326 82.5 228 57.3
Submission 179 45.3 205 51.5
Domination 276 69.9 312 78.4
Integrative outcome 90 22.8 18 4.5

Total N 395 , 398

 

31.

comparisons between the category clusters for Story No. 4 and Story
No. 2 are small and represent non-significant differences.

In general, there were more children in the high socio-economic
level than.were found.among the children in the low socio—economic
level giving responses in the several category clusters. Exceptions
represented only small and non-significant differences.

The differences between Story No. 4 and Story No. 2 in levels
of percentages with clusters which were noted.and discussed under

sex comparisons are also apparent in the socio-economic comparisons.

Story No. 4 — Story No. 2 Comparisons

Table 5 gives the total frequencies and percentages of responses
in all the clusters of themas and in integrative outcomes for Story
Be. 4 and for Story No. 2 for the Birmingham sampling. Percentages
for Story No. 4 were significantly higher at the .001 level, by chi-
square test, for the thema clusters gonstructive behavior by gggggg,
constructiv; behavio; 1m;jhg,ghild, and for integrative outcomes.
Percentages of response for Story No. 4 were significantly higher at
the .01 level for the thema cluster anxiety.. Thema clusters inwhich
percentages of response were leer for Story No. 4 than for Story
Be. 2 were: 5211; taugh (.02 level), 3511; ii; (.05 level), snbmission
(.10 level), and domination (.01 level). Results of only one thema

cluster, resistanc , were not different at any level of significance.

Discussion

There were significant differences between Story No. 4 and

Story No. 2 in frequency of response in all clusters of themas except

32.

one. Some of the reasons for the differences have been pointed out

in the discussion of sex differences. It is to be shown later that

for the most part the differences occur consistently in all the samples,
so that the relationship of each sample to the other samples is fairly
stable from one story to the other, within each cluster of themas.

The stories are structured differently and they evoke different
responses from the children. In Story No. 4 the mother is already
present, the truth is practically self-evident. A: long as there is
no chance of evading, the situation demands that something be done to
rectify it. This is not the case in Story No. 2, where Michael is
alone when he discovers his dilemma -- there is greater opportunity
to take devious routes to a solution. This difference in the nature
of the Incomplete Stories accounts in part for the fact that so many
children reaponded differently to Story No. 4 than they did to Story

No. 2.

1‘2-

VI. CROSS-CULTURAL COMPARISONS

Category C uster: anstructive thinking, e-lir , and agtign
by gather. Table 6 presents the numbers and percentages of children
who wrote themas of constructive bghgzig; by th§_mgthgz in each of the
five samples for Story No. 4 and Story No. 2. Hamburg children gave
the highest percentage of reaponse, and it was significantly higher
than the Knoxville and Munich samples at the .05 level or better. The
order of the samples from highest to lowest percentages of response
is exactly the same as that found by Heber for Story No. 2. Heber found
the percentages of response in Hamburg, Helsinki, and Birmingham
samples to be significantly higher than the Knoxville and Munich

samples, at a .05 level or better.

TABLE 6

nunssas.snn Psacsxiscss or CHILDREN IN’EACH or THE SAMPLINGS
WITH THEMAS conan IN THE CATEGORY CLUSTER, CQNSTRUCTIEE
zglrxing, gastlnc,.igg_gggigg_§1 Morass
r03 STORIES no. 4vAND NO. 2

 

 

 

 

STORI Ho. 4 STORY N0. 2
Size of Size of
git: . Sammde ~‘;§ 1% City Sample gi_ gfi
Hamburg 246 139 56.5 Hamburg 281 5n 19.2
Helsinki 208 102 49.0 Helsinki 209 35 16.7
Birmingham 395 186 n7.1 Birmingham 398 64 16.1
Knoxville 214 99 46.3 Knoxville 217 21 9.7
Munich 164 72 43.9 Munich 165 10 6.1

 

Catfegogz Cluster: Constructive thinking, feeling, gag ggtign by
M. The numbers and percentages of children who wrote theme of
constructive behavior by the child in Story No. 4 and Story No. 2 are
shown in Table 7. The absolute level of percentages is much higher
for Story No. 4 than for Story No. 2.. None of the differences between
samples was significant in either story. The rank order of samples

from highest to lowest percentages was different in the two stories.

TABLE?

NUMBERS AND ERCELMGES OF CHILDREN IN EACH 03‘ THE SAMPLINGS
WITH Twas COD’ED IN THE CATEGORY CLUSTER, CONSTRUCTIVE
W ELLEZQ, mm ACTION LY. SUCn‘xN
IOR STORIES 130.14 AND NO. 2

 

 

STOBI NO. 4 STORY NO. 2

 

Size of Size of
City Sample 11 5% City Samb_1__e L 56
Knoxville 214 79 36.9 Birmingham 398 43 10.8
Birmingham 395 140 35.4 Hamburg 281 27 9.6
Helsinki 208 71 34.1 Munich 165 13 7.9
Hamburg 246 78 31.7 Knoxville 217 14 6.5 ‘
Munich 164 49 29.9 Helsinki 209 9 4.3 '

 

Categog Cluster: M gills. £13113. Table 8 shows the numbers
and percentages of children in each sample who wrote themas of Sugar;
1911; m, for Story No. 4 and Story No. 2. All the samples except
Munich showed for this category cluster fewer responses in Story No. 4
than in Story No. 2, but the differences were not large nor significant.
The percentage of reSponss in the Knoxville sample was significenitly

higher than the percentage for any of the other samples at the .05

35 .

level of significance by chi-square test. The percentage of reSponse
in the Munich sample was significantly lower than the percentages
in the Birmingham, Hamburg and Knoxville samples. In Story No. 2,
the rank ordering of samples by percentage of response, from highest

to lowest, was Knoxville, Birmingham, Hamburg, Helsinki and Munich.

TABLE 8

mmsas m PHHCHrTACHS or CHILDREN IN EACH or THE SAMPLINGS
WITH THELHS 00133.11 IN TH E CATEGORY CLUSTER, SUSAN TELLS
TEETH, FOP. STCHIHS NO. 4 AND 110. 2

 

 

 

STORY N0. 4 STORY N0. 2
Size of Size of
City Sample H 7‘6 City Sample 11 %

Knoxville 214 130 60.7 Knoxville 217 140 64.5
Hamburg 246 127 51.6 Birmingham 398 230 57.8
Birmingham 395 194 49 .1 Hamburg 281 148 52.?
Helsinki 208 89 42.8 Helsinki 209 104 49.8
Munich 164 6+ 39 .0 Munich 165 58 35 .2
m Cluster: _S_1,3_s__a,n tellg lie. The percentages of children

in each sample who wrote themas of £9533 tell; m are shown in

Table 9. In all samples there were lower frequencies of response

in this thema for Story No. 4 than for Story No. 2. In both stories
Munich was significantly higher than all the other samples in frequency
of mm 3.2.11.3. Li; themas. In Story No. 2 Hamburg was found to be
significantly higher than~ Helsinki and Birmingham and these in turn

were significantly higher than Knoxville. In Story No. 4 Hamburg and

36.

“to.

,\

-_-.- -.

-I
n -.
‘ v
.
p..-
.
4
-
fl“

—

_.-_

-_.- .-

V
I
O
a"
..
' O

Birmingham were significantly higher than Helsinki and Knoxville in
frequency of reaponse of child tells lie. Knoxville was lowest in

both samples.

TABLE 9

NUMBERS.AND PERCENTAGES OF CHILDREN IN EACH OF THE SAMPLINGS
WITH THEMAS CODED IN THE CATEGORY CLUSTER, SE§AE_TELLS LIE,
FOR STORIES N0. 4vAND NO. 2

 

 

 

 

STOHY'NC. 4 STOBI NO. 2
Size of Size of
City Sample N 5 City Sample H %
Munich 164 61 37.2 Munich 165 106 64.2
Hamburg. 246 68 27.6 Hamburg 281 123 43.8
Birmingham 395 9 24.6 Helsinki 209 68 32.5
Helsinki 208 35 16.8 Birmingham 398 123 30.9
Knoxville ‘214 30 14.0 Knoxville 217 42 19.3
Qgfiggggy'Clustez: Resistancg. In Table 10 are shown the frequencies

and percentages of children in eadh sampling who wrote a thema of
zgsistance for Story No. 4 and for Story Ho. 2. For Story Ho. 2 the
Birmingham sample was significantly higher in frequency than the other
samples. For Story No. 4-Birmingham had the highest frequency, but

it was not significantly higher than any of the other samples except

the lowest one, Knoxville. It was significantly different from Knoxville
at the .05 level by chi-square test. The order of ranking of samples
was different for Story No. 4 than for Story No. 2 but did not contra-

dict the one significant difference found in Story No. 2.

87.

. I
p . . .. . ‘ ‘ ., .ga .
Q - 7' . 4 _,.., - '
.
. 4. -
. V “\ . o
c A k I i A ‘
o
-._.. ...
, . . . _-
a 7'. . . ,~ . e.
- . Y ‘
C . e
. . . . k . . ..
. . _—
.
- fl
‘ .
P-.. - -.- A ..-_l. - ~___
' e
0 o
' o
‘ o
.4 ., ~
. . . k i ,,
.. _ F”-__..- 7-. -
. . I
‘. : ‘ . . .- -
' ‘ ‘ .u . ‘ .
C
- O ‘ H . v ' ’
'.- q .. _»_
. .- ’r~
. . . C -s- .0
. . .
.‘ t ‘ n . ~ ‘1: - v' ~-
. . _. . . .
. 4
. , .
-
n _ ‘ . e
o .. '~ ' ‘ u
‘7‘ ' D .. - . -
.- ~.. _. ‘ '
‘ _ - - .
. _ , .
i- - V . .

TABLE 10

NUMBERS AND PERCL AGES OF CHILDREN IN EACH OF THE SAMPLINGS
UITH.THEMAS CODED IN THE CATEGORY CLUSTER, RESISTANCE,
FOR STORIES N0. 4vAND N0. 2

 

 

 

STORY HO. 4 STORY N0. 2

Size of Size of
City Sample I! % City Saniple N )4: ~‘
Birmingham 395 71 18.0 Birmingham 398 79 19.8
Helsinki 208 34 16.4 Hamburg 281 29 10.3
Hamburg 246 28 11.4 Knoxville 217 20 9.2
Munich 164 18 11.0 Helsinki 209 18 8.6
Knoxville 208 15 7.0 Munich 165 12 7.3

 

Qateggzz Clustez: Angigty. The frequency and percentages of
response in each sample for the category cluster of gpgigtz are shown
in Table 11 for Story No. 4 and Story No. 2. The absolute level of
the percentages was much higher for Story No. 4 than for Story No. 2.
In Story No. 4 the Birmingham sampling had a higher percentage of
reSponse than the other samples, significant at the .10 level by
chi-square test. The percentage of the Helsinki sample was higher
than the Hamburg and Knoxville samples, significant at the .02 level,
and the Hamburg and Knoxville samples in turn were higher than the
Munich sample, significant at the .01 level. For Story No. 2 the
Birmingham sample had the highest percentage, but it was not signif-
icantly higher than the Hamburg and Helsinki samples. Birmingham,
Hamburg and Helsinki percentages were significantly higher than the

Munich percentage for Story No. 2, and the Munich.percentage was

38.

-._ -0—

 

.
..g
a.
pew
‘a
.v
a“.
..
~ -
.
. -..-

in turn significantly higher than the Knoxville sample. The significant
differences found in Story'No. 2 for this cluster, where Munich was
significantly higher than.Knoxville in percentage of response, was
contradicted in Story Ho. 4, where Knoxville was significantly higher

than Munich in percentage of reaponse.

TABLE 11

NUMBERS ANDPERCENTAGES OF CHILDREN IN EACH OF THE SAMPLINGS
WITH THEMAS CODED IN THE CATEGORY CLUSTER. NXI T ,
FOR STORIES N0. 4-AND NO. 2

 

 

STORE'NO. u srosx NO. 2
Size of Size of
City Sample N at City Sample N %

 

Birmingham 395 326 82.5 Birmingham 398 228 57.3

Helsinki 208 160 76.9 Hamburg 281 152 54.1
Hamburg 246 16L» 66.7 Helsinki 209 110 - 52.6
Knoxville 214 140 65.4v Munich ' 165 63 38.2
Munich 164 81 49. Knoxville 217 28 12.9

 

Categggy Clustegx Submission. The frequencies and percentages
of children who gave a response of gghmiggigg in Story No. 4 and in
Story No. 2 are shown in.Table 12. The frequencies were all much
higher for Story Ho. 2 than they were for Story No. 4. The Birmingham
sample gave a higher frequency of response than the Knoxville and
Helsinki samples, which in turn gave higher frequencies than Hamburg
and Munich. Differences were significant at the .01 level by chi-
square test. For both stories, the Knoxville and Helsinki samples
were significantly higher than the Munich sample in frequency of
response to the submission thema. The significant difference found

in Story Ho. 2 was therefore not contradicted in Story No. 4.

TABLE 12

mamas mnmscmmsss or CHILDREN IN EACH or THE smmmss
WITH rssmas consn IN THE carsoonr CLUSTER, SUBMISSION,
FOR STORIES no. 4 Am) N0. 2

 

STORY HO. 4 STOHI NO. 2
Size of Size of
City Sample N % City Sample 3 f

 

Birmingham 395 179 45.3 Knoxville 217 124 57.1

Knoxville 214 72 33.6 Helsinki 2209 115 55 .o
Helsinki 208 68 32.7 Hamburg 281 148 52.7
Hamburg 246 53 21.5 Birmingham 398 205 51.5
Munich 164 32 19.5 Munich 165 74 44.8

 

Categogz Clustgr: Dominatiog. Fre quencies and percentages of
reSponse of the children in the five samples with a thema of domination
are shown in Table 13. The frequencies of the Birmingham, Munich and
Hamburg samples, in that order, were significantly higher at the
.05 level than the frequencies of the Knoxville sample, which was in
turn higher than the frequency of the Helsinki sample, significant
at the .001 level. For Story No.l+, also,Birmingham was highest and
significantly higher than all other samples in frequency of reaponse;
there were no other significant differences. In this cluster, too,
the significant difference found in Story No. 2 was not contradicted

by results of Story No. 4.

TABLE 13

NUMBERS AND PERCERTAGRS OF CHILDREN IR EACH OF THE SAMPLINGS
WITH THEMAS CODED IN THE CATEGORY CLUSTHR, DOMIIATION.
FOR STORIES NO. 4.KND NO. 2

 

 

 

STOBI Ho. 4 STORY Ho. 2
Size of Size of
City Sample H 2 City Sample H %
Birmingham. 395 276 69.9 Birmingham 398 312 78.4
Munich 164 108 65.9 Knoxville 217 141 65.0
Hamburg 246 159 64.6 Hamburg 281 171 60.9
Knoxville 214 116 54.2 Helsinki 209 121 57.9
Helsinki 208 78 37.5 Munich 165 93 56.4

 

Table 14 gives for Story No. 4 the average number of thema tallies
per story in each of the samples, according to sex groups, socio-
economic groups, and, for the Munich and Hambung samples, sex compar-
isons of Stories No. 4 and No. 2. In all of the samples, girls wrote
longer stories than boys. In all of the samples except Hamburg, the
high socio—economic group wrote longer stories. In Hamburg and Munich,
girls wrote longer stories for Story No. 4 than for Story No. 2. In
Munich, boys wrote longer stories for Story No. 2 than they did for
Story No. 4. In Hamburg, boys wrote slightly Shorter stories for

Story No. 2 than they did for Story No. 4.

Ln -

when a 1'4

.AVERAGE NUMBER.OF THEMA TALLIES P33 STORY 303 SHORT HO. 4 TN
EACH OF THE SAMPLES ACCORDING TO SEX GROUPS AND
SOCIO—ECONOMIC GROUPS: FOR THE MUNICH AND
HAMBURG SAMPLES. SEX COMPARISONS OF
STORIES N0. 4'AND NO. 2

 

 

Sex Comparisons

Boys Girls
Birmingham 9.8 155.5
Helsinki 5.6 8.1
Hambung 7.4 10.1
Munich 4.7 8.8
KnoxVille 6.5 706
Socio—economic Comparisons
High 110w
Birmingham 13.0 10.7
HB1 8111121 8.? 5.8
Hamburg 8.9 6.8
Munich 7.1 6.9
Boys Girls
Story 4 7.4 10.1
Hambung
Story 2 700 903
Stnry L" be? 808
Munich
Story 2 5.6 6.6

 

Discussion of cross—cultural Comparisons

One question which might come to mind about the Anderson Incomplete
Stories, and about any projective device, for that matter, is, Do the
children give consistent responses from story to story in the series?
.Another question: Do children make up a plot for one story, and then
just repeat the same plot in the rest of the stories?

A partial check on these questions was made by comparing the out-
come level of Story No. 2 for each child with the outcome level for
his Story No. 4. In the total sampling of 1,227 children, there were
292 children whose outcomes for Stories No. 2 and No. 4 were coded at
the same level. This means that the preponderance of the children did
not repeat the same story for Story No. 4 that they had written for
Story No. 2.

Now the question arises, If children write different stories from
one to another of the.Anderson Incomplete Stories, can there be con-
sistency in the results? The data in Tables 7 through 14 indicate that
there was consistency.

Tables 7 through 14 have presented numbers and percentages of
children in each of the hamplings‘with themas coded in the eight category
clusters, for Stories No. 4 and.No. 2, respectively. In comparing
the percentages of the samples relative to each other, it was found
that the results for Story No. 4 were in general agreement with the
significant results found for Story No. 2, with only one exception.

The following statements may be made for both stories:

(a) In ggngtnlgtive behavio; p1 _t__h_e_ m, Hamburg, Helsinki, and

Birmingham had higher percentages than Knoxville and Munich:

43.

(b)

(e)

(d)

(a)
(f)

(s)

In my}: _t_;l_1_s_ m, Knoxville had the highest percentage, and
Munich had the lowest percentage;

In ghild_jgllg_ ie, Munich, then Hamburg, had the highest percent-
ages, and Knoxville had the lowest:

In zgsistancg, Birmingham had the highest percentage:

In anxigty, Birmingham had the highest percentage;

In submission, Knoxville had the highest percentages for Story
No. 2 and second highest in Story No. 4; Munich had the lowest in
both stories;

In dominatigg, Birmingham had the highest percentages, and Helsinki
had the lowest percentage for Story No. 4 and the second lowest

for Story No. 2.

The one contradiction occurred in the agxietx category cluster,

where Munich percentages were significantly lower than Knoxville for

Story'No. 4, and Knoxville percentages were significantly lower than

Munich for Story No. 2.

Consistencies as numerous as these support the proposition that

the children in a sample may change their stories individually, but

when the results of the whole sample are taken at once, those charb

acteristics which differentiate the sample from other samples are

maintained from story to story.

Further support of this proposition was found in the analysis of

sex differences in the Birmingham sample in Chapter V of this study,

where the differences between boys and girls were for the most part

maintained from Story No. 2 to Story No. 4.

The changes in the levels of the percentages for many of the
category clusters from Story No. 2 to Story No. 4 appeared here in
all the samples as it did in the analyses of Birmingham sex comparisons
and socio—economic comparisons. To review the discussion of reasons
for the changes in levels, the cw W by the ot er,
constructive behavigr :91 3h; g_h_i_1_d_ and w category clusters were
thought to be raised in The Ink Spots story by the many statements
in the children's stories involving behavior of both mother and child.
The my; and 1.12 clusters were thonght to be at a lower level of
percentages in The Ink Spots story because the storywriter begins at
a point where the mother can be construed to know the truth already,
whereas in The Lost Meat story the mother has to be informed in some
manner about what happened.

The domiggtiog and gubmiggign clusters were lower in level of
percentages because the structure of Story No. 4 and the questions
asked after it are conducive to more themas of working together of
mother and child and consequently less domination and submission.

It may be noted that the zggigtgngg category cluster remained at
the same level in both stories. The factors which Operated to change
the levels of the other clusters did not affect this cluster.

Just as the differences between boys and girls for the Birmingham
sample were maintained for the most part in spite of the shift in
levels of percentages for the category clusters, so also the relative
positions of the cultural samples were well maintained.

Birmingham. The analysis of category clusters for Birmingham

has shown that for both stories the Birmingham sampling falls near

45.

the middle of the range of percentages with respect to the other
samplings in mm M Q}; Ml. m, and 1;; clusters.
In Story No. 4 Birmingham came second to the highest in ggpgtgugtigg
behavio; pygphg,ghild, but differences in this cluster were not signif-
icant. Themas of g2n§t§2g3113_3§hg1123, of 1gpth, and of 1;; are

not differentiating factors for the children in the Birmingham sample.
This neither confirms nor denies the hypotheses about behavior in a
democratic culture.

The Birmingham children gave the highest frequencies of response
for category clusters for resistance and domination in both stories.
This is consistent with the results for outcomes discussed in Chapter
IV. The Birmingham sample had the highest frequency of dominatiog-
sugmission outcomes and was second only to Knoxville in domination-
submissiog and was not significantly different from Munich, Hamburg,
and Knoxville in domination-submisgigg outcomes. The resistance
shown with highest frequencies by the Birmingham children is a higher
level of relating to domination than submissive behavior shown by
children who do not resist.

In the category cluster of anxiet , the Birmingham sample had
the highest frequency, contrary to the hypotheses. Included in the
anxiety cluster is category 22, bggg‘ re uests, lepiency, forgiveness.
Statements in which Susan said she was sorry or that she did not
mean to do it were coded in this category. This type of statement was
very frequent in the Birmingham sample and is probably the reason

Birmingham was highest of the samples in this cluster.

In Story No. 2 the irmingham children gave second to the lowest
frequencies of response for the category cluster for suhmigsign. In
Story No. 4, contrary to this finding and to the expectations set up
by the hypotheses, the Birmingham sample was the highest in frequency
of submission themas. Included in the submissign_cluster in category
24, submissign: thinking and feelin , in which were coded statements
that Susan was sorry that she her3 Spoiled her mother's new coat.

One section of Hypothesis No. 3 in Chapter I states, ”Children
in the more authoritarian culture will...show significantly fewer
responses of...sense of fair play...,” which statement implies that
children in a democratic cultzre will show more reSponses of fair play.
Story No. 4 is different from Story No. 2, because the new coat in
Story No. 4 is a person 1 property of the mother's, something which
she doubtlessly values. If the child writing the story possesses a
"sense of fair play," he will probably portray Susan as showing con-
cern over having harmed someone else's prOperty. If the storywriter
showed no concern over what was done, it would be construed as a
pathological attitude,

It is not really surprising, therefore, that children in the
democratic cultures wrote stories containing themas of Susan saying
she was sorry or feeling sorry about getting ink Spots on the mother's
coat. It is part of the tradition of democracy of sensitivity to
the feelings of others and reapcnsibility for one's own acts, or

”sense of fair play."

“'7e

Categories 22 and 24 each need to be subdivided into two dif-
ferent categories because two different kinds of psychological values
are included in 22 and 24. One value is an independent, reaponsible
recognition of error on the part of the children combined with a motive
of not repeating it; Spontaneity is high and fear is low.. In the other,
which is also coded in 22 and 24, fear is high and Spontaneity is low,
and therefore the attitude does not represent re8ponsible behavior,
but conformity and avoidance of punishment.

Helsinki. The Helsinki sample, like the Birmingham sample, was
for the most part near the middle in rank with respect to the other
samples, in percentages of reaponse for categories of constructiyg
behavio; 31 13h; m, gopstmcjm behavig; by 313; ppm, c_:hi_l_d_ Bile.
.tggph, and child_jgllgmlig. These themas do not differentiate the
Helsinki sample, which again neither confirms nor denies the hypotheses

about a democratic culture. Further, the category cluster of

ocictanc

   

did not differentiate the Helsinki sample.

In the category cluster of anxiety the Helsinki sample was second
only to Birmingham and was significantly higher than Hamburg, Knoxville
and Munich in Story No. 4. This is consistent with the large number
of ambiguous outcomes found in the Helsinki sample, as pointed out
in Chapter IV. The traditionally low-communicative nature of human
relations in the Finnish culture is thought to be related to the
ambiguity.

In the submission category cluster, the Helsinki percentage was

in the middle of the samples, being significantly higher than Hamburg

48.

and Munich. In the dominatiog cluster Helsinki was significantly
lower than all the other samples at the .001 level. It was found in
Chapter IV that Helsinki was significantly lower than the other samples
in frequency of domination-submission outcomes. The analysis of
categories suggests that among the Helsinki children, though parental
domination is low, there is again the ”sense of fair play“ or respons-
ibility felt by the child for Spoiling the mother's new coat.
Responsibility should not be confused with submission. There were
many tallies in category 24 for the Helsinki stories, which resulted
in a fairly high percentage in the submission category cluster.
Hamburg and Munigh. In the analysis of categories for Story
No. 4, Hamburg and Munich were found to be similar to each other in
some respects and quite different in others. The percentages of
response in the ggngtzugtizg behavig; by;thg,ghild, resista e, 333?
missign, and dgminatign clusters were almost id mtical, which is
consistent with the findings for outcomes, where these two samples
were very much alike at all levels of relating. In the constructive
W )1 j_h_e m, m, 11.3, and anxiety clusters Hamburg and
Munich were significantly different from each other.
T;g¢h,and_lig,category clusters did not bear very much relation—
ship to the outcomes of the Stories No. 4 and No. 2. This is not to
say, however, that amounts of truth and lie were unrelated to the
patterns of human relationships within a culture. Where there is
more permissiveness of parents in a culture and more spontaneity,

there is likely to be more telling of t ruth by the children, since

49.

they need not fear retaliation.when they have made a mistake. But

the structures of Stories No. 2 and 4 are such that the truth—telling
or lying usually comes near the beginning of the story, and it is what
develops after that in the story that determines the outcome.

The fact that Munich and Hamburg were alike for the cluster of
993structivg bgh§1195_by_§hg,§hild in Story No. 4, ranking as the
lowest two of all the samples, while Hamburg was highest and Munich
was lowest in rank for the cluster of constructive behavior 31 Egg
pgthgg, with significant differences between them, suggest differences
in parent-child relationships in the two cultures. In Hamburg, it
would seem to the child-authors that the parents were prone to solve
the problems for their childrenwhen things went wrong more frequently
than in Munich where the children did not take initiative in making
amends for damage they may have caused. In Munich, like Hamburg, the
children did not seem to take initiative, but neither did the parents
initiate constructive behawor.

In the domination cluster the Munich and Hamburg percentages were
significantly higher than the Knoxville and Helsinki samples, and in '
the gubmissign cluster they were significantly lower than the Birmingham,
Knoxville and Helsinki percentages. The categories composing the
submisgign cluster are child categories. When a sample is found to
give themas of domination in large numbers of its stories, it means
that the children who wrote the stories have given these themas to
the mother character in the story. When few of the stories in a.sample

contain themas of submission, it is the children who wrote the stories

50.

who projected little such behavior onto the child character in the
story. In Munich and Hamburg, which are assumed to be authoritarian
cultures, the children reflected the greater activity of the authority,
the parent-character, by giving the mother more themas in their stories
and by giving the subordinate, the child—character, fewer themas.
Knoxvillg. The pattern of constructive behavior in Knoxville for
Story No. 4 is in contrast to that for Munich and Hamburg, and in fact

is directly opposed to the pattern in Hamburg. The Knoxville per-

centages rank highest in cogsgggctive behaviog Ex the.ghi;d,and

second to lowest in ggnstructive behavigg,py,ghg mgther, being signif-

icantly lower than.Hamburg in the latter category. Apparently in
Knoxville the children took relatively more initiative in constructive
problempsolving and the parents took less initiative, in this story.
This is consistent with hypotheses about individual reSponsibility and
spontaneity in a democratic atmoSphere.

At the dgmination-submission level, too, the pattern was apposed
to that in Hamburg and Munich. KnOXVille was significantly lower than
Hamburg and Munich in percentages for the domination cluster and signif-

icantly higher than Hamburg and Munich in the submission cluster. The

 

children in the Knoxville sample showed that the activity of children
assumes more importance to them, because more of them gave themas in
a child category cluster and fewer of them gave themas in the corre3pond-
ing parent category cluster.

As has been found consistently in analyses of Story No. 2 (2, 8),
so in Story No. 4, the Knoxville sample had the highest percentages of

truth and the lowest percentages of ii; clusters in its stories. This

finding adds to the evidence that in the Knoxville culture, as compared

5,1-

with the other cultures represented, there may be more telling of
truth and less telling of lies.

The amount of gnaig§z_shown by the percentage for the Knoxville
sample was significantly more than that shown by the Munich percentage
in Story Ho. 4. In Story No. 2, Knoxville was significantly lower
than all the other samples, including Munich. This was thought to
occur because Knoxville was the only sample (except Mexico City) which
is not still laboring under the economic dislocation caused by'World
War II. This interpretation seems to be denied by Story No. 4 results.
However, as was pointed out above, the categories in the anxiety cluster
included themas which should not be called anxiety. The anxiety percent-
ages in Story No. 4 are therefore not really comparable to those in
Story No. 2. The categories need re-defining and recasting in order
for the argiety cluster to be comparable for the two stories.

Length 9; Stories. Table 14 gives the average number of tallies
per story for boys as compared to girls in each sample, for high, middle,
and low socio-economic groups in all samples except Knoxville (which
could not be subdivided by socio-economic groups), and for boys as
compared to girls for Story No. 4 and Story No. 2 in Hamburg and Munich.
It is seen that the Birmingham children averaged more tallies per
story than any of the other samples. The stories from Birmingham were
long and contained greater variety in their plots than the other samples.
They more often had good closure than stories from other samples.

The significantly higher percentages of integration and domination—
resistaneg_outoores for the Birmingham sample may be related to the
length of their stories. In general, stories which were short more

often had ambiguous outcomes.

The processes of democracy require that through two—way communica-
tion problems be worked over until a solution is reached, and this
takes time. It would be expected, then, that in a culture high in
integrative problem—solving there would be more verbalization and

communication.

53.

VII. .CROSS-CULTUBAL SEX.COMPAEISONS

Table 3 in Chapter V and Tables 19 through 22 of.Appendix.A give
the data of which Table 15 is a composite. Table 15 shows whether girls
or boys are higher in frequency of reSponse in each category cluster
and outcome level for each of the five samples for Story No. 4. The
letters B and G are used to indicate boys or girls. If the same sex
was found to be higher in Story No. 2 as it was here, the letter B or
G is underlined.

Girls were higher than boys in 30 of the 40 category clusters.

In the Birmingham sample girls were higher in seven of the eight
clusters; in Hamburg they were higher in all the clusters.

Differences were in the same direction for Story to. 2 as for
Story Ho. 4 in 28 out of 45 category clusters plus integrative outcomes.

The category clusters showing a definite trend toward sex consistency
were:

Constructive behavior by mother, where girls were higher in all
samples and in three of the five samples sigr ficantly so;

angtgugtigg behavior gy;§§§;n, where girls were higher than boys
in four of the five samples and significantly so in the Birmingham
sample;

§3§§g,3gllg_lig, where girls were significantly higher than boys
in four of the five samples;

gesistangg, where girls were higher than boys in four of the

five samples and significantly so in the Birmingham sample:

TABLE 15

caosamnm sax conmmsoxs. RBCAPITUIATION or DIFFERENCES
sen-mm: BOYS mm GIRLS mom names 3, 19, 20, 21, 22

. "B'I designates a higher percentage of boys, ”G' designates
a higher percentage of girls. '*“ indicates that the difference
was significant at the .05 level or better, by chi-square test.
A line under "B“ or ”G” indicates that the same sex was also
found to be higher in Story No. 2.

 

Birming- Helsinki Hamburg Munich Knox-

 

Category luster ham . ville
Constructive (mother) (2" Q _G_ G“ G
Constructive (child) ,fif B G Q, Q!
Child tells truth B G _G_ G B
Child tells lie _G_ G G '_'..'_ B
Resistance G"I B_ G G G
Arxiety Q“ 9* Q G" 5‘;
Submission Q“ Q“ _G_ (_‘xj" Q“
Domination Q; _G_ G B Q

m:

Integrative G“ B _G_ G B
Domination-resistance B ‘B G G B
Ambiguous B‘ B* B B* B*
Domination-submission G G‘ B G G‘

 

54a.

Anxiet , where girls were higher than boys in all of the samples
and significantly so in three 0f the five; and

Sub issi , where girls were significantly higher than boys in
all of the five samples and significantly so in four of the five.

As explained previously under the discussion of sex differences
in Birmingham, the higher frequency of responses by girls to this story
may be due to the greater ease for girls of identifying with the character
in a story about a girl.

The trends in Story No. 2, The Lost Meat, which most strongly
supported the findings in the Ink Spots Story were in the category
clusters of submission and anxiet , where girls were higher than boys
and differences in some samples were Significant. For the rest of the
category clusters, boys responded with higher frequencies more often
than in Story No. 4, which could be predicted if it is easier for boys
to identify with the boy in Story Mo. 2 than it is for girls.

The boys were higher than girls in number of ambiggoug outcomes
to stories in all five samples and significantly so in four of them.
The girls were higher than boys in four of the five samples in number
of domipgtigppsubmissign outcomes and three of the differences were
significant. The boys' stories more often had less closure than the
girls' stories. The closure which the girls gave their stories was

more often in the domination-submission level.

 

Integrative and domination—resistance outcomes showed no consistent

trends of one sex having more stories in these outcome levels.

55.

 

VIII. CROSS-CULTURAL SOCIO—ECOHOMIC COMPARISONS

Table 16 is a recapitulation of Table # in Chapter 5 and Tables
23 through 25 in the Appendix and indicates the socio—economic group
which gave the highest frequency of reaponse to each thema cluster and
outcome level for Story No. 4. “H" means the high group had the
highest frequency, "M“ means the middle group gave the highest frequency,
and “L“ means the low group gave the highest frequency. Where both
high and middle groups were significantly higher than the low group,
but not significantly different from each other, it is indicated by
H-MT. If all three groups were significantly different from each
other, it is indicated by an asterisk beside both of the two highest
ones in order, as H’ L'. If the same group was found to be higher in
Story No. 2 as it was here, the letter is underlined.

The high or high-middle socio-economic group was higher in
frequency of response or in outcome level in 32 of the 48 compar-
isons. Sixteen of the differences were significant.

The consistent trends were in the following category clusters:
_Qonstructive behavior by_mother, where the high socio-economic group
was higher in all four samples, one of the differences being signif—
icant; assistance, where the high or high and middle socio-economic
group was higher in three out of four samples, and the middle socio-
economic group was higher in the fourth, two of the differences being
significant; and dominatiog, where the high and middle socio-economic

groups were higher in all four samples, one difference being significant.

56.

 

TABLE 16.

CRCSS— “Tm-AL soc: .scoromc cost-meszscrs. FEWITCLATION
or DIFFEBIXICES Bartram HI c: , I~iIDDLE mm Low
SOCIO-ECONOMIC LEVELS FROM TABLES u, 23, 24, 25

”H“ designates a higher percentage for the high socio-
economic group, "N“ for the middle socio—economic group, and "L"
for the low socio—economi: group. “H—M*" indicates that high and
middle socio-ecoromic groups were both significantly higher than
the low group. "3* L*" indicates that the high socio-econonie
group was higher than the low socio-economic group, which in turn
was higher than the middle socio—economic group. "*" indicatos
that the difference was significant at the .05 level or better.

A line under “H", “M“ or ”L" indicates that the same socio-economic
group was also found to be higher in Story No. 2.

 

 

Birming-

Category Cluster ham Helsinki Hamburg Munich
Constructive (mother) 2 g 11* H
Constructive (child) 3 M g :1
Child tells truth L L =rL‘ H

' Child tells lie :1 g L. IN
Resistance “* E, §}d* H
Anxiety g, L M H
Submission 3 ELM" M3. g
Domination E_ ' HaM* X ‘3

Outcome
Integra ive g _._._ L 3*
Domination-resistance H d L H
Ambiguous L L* H* ’* L*
Domination-submission H H-d* M* L‘ H

56a.

The high socio—economic group was higher in ipggggggizg outcomes
in three of the four samples- one difference was significant. The
low socio-economic group was higher in all four ggpiggggg outcomes, and
three of the four differences were significant. The socio-economic
groups did not show consistent trends in domination-resistancg_and
gpggpgtion_subnissign outcomes.

Twenty out of 36 of the category clusters and integrative out-
comes showed the same group to be highest in frequency in Story No. 2
as in Story No. 4.

It is interesting to note in Table 16 that in the truth and lie

 

category clusters, where one socio-economic group was highest in truth,
I the other group was highest in ;i_. The cluster of child tells lie

is not necessarily a reciprocal of child tells truth. It has been
found in the other studies (2, 8), however, that frequently these
clusters reciprccate each other. The consistency of the responses
made to the measuring instrument is supported by these findings.

Table 16 shows that in domination and gpppiggigg category clusters
the same socio-economic groups gave the highest percentages in all
samples. That is, where the high socio-eccnomic group was highest
in submission in Birmingham, Helsinki and Munich, it was also highest
in domiggtion. In Hamburg the middle socio-economic group was highest
in these two category clusters. It would'be eXpected that submission
would be associated.with domination when the gpgipgglgg is very strong:
and severe and the child's Spontaneity is relatively low.

Heber (8) found that the middle socio-economic groups in.Belsinki

and Hamburg did not give percentages of reSponse to Story No. 2 that

57.

were significantly different from the other socio-economic groups.
In Story No. 4 the middle socio-economic groups were in some cases
different from both the high and low socio-economic groups and did
not fall in between the high and low groups in percentages of response.

In Helsinki the middle socio-economic group was highest in dominetiqge

 

resistance outcomes. In the Hamburg sample the middle socio—economic
group was highest in category clusters of gpgigtz and domination.

In outcomes of ambiguity and dominationesuhmission the Hamburg middle
socio—economic group was significantly higher than either the high

or low socio~economic groups.

:0

IX. TYPES or Dommrlom

When the coding system for analysis was designed. provision was
made for finding what types of punishment were most frenuent in the
samplings being studied. Categories were provided for several
Specific types of domination. and tallies were given in the appro-
hriate categories whenever a child's story mentioned a particular
type of punishment or domination. The results are given in Table 17
as the “ercentages of chi‘dren in each sample giving reaponses in
each of the domination categories. The percentages in each category
are not necessarily mutually exclusive, because a child may have
mentioned two or more types of domination and would then be counted
in each type he mentioned. Category 64, Jother ggllg fathgr, is not
included in the cluster of themas of domigatign, but it is included
in Table l? as a matter of interest in the frequency with which
children in the different samples brought the father into the story.
From Table 17 several observations can be made:
Esprivatigg‘gfi‘iggdiwas mentioned as punishment more frequently in
Birmingham than in the other samples.

Dgprivatignigfi g;aytimg was more frequent in the Hamburg and Munich
:samples than in the others.

Hgstitutioglugdggfldugg§g_was more frequent in the Birmingham, Hamburg
and Knonille samples than in the Helsinki and Munich samples.

Ehygigglipggighmggtgwas more frequent in the Munich and Helsinki

samples than in the others.

59.

TIBLE l7

STOPY YIO. 4. PITCIIT 0.37 OF THE TOT L IUKBIRS 0F CHIIDII'I'I1 THE
FIVI’D EIELI ES GIVIl} TEE": 1‘3 03‘ DOI- I." .1.“ 1011 03' THE CHILD,
AID PHRCIL ;333 OF TOT IL LJIE HS 0‘ T'TLDIII II E10H
SAMPLING GIVIIG THIF1S II T NLVI n1nu0
D0111 111101I
In the percentages of total domination each child is counted only
once even though he may have given themes for two or more categories of
punishment or other domination.
In the percentages for each category a child is counted only once,
though the same child may be counted in more than one category.

17‘
:JS .1.

 

 

Percenta 3es

Type of Domination Birmingham Helsinki Hamburg Munich Knoxville

 

Total domination 69.9 37.5 64.6 65.9 54.2
Deprivation of food 2., 0 O .6 .9
Deprivation of pl% 7time 2.5 1.9 4.1 8.5 1.4
Restitution under duress 21.0 3.4 16.3 7.3 14.0
Physics 11 punishment 7.6 11.1 4.1 14.0 6.1
Mother tells father 4.6 1.4 2.4 1.8 1.4
Father punishes 3.8 .5 .8 .6 .5
Susan sent to bed 15.7 .5 5.7 .6 2.8
Miscellaneous punishment 8.6 7.2 4.9 3.7 7.5
Scolding 27.6 18.3 48.8 34.8 6.1
Susan sent to cleaners 6.8 2.4 8.1 3.0 1.9
AdmoniShment (past behavior)15.2 5.8 4.1 1.8 10.7
Admonishment (future 9.9 2.4 4.1 1.8 17.8

behavior)

 

60.

Mother ggllg father was most frequent in Birmingham and second most
frequent in the Hamburg sample.

Egnishmentlpyufgghgg_was more frequent in the Birmingham sample than
in the others.

‘§p£§n‘§ggt.tg pg; was most frequent in Birmingham and second most
frequent in Hamburg.

Misgellaggqgg‘punighmgnt‘was more frequent in Birmingham, Knoxville
and Helsinki than in Hamburg and Munich. Statclents like "The mother
punished Susan” were coded here.

§gglging showed a wide range, with large differences between the frequen-
cies of all samples. Che Hamburg frequency was highest, then Munich,
Birmingham. Helsinki, and Knoxville. in that order.

Sgsan.gent to §§g£§_(or cleaners) was most freouent in Hamburg and
next highest in Birmingham.

Admonishment ganggrning‘pggt behavior was most frequent in Birmingham
and next most frequent in Knoxville.

Admonishment gpqut.§g§ggg behavior was more frequent in the Knoxville
sample than in any of the others, and in the Birmingham sample was
second.

Table 17 shows the amount of each type of domination in a sample
relative to the other samples, but it is difficult to tell from this
table which types of domination were most prevalent within a given
sample with respect to those most prevalent in the other samples.

The types of domination were therefore ranked according to their

frequency in each sample. The rankings are shown in Table 18. In

Table 18 it can be seen that:

Deprivation,g§.£ggg was not a relatively prevalent form of punishment
in any of the samples.

Deprivation 91 playtimg ranked high only in Munich.

Restitution.gn§g; duress was a relatively prevalent form of domination
in all the samples, eSpecially in Birmingham, Hamburg, and Knoxville.

Physical pgpishment ranked high in Helsinki and Munich. The ranking
of the father as an agent was quite low in all the samples.

§p§gptggp§_tp,bgg was a prevalent form of punishment in Birmingham
and Hamburg.

Miscellaneous punishment ranked fairly high in Helsinki and Knoxville.

chlding was the most prevalent form of domination in Birmingham,
Helsinki, Hamburg, and Munich. It was not very prevalent in
Knoxville.

§p§§pL§§ntltg.thg‘§tggg was fairly prevalent in Hamburg.

Admonishment,a§gp&_p§§t penivigz ranked fairly high in Knoxville,
Birmingham and Helsinki.

Admopighment‘gpgg§,£u§ugg behavior ranked highest as a form of domina-
tion in Knoxville, and it was not very high in the other samples.

It should be borne in mind that the figures in Table 18 are
independent of the total domination in each sample; that is, the rank
of one for scolding in Helsinki and Birmingham does not mean that
there is as much scolding mentioned in the Helsinki sample as there
is in the Birmingham sample. It only means that 2:.ihé.&2£§l kinds
of domination in the Helsinki and Birmingham samples, scolding was
the most prevalent in both. ‘As a reminder, the total percentage

of domination which was found in each.sample is given at the top

TnBLE 18

3113K ORDER OF TYPES OF DUI-11173101? WITHIN ELCH OF THE
F115.) 5.3139138 Till-£11 330M TILBIE. 17

 

..-.—

 

 

.5

Rank Orders

 

Type of Domination Birmingham Helsinki Hamburg Munich Knoxville
Total domination (per cent) 69.9 37.5 6L.6 65.9 5n.2
Deprivation of food 11.5 12 12 11 11
Deprivation of playtime 11.5 8 7.5 3 10
Restitution under duress 2 5 2 4 2
Physical punishment 7 2 7.5 2 5.5
Mother tells father 9 9 10 8 10
Father punishes 10 10.5 11 11 12
Susan sent to bed 3 10.5 4 11 7
Miscellaneous punishment 6 3‘ 5 5 U
Scolding l l 1 1 5.5
Susan sent to cleaners 8 6.5 3 6 8
Admonishment (past behavior) 4 4 7.5 8 3

Admonishment (future
behavior) 5 6.5 7.5 8 1

 

of each column in Tables 17 and 18.

Discussion of Types of Domination
In Chapter IV, in the discussion on outcomes of stories, it
was found that Hamburg, Munich, and Knoxville were very much alike
in outcomes. The question was raised as to whether or not the
domination-submission in a more democratic culture is the same as
the domination-submission in a more authoritarian culture. Some
insight into the answer to this question may be gained by comparing

figures for Hamburg, Munich and Knoxville in Table 18.

In the Knoxville sample, admonishment about 1353;; behavior
ranked first and admonishmept apggtlpggt pghgvigz ranked third.

In all the other samples, scolding ranked first. Furthermore,
admgnishment did not rank high in any of the other samples, and
scgldigg did not rank high in Knoxville. Admonishing is a milder
form of domination than scolding, and this is one clear difference
between domination in the Knoxville sample and in the other samples.
From the evidence of several of the findings it is proposed
that the domination of parents in Knoxville is of a milder form than
the domination of parents in Hamburg and Munich. .Admonishing usually
has reference to the act of the child: scolding has reference to the
child. Physical punishment may have reference to the behavior but by
definition is addressed to the child as a person. The data suggest
further that the submission of the children in Knoxville is different
from the submission of children in Hamburg and Munich. It is marked

by high acceptance of the ethical code including telling the truth,

64.

accepting the blame for one's own mistakes, and taking initiative in
making amends for those mistakes.

The Specific findings which supported this interpretation include
(1) the high ranking of admonishmgnt and of restitution m m
as two of the most prevalent kinds of domination in the Knoxville
sample; (2) the significantly high percentage of response for the
Knoxville sample in the j§;;§_tzgth cluster of themas, and the low
percentage of reSponse in the tell§_lig cluster of themas; (3) the
fact that the Knoxville sample showed the highest percentage of reSponse
in the gonstzuctive behavior Ly §g§an category cluster but next to
the lowest percentage of children with themas in the construgtivg
behaviozlhx mother category cluster. Munich and Hamburg were lowest

in frequency of constructivg behavigruhx Sgsag themas, and Hamburg

was highest in frequency of constructive pehavior.hzumgthg; themas.

65.

X. SUEEARY

This study was an analysis of Story No. 4, The Ink Spots, Series A,
of the Anderson Incomplete Stories. Story No. 2, The Lost Meat, had
been analyzed in a previous study, and it was possible to make com-
parisons between the two analyses and further evaluate the Anderson
Incomplete Stories as a measuring instrument of intercultural
differences.

Samples used were from data collected in Birmingham, England;
in Helsinki, Finland; in Hamburg and Munich, Germany; and in Knoxville,
Tennessee, U.S.A. Subjects were 1,227 children in the seventh school
year in these cities.

Children's stories were coded in two ways: ‘by four defined levels
of outcome (integrative, domination-resistance, ambiguous, or domination-
submission); and by categories representing the psychological themas
in the stories. In Story No. 2 there were 33 categories for thought,
feeling and action by the child-character and 46 for the mother-
character. In Story No. 4 there were 37 categories for thought,
feeling, and action by the child-character and 53 categories for the
mother-character.

Aims were to analyze Story No. 4 and make sex and socio-economic
comparisons in category clusters and outcome levels, and these same
comparisons in cross-cultural samples, and to make comparisons of
Story No. h with Story'Ho. 2.

Reliability of coding showed percentages of agreement ranging,

66.

from 93.57% to 1007?.
The Birmingham sample was highest in the two outcomes of spontan-

eity of the child, intczrative outcome and dom nation—resistanc .

   

It was lowest in ambiguous outcomes. That is, human relations in
Birmingham were more clearly defined by these children in their stories.
Helsinki was significantly different from all other samples in ambiggous,
unstructured outcomes of the stories, indicating a low order of communi-
cation, and significantly lower in ggmination-submissign. Hamburg,
Munich and Knoxville were alike in all four outcomes and showed no
significant differences among themselves.

More boys showed ambiguous, unstructured outcomes than girls

 

in all five samplings, the differences being significant in four.
More children in the high socio-economic groups showed integrative
outcomes in both Story No. 2 and Story No. 4, the differences for
Story No. 4 being significant in Hamburg and Munich. More children

in the low socio-economic groups had ambiguous outcomes than in the

 

high groups in all samples, the differences being significant in Helsinki,
Hamburg and Munich.

Analyses of the data showed that in completion of the stories
children were sensitive to the structure of the stimulus material.
There were two structural differences in the two stories that showed
effects in the analysis of the data. Story Ho. 4 had a question after
it: ”How do they both feel about it?" Calling attention to both the

mother and the child seemed to account for higher numbers of integrative

S

outcomes, higher percentages of constructive_behavigr by both the

mother and the child and higher frequencies of anxiety than were found
in Story No. 2, while themas of domination and submission were lower .
*in frequency. Second, in Story No. 4 the mother enters Just as Susan
finds herself in a predicament. There was less opportunity or necessity
to tell truth or lie, and there were much lower frequencies of both
teng truth_and.1gll§.lig in Story Ho. 4.

The girls wrote longer stories with wider variety of themas than
boys. As Heber found in Story Ho. 2, girls gave higher frequencies
in category clusters of constructive behavior Ly mothe , submission, and

naggiggz. Consistently in all five samples girls were high in constructive

mother category clusters; with the exception of Helsinki, gonstgugtivg

 

bghayig; by 33g.ghild; and, with the exception of Knoxville, child tgllg
lie. For all five samples girls had higher frequencies of anxiety and
submission, the differences being generally in the same direction as
for Story No. 2, and six of tle ten differences for these two categories
being significant.

Analysis of data for socio-economic levels showed that children
from the high groups generally wrote longer stories. They had consistent-
ly higher frequencies of integrative outcomes (as in Story No. 2), the
differences in Hamburg and Munich being significant. The high groups
had consistently higher frequencies in the category cluster of gonstruct—
.11; behavior by tgg_mother and, with the exception of Helsinki, construct-
izg behavig; by_th§ child. Children in the high group had higher fre-
quencies of.tg;;§.lig i111 Birmingham and Helsinki. Children in the

low group had higher frequencies of tells lie in Hamburg and Munich,

68-

the difference in Munich being significant. Socio—economic levels in
the four cities were exactly reversed for child tells truth. Children
in the high socio-economic group had high percentages of 3g;;§,tguth
in Hamburg and Munich, while tgllg taggh was found more frequently in
Birmingham and Helsinki among children at the low socio-economic level.
Children in the low groups tend to write shorter stories.

'Knoxville stories reveal more initiative and personal integrity
in the child and more resPect for the child for both Story No. 2 and

Story No. 4. Knoxville children showed the highest percentage of tells

 

tgpgh and the lowest percentage of tells lie. They showed high attention
to the child in having highest percentages of constructive behgvigr Ly
Sgggn, and lower attention to the mother in having next to lowest
percentages of children with themas in the gonstructivg fighgxigz;mz
mother category cluster. Munich and Hamburg were lowest in frequency

for constructive behavior hy_§p§gg, and Hamburg children, who gave little

attention to constructive behavior §y_tre child, were highest in fre—

 

quency of constructive behavior L; the mother. Evidence shows that

 

according to the children's stories highest frequencies of truth and
lower frequencies of tells lie were accompanied by milder punishment
and milder forms of parental domination.in Knoxville.x Admonishment

about future behavior ranked first in frequency and admonishment about

 

past behavior ranked third. In all the other samples scolding ranked

 

first among the types of domination. Adnonishment did not rank high

in any of the other samples and scolding did not rank high in Knoxville.
It was pointed out that admonishment represents chiefly disapproval of
some kind of behavior. Scolding and physical punishment represent an

attack on the child himself.

69.

Hamburg and Munich were significantly different from each other in
category clusters of constructive behavior by the mother, truth, i ,
and anxiet*, supporting the findings for Story Ho. 2, which leaves Open

the question of a Cerman national character.

 

Anderson, H. H. Domination and Integration in the Social Behavior
of Young Children in an Experimental Play Situation,
G net. Psychol. Mopqgg., 19: 341-409, 1937

Anderson, H. H. Children's Judgment of Social Conflict: An

Intercultural and cross-national Study. (A
we er presented to the Mich. Acad. Sci., Arts,
and Letters. 19553

Anderson, H. H. and Anderson. G. L. Children's Perceptions of
Social Conflict Situation: A Study of Adolescent
Children in Germa.‘. Apggjpfp.gggrn;l pf Qrtho—
psychiatry, Vol. XXIV, No. 2, 195D, 246-257

Anderson, H. H., and Brewer, H. M. Studies of Teachers' Classroom
Personalities, I: Dominative and Socially
Integrative behavior of Kindergarten Children.

Anpl. Psvgthgl. Mop. 0:31., No. 6, 1945

Anderson, H. H., and Brewer, J. E., and Reed, M. F. Studies of
Teachers' Classroom Personalities, III: Follow-up
Studies of the Effects of Dominative and Integra—
‘ive Contacts on Children's Behavior. Appl.
Psychgl. Monogr., Stanford University Press, 1946

Geierhaas, F. G. Problems of Reliability in Evaluating Story
Completions about Social Conflicts by German
Adolescent Children. Unpublished M. A. thesis,
Michig’n State College, 1955

Geierhaas, F. G. Problems of reliability in Evaluating Story
Completions about Social Conflicts by German
Adolescent Children. (A paper presented to the
Mich. Acad. Sci., Arts, and Letters, 1955

Heber, H. F. A Cross—cultural Comparison of Children's judgment
of Parent-Child Conflict in Germany, England,
Finland, United States and Mexico. Unpublished
M. A. thesis, Michigan State College, 1955

HELSINKI, FINLAND:

" ‘ I
APPLH

DIX A

TABLE 19

ITUlZBLHES

"TD PEPE} . HT G..38

OF

BCYS.RHD GIRLS 31TH THJMAS CODED IH THE E GHT
CA? CORY CLUSTBRS, nzn ACCORDIHG TO OUTCOME

 

 

Story Ho. 4

Story No. 2

 

 

Boys Girls Boys, Girls

Category Cluster N % N fl N 3 H fl
Constructive (mother) 48 46.6 54. 51.4 17 16.5 18 17.0
Constructive (child) 39 37.9 32 30.5 4 3.9 5 4.7
Child tells truth 42 40.8 47 44.8 49 47.6 55 51.9
Child tells lie 15- 14.6 20 19.0 32 31.1 36 34.0
Resistance 17 16.5 17 16.2 11 10.7 7 6.6
Anxiety 71 68.9 89 84.8* 51 49.5 59 55.7
Submission 25 24.3 43 40.9* 48 46.6 67 63.2*
Domination 37 35.9 41 39.0 58 56.3 63 59.4

9232222.

Integrative 18 17.5 14 13.3 7 6.8 7 6.6
Domination-Resistance 10 9.7 2 1.9 (Not reported by Heber)
Ambiguous 55 53.4* 38 36.2 " " " "
Domination-Submission 20 19.4 51 48.64l " " " "

Total 103 105 103 106

 

 

*Sifinificantly higher than the other sex at the .05 level or better, by

chi-square test.

\J

f\)

th3033, 9374;; :

x'v firm
1“ r3111.)

TABLE 20

AED PERCENT.GES OF BOYS AND GIRLS

WITH THINKS CODED IN TIE EIG’T CaTEGORY CLUSTERS,

AND ACCORDIKG T0 OUTCOME

Story no. 4

Story No. 2

 

 

Boys Girls Boys Girls
Category Cluster h % H % N s N %
Constructive (mother) 66 53.7 73 59.3 20 14.1 34 24.55
Constructive (child) 34 27.6 44 35.8 14 9.9 13 9.4
Child tells truth 62 50.4 65 52.8 65 45.8 83 59.7*
Child tells lie 32 26.0 36 29.3 72 50.7* 51 36.7
Resistance 11 8.9 17 13.8 16 11.3 13 9.4
AflXiety 76 61.8 88 71.5 72 50.7 80 5706
Submission 21 17.1 32 26.0 70 49.3 78 56.1
Domination 76 61.8 83 67.5 88 62.0 83 59.7
Outcome
Integrative 13 10.6 17 13.8 14 9.9 14 10.1
Domination-resistance 6 4.9 7 5.7 (Eot reported by Heber)
Ambiguous 44 35.8 40 32.5 " " " "
Domination—Submission 60 48.8 58 47.2 " " " "
Total 123 123 142 139

 

*Significantly higher than the other sex at the .05 level or better, by

chi-square test.

73-

TABLE 21

MUNICH, GERMANY: NUNBERS AL PER 33 AGES 0? BOYS AND GIRLS
WITH THEMAS CODED IN THE EIGHT CATEGORY CLUSTERS,
AND ACCORDING TO OUTCOME

 

 

 

 

Story No. 4 Story Ho. 2
Boys . Girls Boys Girls
Category Cluster N 923 N 55 1‘1 36 N ~35
Constructive (mother) 24 32.9 48 52.7‘ 5 6,8 5 5,4
Constructive (child) 20 27.4 29 31.9 5 6.8 8 8.7
Child tells truth 25 34.2 39 42.9 26 35.6 32 34.8
Child tells lie 27 37.0 34 37.4 45 61.6 61 66.3
Resistance 6 8.2 12 13.2 8 11.0 4 4.3
Anxiety 29 39.7 52 57.1"l 28 38.4 35 38.0
Submission 7 9.6 25 27.5“ 26 35.6 48 52.2*
Domination 49 67.1 59 64.8 42 57.5 51 55.4
Outcome
Integrative 8 11.0 13 14.3 3 4.1 2 2.2
Domination-Resistance 1 1.4 11 12.1 (Not reported by Heber)
Ambiguous 34 46.6"I 21 23.1 " " ' "
Domination-Submission 30 41.1 46 50.6 " u " "
Total 73 91 73 92

 

*Significnntly higher than the other sex at the .05 level or better, by
chi—square test.

74.

TABLE 22

KKOXVILIE, TEDNESSEE: I'IRL

ETD PSRC "TrG OF BOY}

"133D G II—lLS

1' I23 TESL-.8 CODED IN '1'.) EIGHT C.TI~3GORY CLUST'1 .3113. 113118,

211.1) 210 C 0131:) 1 11 rG- T 0 OUT C 01-23

 

 

Story No. 4

Story Ho. 2

 

 

Boys Girls Boys' Girls

Category Cluster N ,3 I: ,1; R 7‘6 N ,5
Constructive (mother) 2 45.2 57 47.1 2 12.2 9 7.6
Constructive (child) 27 29.0 52 43.0* 4 4.1 10 8.4
Child tells truth 58 62.4 72 59.5 60 01.2 80 67.2
Child tells lie 16 17.2 141.6 23 23.5 19 16.0
Resistance 6 6.5 9 7. 4 14 14.3 6 5.0
Anxiety 58 62.4 82 67. 8 12 12.2 16 13.4
Submission 24 25.8 48 39.’7 ., 41 41.8 83 69.7*
Domination 46 49.5 70 57. 9* 56 57.1 85 71.4*

Outcome
Integrative 15 16.1 14 11.6 4 4.1 5 4.2
Domination—Resistance 5 5.4 2 1.7 (Not reported by Heber)
Ambiguous 2 34.4* 27 22.3 " " " "
Domination-Submission 41 44.1 78 64.5* " " " "

Total 93 121 98 119

 

*Significantly higher than the other sex at the .05 level or better, by

chi—square test.

75.

TABLE 23

HELSIIII, FIULAHD: NUMBERS AND PBHGHITAGES OF CHILDREN IN E 8
HIGH, MIDDLB.AND LOJ SOCIO—ECOHOMIC GROUPS U TH THEMnS CODDD
IN THJ EIGHT CATEGORY CLUSTERS, ADD ACCORDIHG TO OUTCOI”S
FOR STORIES HO. 4 AND NO. 2

 

 

 

 

Story To. 4 Story 10. 2
Category High hiddle Low High kiddie Low
Cluster n 3 N s N t n 9 N 5 n 9
Constructive 39 52.7 37 46.3 26 48.1 22 29.7* 8 10.0 5 9.1
(mother) .
Constructive 21 28.4 30 37.5 20 37.0 3 4.1 5 6.3 1 1.8
(child)
Child tells
truth 29 39.2 34 42.5 26 48.] 40 54.1 37 46.3 27 49.1
Child tells
lie 17 23.0 10 12.5 8 14.8 32 43.2* 27 33.8* 9 16.4
Resistance 13 17.8 14 17.5 '7 13.0 10 13.5 5 6.3 3 5.5
Anxiety 58 78.4 59 73.8 43 79.6 49 66.2* 35 43.8 26 47.3
Submission 26 35.1 31 38.8 11 20.4 36 48. 47 58.8 32 58.2
Domination 33 44.6* 35 43.8* 10 18.5 38 51.4 49 61.3 34 61.8
Outcome
Integrative 15 20.3 12 15.0 5 9.3 8 10.8 3 3.8 3 5.5
Domination- 4 5.4 5 6.3 3 5.6 (Not reported by Heber)
resistnnce
Ambiguous 27 36.5 34 42.5 32 59.3* " " " "
Domination- 28 37.8 29 36.°* 14 25.9 " " " "
submission
Total 74 80 54 74 80 55

 

*Significnntly higher than

.01 level.

the other

76.

socio-economic group or groups at we

immune, 03m;
MIDDLE ADD

TICUT

113..»

L05 SOCIO-
CiTZGORY CIU

-1"
.lJU

TABLE 24

.ITY: NIH-332338 -‘ITD PTT~€CITT£GLS OF CHILDREN III THE HIGH,
ONCE-.10 GROUPS '9’ TH TEX-lifts

CODHD III THE

STDRS, ADD ACCORDIHG TO OUTCOM18
FOR STORIQS NO. 4 AND NO. 2

 

 

Story No. 4

Story Ho. 2

 

 

Category High Middle Low High. hiddle Low
Cluster n s N 8 h 8 n s n 9 N 3
ConstruCtive 51 73.9* 55 50.0 33 49.3 16 23.5* 33 22.6* 5 7.5

(mother)
Constructive 30 43.5 27 24.5 21 31.3 8 11.8 16 11.0 3 4.5
(child)
Child tells 44 63.8* 45 40.9 38 56.7* 40 58.8 73 50.0 35 52.2
truth
Child tells 18 26.1 30 27.3 20 29.9 29 42.6 63 43.2 31 46.3
lie
Resistance 10 14.5* 15 13.6* ‘3 4.5 8 11.8 15 10.3 6 9.0
Anxiety 45 65.2 83 75.5 36 53.7 40 55.8* 90 61.6 22 32.8*
Submission 12 17.4 34 30.9* 7 10.4 38 55.9 79 54.1 31 46.3
Domination 37 53.6 80 72.7 42 62.7 45 66.2 83 56.8 43 64.2
Outcome
Integrative 9 13.04 12 10.9 9 13.4 8 11.8 18 12.3 2 3.0
Domination-~ '3 4.4 4 3.6 6 9.0 (Hot reported by Heber)
resistance
Ambiguous 38 55.1* 24 21.8 22 32.8* " " " "
Domination— 19 27,5 70 63.64 30 44.8* n n n u
submission
Total 69 110 67 68 146 67

 

*Significantly higher than the other socio-economic group or groups at the
.05 level or better, by chi-square test or t—test.

77.

TABLE 25

MUNICH, Gnhhnnr:

I. TLBLLRS rdTD PERCEITTAG

.38 OF CHILDREI.T IL? T1133

 

 

 

 

 

EIG: HUD LO'. SOCIO-ECOKOI-LIC GROUPS 1'." TH TEX-.115 CODED 1LT T1113
EIGHT .LTFBORY CLUSTERS , ALD JiCCORDIl-TG TO OUTCOI‘LJJS
FOR STORIES ITO. 4 1112') ITO. 2
Story Ho. 4 Story Ho. 2
High Low High low
Category Cluster N % N 9 N % N %
Constructive (mother) 37 4410 35 43.8 5 5.8 5 6.3
Constructive (child) 29 34.5 20 25.0 7 8.1 6 7.6
Child tells truth 37 44.0 27 33.8 . 26 30.2 32 40.5
Child tells lie 24 28.6 37 46.3* 59 68.6 47 59.5
Resistance 10 11.9 8 10.0 4 4.7 8 10.1
Anxiety 42 50.0 39 48.8 31 36.0 39 40.5
Submission 17 20.2 15 18.8 39 45.3 35 44.3
Domination 57 67.9 51 63.8 49 57.0 44 55-7
misses
Integrative 14 16.7 7 8.8 3 3.5 2 2.5
Domination-resistance 7 8.3 5 6.3 (Not reported by Heber)
Ambiguous 21 25.0 34 42 .5‘ " " " "
Domination-submission 42 50.0 '34 42.5 " ” " "
Total 84 80 86 79

 

*Significantly higher than the other socio-economic group at the .02

level, by chi-square test.

78-

“P3111712: B
Anderson Incomplete Stories: Form 2. Age at last

 

 

 

 

 

Birthdats: birthday! 30y: Girl
Birthplace: Church attendst

HolongY onghavo you fved in
School: ' Grado:_____ th 3 community. ciiy. town?_________

 

Occupation of father: - _1 Deception of mother:

 

1. Frank and Peter were walking to school. Suddenly Frank grabs Peter's cap and
throws it high up into the nearest tree so that Peter could not reach it with his
hand. Frank had never done anything like this before. Neither had Frank and Peter
quarrolod with each other the day before. -

Why did Frank do this? What does Peter do?
What does Frank think? How does Peter feel about it?

Think about those questions. then finish the story quickly with a few sentences.

2. Michael's mother sends him to the store to get one pound of wieners. On tho wry
home he puts the package of meat on the edge of the sidewalk and plays for a. short
time with his friends. A shepherd dog darts forward quickly. claws half the wither:-
out of the package and rushes away with them. Michael wraps up the remaining wieue:
and takes them home.

What does Michael say to his mother?

What does his mother do?

How does Michael then feel about it?

Think about those questions. then finish this story quickly with a few sentences.

79.

~2—

3. The teacher suddenly discovers that fifty cents has disappeared from her desk.
She looks up and sees that all the class are working on their arithmetic assignment.
She wonders what happened to the money and what she should do.

What does the teacher do?

Finish this story also with a few sentences. Tell what happened to the money and also
exactly how the teacher feels and what she does.

 

1}. Elizabeth is sitting in the living room working on her homework. Suddenly she
remembers her mother's new coat. She wonders how she would look in it. When taking
the coat off she notices that she had got some ink spots on her mother's new coat.
Just as Elizabeth is trying to wipe out the ink spots her mother enters the room.

What does her mother say? How do they both feel about it?
What does Elisabeth say? What does each one do?

Think about these questions and finish this story quickly with a few sentences.

~3—

5. George and Karl are playing with a football. They know that they should not play
football on the narrow space in front of the house. George gives a strong kick and
the ball hits a window and makes a big crack in it. Karl thought that someone had

appeared behind the window. No one could have seen who lad kicked the ball against
thO “Nous

Finish this story with a few sentences. Tell how the two boys feel about it and what
they do.

6. Else. often handed in her home composition late to the teacher. This time it was
an especially important composition and she had. moreover. written it on time. On
the way to school she lost her composition notebook and could not find it anywhere.

What does Elsa say to her teacher?
What does the teacher say?
Think about these questions and may this story with a few sentences.

495
7-31—53
an 81.

. k

71‘11 .._ .-
‘-th§~‘ ' -‘ ' ““I...,I...'._".?-
'u

W0
W35

A; "V 18 WW};
Wa-

 

 

MICHIGAN STATE UNIVERSITY LIBRARIE

LILIL LILILLIL IIILLLIII