A MULTIPLE AGREEMENT ANALYSIS AND
ENTERPRETAHON 0F COLLEGE

$ELECTEON PATTERNS

Thesis {‘or flu Degree of DH. D.
MECEEGAR STiTE UKWERSETY
Roderick Arthur Hilsinger
1963

 

 

THESIS LIBRARY l
Michigan State '

University

 

This is to certify that the
thesis entitled

A MULTIPLE AGREEMENT ANALYSIS AND INTERPRETATION
OF COLLEGE SELECTION PATTERNS

presented by

Roderick Arthur Hilsinger

has been accepted towards fulfillment
of the requirements for

.m—degm in Education

 

Date JUIY 31. 1963

0-169

ABSTRACT

AJMULTIPLE AGREEMENT ANALYSIS AND INTERPRETATION
OF COLLEGE SELECTION PATTERNS

by Roderick Arthur Hilsinger

The purpose of the study was to isolate groups of students with
similar reasons for selecting their colleges, and to identify, if
possible, the distinguishing;characteristics of these groups. The
questions this research attempted to answer are: (1) Do patterns
exist in the reasons students give for selecting colleges? (2) Are
students with these patterns distinguishably different in relation
to other selected variables? (i.e., socio-economic status, level
of intelligence, education of parents, peer group acceptance, self-
-concept, type of high school attended, and type of college chosen.)

The sample employed in this research was a small portion of the
high school papulation surveyed in 1957 and 1958 by James Coleman,
and reported in his book, The Adolescent Society. Five hundred and
eighty-one of the approximately 10,000 original students from ten
northern Illinois high schools completed a questionnaire in the spring
of their college freshman-year, as part of a.follow-up study by Coleman
and Peter Rossi, of the National Opinion Research Center in Chicago.
{NORC Survey 412.) From among these 581, 276 were-extracted who had
completed fall and spring questionnaires their senior year in high
.school, indicating their choice of a specific college and their

‘reasons for choice, and who were enrolled at that particular college

..
a

Roderick Arthur Hilsinger

when completing the third questionnaire the following spring. For the
purpose of validation and cross-validation, this sample was randomly
divided into two groups of 138 students each.

Hemingway's Multiple Agreement Analysis was employedjto separate
each of the two sub-samples into groups exhibiting a commonality of
reasdns. When positive responses were analysed together with the lack
of response to eight "reasons for choice," 72 percent of the students
in the validation sample, and 73 percent in the cross validation, were
accounted for in four categories of reason response types: Types A,
B, C, and D. The remaining students, whose unique combinations of
reasons did not coincide with any of the established patterns, were
designated as "Type" N.

Chi square analysis was made, setting alpha at .20, of the
respOnses of students in the validation sample to 153 other available
items. Forty-eight significantly differentiated the students in at
least one reason-response category from the remaining students. When
these 48 items were subjected to another chi square analysis with the
across~validation group, setting alpha at .10, only 16 of the items
significantly differentiated. Thus sixteen characteristics were~dise~
covered which served to "identify" or describe the four major reason-
response types isolated, plus N.

The major hypothesis that students could be classified according
to their reasons for college selection was supported. The minor
hypotheses-—that the content of the questionnaire items would identify

the characteristics of students in these classifications--were

 

Roderick Arthur Hilsinger

partially suppOrted. Because of the limited number of items found to
hold after cross-validation, it was not possible to crystallize the
identifying characteristics of the students in each type of reason-

response group to the extent originally anticipated."

 

A MULTIPLE AGREEMENT ANALYSIS AND INTERPRETATION

OF COLLEGE SELECTION PATTERNS

By.

Roderick Arthur 'Hilsinger

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the-degree .of
DOCTOR'OF PHILOSOPHY

College of Education

19 63

ACKNOWLEDGMENTS

This research was part of a project sponsored by the College
Entrance Examination.Board. Grateful appreciation is extended to
iMru Douglas Dillenbeck of that organization, whose initial encourage-
=ment and assistance helped crystallize the direction of this research.
In addition, thanks are due to Dr. Peter Rossi, Director of the I
National Opinion Research Center, for kindly granting access to the
data from NORC Survey 412.

Among the many members of the Michigan State University faculty
who gave generously of their time and assistance, Drs. David Krathwohl
and David Gottlieb deserve particulaerention. Special thanks are

given to Dr. Buford Stefflre for his patient and unstinting help in

athe design and analysis stages of this project, and to Dru‘William

Farquhar for advice on-the statistical treatment of the data and aid
in the preparation of the final capy.

Finally, the report itself has been immeasurable-enriched by
the constructive criticism, editing, and typing skills of my wife,

Patricia Dean.Hilsinger.

ii

ACKNOWLEDGMENTS . . . . . . .

LIST OF TABLES I I I C C O 3

LIST OF APPENDICES. . . . . .

Chapt

II

III

er
I THE PROBLEM . . . . .

Purpose of the Study..

Need for the Study .

TABLE OF

0

Theoretical Assumptions
Description of the Study. . . .
Definition of Terms . . . . . .
Limitations of the Study.
Hypotheses to be Tested .

CONTENTS

0

Underlying the Study.

REVIEW OF THE LITERATURE.

Nature of the Review. . .

Explanation of the Tables

O

Criticism of the Literature

Trends revealed by the Literatu

re

Summary . . . . . . . . . . . ...

DESIGN AND METHODOLOGY. . . . .

Description of the Sample . . .
Coleman's Original Sample. .
The Follow-up Sample
The Present Sample .

Design of the
Methodology .

Study .

Stages One and Three

Stage TWO.
Stage Four . .

by NORC

Reliability Estimates.
Summary . . . . . . . . .

ANALYSIS AND INTERPRETATION

Stage One:

Multiple Agreement Analysis

validation Sample . . . . .

Stage Two:

Identifying Characteristics

of the

O I I O I O O C O O O

of the

validating Groups . . . . . . . . . .

‘Significant Differences Among Reason-Response

Types Within the validation Sample. . . . .

iii

0 O . 0 O O I C I I I

I O O O O I O I O O I

 

Page

11

59
65
66

Chapter

Iv ANALYSIS AND INTERPRETATION,

VCont.

Item Content Associated With Type A Students .
Item Content Associated With Type B Students .
Item.Content Associated With Type 0 Students .
Item Content Associated With Type D Students .
Item Content Associated With "Type" N Students

Item Content Which Did Not Discriminate Between

Reason-Response Types
Stage Three:

Stage Four:

Chi-Square Analysis of the

Reliability Estimates . .
Summary

V SUMMARY AND CONCLUSIONS .

Purpose
The Sample . . . . . . . .
Design of the Study . . .

Reason-Response Types Identified

Conclusions

A Comparison with Previous Literature

iMultiple Agreement Analysis of the

Cross-Validation Sample
Identifying Characteristics of the
Cross-Validation Sample

Cross-Validation Sample

Limitations of the Conclusions . . . .
Implications for Further Research

General Recommendations

BIBLIOGRAPHY . . . . . . . . .

iv

Recommendations for Future Questionnaires

O O O O O O O O O O

Page

66
66
68
69
69

71
76

78
79
88
89

91

91
91
92
93
95
96
101
102
102
106

108

Table No.

2.1

2.2

4.7

4.8

LIST OF TABLES

A Topical Summary of Research on Factors Influencing
Students in the Selection of their Colleges . . . . . .

The Rank Order of Factors that Influenced Students'
Choice of a College to Attend as Reported in Studies
ft” 1926 to 1962 I I I I I I I I I I I I I I I I I I I

The Frequency with Which Influential Factors in
College Selection were Ranked... . . . . . . . . . . .

The Refinement of the NORC Follow-up Sample. . . . . .

The Frequency Distribution by High Schools of the

Samples used in this Research and Coleman's

Original Sample and Population . . . . . . . . . . . .
Responses to Item 53 of the Fall Questionnaire . . . .
Direction of Responses to Item 53 of the Fall Questionnaire

Chi-Square Table for Item 53 of the Fall Questionnaire

Multiple Agreement Analysis of the Validation
Sample's Reasons for College Choice . . . . . . . . . .

Significant Differences in the Responses of Type A
Student 8 I I I I I I I I I I I I I I I I I I I I I I I

Significant Differences in the Responses of Type B
s tudent s I I I I I I I I I I I I I I I I I I I I I I I I

Significant Differences in the Responses of Type C
Student 8 I I I I I I I I I I I I I I I I I I I I I I I I

Significant Differences in the Responses of Type D
sweats. .' . . '. g '.*y. g» . . . ¢ 0 a a a o a o o a o 0

Significant Differences in the Responses of the
Unclassified Students, "Type" N . . . . . . . . . . . .

Item Content Associated with the Majority of Students
in the Validation Sample. . . . . . . . . . . . . . . .

Multiple Agreement Analysis of the Reasons for College
Choice of the Validation.and Cross-Validation Samples .

 

Page

14

22

34

45

47
53
54
55

63

67

68

69

70

71

73

77

Table No.

4.9
4.10
4.11
4.12
4.13
'5.1

Item Content Associated with Type A Students .
Item Content Associated with Type B Students .
Item Content Associated with Type C Students .
Item Content Associated with Type D Students . . .
Item Content Associated with "Type" N Students .

Percentage Order of Reasons for Selecting Colleges
of 276 High School Seniors . .

vi

 

Page
80
83
84
85

86

96

Appendix

A

 

LIST OF APPENDICES

Page
Items Received from the National
Opinion Research Center, together
with Tables of Responses . . . . . . . . . . . . . . . 144
An Illustration of Hemingway's
Multiple Agreement Analysis . . . . . . . . . . . . . 199

 

CHAPTER I

THE PROBLEM
Purpose of the Study

In the decision-making process of college selection, the students'
"reasons for choice" frequently become the prime determinants. The
college admissions counselor, whose time is limited by the requirements
of a heavy counseling load, must advise according to the factors which
the student indicates are significant for him. ‘His problem is further
complicated by the fact that stated reasons often are not well deveIOped
or articulated. Clearer definition in this area of students' selection
criteria should have immediate practical application for the counselor,
as well as for college officers concerned with enrollment.

It is a reasonably sound assumption that there is a basic core of
reasons for choice that are common to all students and all institutions.
Indeed, the-repetition of reasons encountered in previous studies con-
firms this impression. In addition, the standard guidebooks on college
admissions tend to emphasize and repeats a few basic variables.

: James Coleman, in pretested questionnaires reported in his book,
Th; Adolescent Societ , offered the following eight reasons for checking
by students who were subsequently to comprise the sample for the present
study: "(1) tuition costs and living expenses, (2) closeness to home,
(3) the high educational standards, (4) course of study offered,

(5) chances of a scholarship, (6) the-religious character of the school,

 

(7) your chances of getting admitted, and (8) peOple you know’who went

to those schools."1

Th2 purpose 2; £hi§.§£g§y was to isolate groups of students with
similar reasons for selecting their colleges, and to identify, if
possible, the distinguishing characteristics of these groups. The
questions this research has attempted to answer are: (1) Can one find
patterns in the reasons students give for selecting colleges? (2) Are
these groups distinguishably different in relation to other selected
characteristics? (i.e., socio-economic status, level of intelligence,
education of parents, peer group acceptance, self-concept, type of

high school attended, and type of college chosen.)
Need for the Study

With-the increasing necessity for higher education in American
society, the multiplying numbers of students entering college, and the
subsequent rise in admissions standards of our institutions, the role
of the college admissions counselor is daily becoming more important.
But at a time when research in the broad field of college admissions
is rapidly proliferating, knowledge of how a student decides which
-college to attend is pitifully incomplete. Douvan and Kaye, in the

study The American College, recognize this problem:

 

IInterview with Peter Rossi, Director, National Opinion Research
Center, Chicago, January 9, 1963. (Referring to the eight reasons
used in item 142 of the Spring High School Questionnaire, NORC Survey
412.)

ls

 

We have little systematic information about the

decision to go to college. The current renaissance

of research on the college student has not concentrated

on determinants of college-going, and the older studies

either focused on objective determinants like family

income and residence or stirred motivational variables

into one pot with these so that it is impossible to say

anything very clear about the independent Operation of

either type of factor.
The authors add, "If we know little about the decision to go to
college, we know even less about how adolescents choose the parti-
cular schools they enter."3

The available published material on college selection may be
divided into three categories: (1) survey research on college-going
trends; (2) research done by Specific institutions of higher education
investigating the reasons why students chose to enroll there; and
(3) philosoPhies and strategies on how to select a college. While this
material is relevant and useful, it offers little assistance to the
practicing:counselor in evaluating students' reasons for choice.

The first group of materials reports patterns in college atten-
dance and takes into account such variables as socio-economic status,

parental education, level of intelligence, academic performance, climate

of the high school and drOp-out patterns. It is illustrated by the

 

2‘Elizabeth Douvan and Carol Kaye, "Motivational Factors in College
Entrance," The American College, ed. Nevitt Sanford, (New York: John
Wiley and Sons, 1962), 199.

31mm. , 216.

 

works of the College Entrance Examination Board and Educational Testing

5 8

Service,4 Wise, Hollingshed,6 Coleman,7 Holland, and Iffert,9 While
generalizations about selection patterns are reported in this type of
publication, the specific reasons which influence students to select
particular institutions are virtually untouched.

Perhaps the most significant contributions in the latter area are
three articles by John Holland using the 1957 National Merit Scholarship
finalists.10 However, since these included only students in the top two

percent of the population intellectually, their results are not genera-

lizable beyond that group.

 

4G. Stice, W.G. Mollenkopf,.and W.S. Torgerson, Background Factors

and College Goigg Plans Among High Aptitude Public High School Seniors,
(Princeton: Educational Testing Service, 1956); and Background Factors

Relating 52 College Plans and College Enrollment Among Public figgh
School Students, (Princeton: Educational Testing Service, 1957.)

 

 

 

 

5 W. Max Wise, They Come for the Best gf'Reasons: College Students
Today (Washington, D.C.: American Council on Education, 1958.

 

 

6Byron S. Hollingshed, Who Shouldk __.Tg_CollegeZ, (New York:
MbGraw Hill, 1957. )

7James S. Coleman, The Adolescent Society, (New York: Free Press
of Glencoe, 1961.)

8John L. Holland, "Student Explanations of College Choice and their
Relation to College Popularity, College Productivity, and Sex Differ-
ences," College and University, Vol. 33, No. 3, (1958), 313-320;
"Determinants of College Choice," College and Uhiyersity, Vol. 34, No.
l, (1959), 11-28; "Parental Expectations and Attitudes About College,"
College and University, vol. 34, No. 2, (1959), 164-170.

Robert E. Iffert, Retention agd Withdrawal gf College Students,
(Washington, D.C.: U.S. Department of Health, Education, and Welfare,
1958.)

 

fl

Ines similar manner, studies which analyze the reasons student
populations gave for selecting particular institutions, are not
generalizable beyond these institutions. An intensive search through
the literature has revealed twenty-four such.studies, ranging from the
1926 survey by Hollingshed of freshmen in-eight Colorado colleges,11

to Foskett and Martin's investigation of university of Oregon fresh-

men in 1961.12 One of the most thorough of these investigators,

Kenneth Cleeton, investigating 3,329 freshmen in 14 white, public
Virginia colleges, concluded that students"reasons for enrolling in.a

specific institution were significantly related to that institution

and no other.13

A factor worth mentioning in connection with these studies is that

only four of them surveyed high school seniors.14 The remainder asked

11Ray B. Bollinghsed, "Why Do Persons Go Tb College?": School and
Society, V01. 23, (1926), 563-565. .

12John Foskett and Walter Martin, "How Students Choose Their Alma
Mater," Ovegyiew,'vol. 2, (June 1961), 23.

13Kenneth N. Cleeton,-"An Analysis of the Factors Which Influence
Students When They Choose a College," (unpublished Ph.D. Dissertation,
University of Colorado, 1951.)

Robert S. McCranahan, "A Study of the 1949 Iowa High School
Seniors to Determine Their Future Plans After Graduation," (unpublished
MgA. Thesis, State University of Iowa, 1949); John L. Holland, "Student
Explanations of College Choice and Their Relation to College Popularity,
College Productivity and Sex Differences," College and University,

Vbl. 33, No. 3, (1959), 313-319; Samuel Basta, "An Analysis of the
College Selection Patterns of Recent High School Graduates in Selected
Western States with Special Emphasis upon Nevada,"(unpublished Ed.D.
Dissertation, university of Southern California, 1960); Alexander J.
Plants, "An Analysis of the Plans for Post-Secondary Education of 1959
Secondary School Seniors in Connecticut and Certain Factors Which
Influence Their Plans," (unpublished Ph.D. Dissertation, University of
Connecticut, 1960.

 

college students why they had chosen the particular colleges in which

they were-enrolled. The validity of seeking the nature of the college
choice in retrOSpect seems questionable at best.

The third category in the literature on college selection includes
a host of books, pamphlets and articles on how to select a suitable
college. Typical of these are the writings of Lovejoy,15 Fine,16
Shosteck,17 Hawes,18 and Bowles.19 While these works offer sound
advice, sensible admissions strategy and encyc10pedic information about
colleges, there is no evidence to indicate that the variables they
suggest for consideration are, in fact, the variables actually employed
by students in their college selections.

The lack of available research on the selection process was
-recently reaffirmed in a letter from Dr. Holland to the author. He

wrote, "The difficulty you are having finding literature about the

 

15Clarence E. Lovejoy, Lovejoy's College Guide, (New York:
Simon and Schuster, 1962.)

 

16Benjamin Fine, How To Be Accepted By the College 2; Your

Choice, (New York: POpular Library, 1961.)

 

17Robert Shosteck, The College Finder, (Washington, D.C.:
B'nai B'rith, 1955.)

18Gene-R. Hawes, The New American Guide Ig_Colleges, (New York:
Signet Key Books, 1962.)

19Frank H. Bowles, M T_o Get Into College, (New York: E. D.
Dutton and Company, 1959.)

 

choice of college is simply because there is no literature. I experi-
"20

enced the same difficulty earlier.

At least one study has pinpointed inapprOpriate initial choice as
a significant factor in the tragic and costly college drap-out rate.21
This would seem to underscore the need for further exploration. In
addition, a clearer understanding of students' selection motivations

could conceivably affect policquaking in both individual institutions

and publicly-supported systems of higher education.

Theoretical Assumptions Underlying the Study

Discounting the aboveqmentioned books which expound systems of
admissions strategy, there appear to be no published theories on the
decision-making process in college choice. At first glance theories
in the area of occupational choice might appear to be relevant.
However, since the choice of an educational institutions is only one
segment in the process of vocational deve10pment, these are not
readily applicable.

There 25g theories, deve10ped on a statistical basis, which lend
themselves to analyzing reasons for choosing a college. These are the
typal theories dealing with pattern analyses of reSponses. Theyassume

that in any subject pOpulation "there are various kinds of underlying

 

20Letter from John L. Holland, Director, National Merit Scholar-
ship Corporation, January 8, 1963.

21Robert E. Iffert, Retention and Withdrawalo_f College Students,
(Washington, D.C.: U.S. Department of Health,Education, and Welfare,
1958), 109.

 

psychological structures or predispositions (not just dimensional ones)
"22

which result in patterns of responses. These patterns can be thought
of as expressions of the particular class or category of subjects in the
population. Each particular class or type is then defined as "a set of
subjects who are sufficiently similar so that the behavior (i.e., pat-
terns of responses) of any one member is the expected (most probable)
response of any other member."23 The classes are defined by the subject
matter of the-particular investigation and are not assumed to be
generally relevant, although they may be.24

On the basis of the method of pattern analysis, it should be
possible to locate sub-classes in a matrix of individuals times reasons.
While there may be equally valid ways of classifying students in college
choice research, the reasons for choice seem to be the most relevant.
The logic is simply that, given all the related information a counselor
knows about a college-bound student, the use and relative importance of
the information in counseling depends in large part on the reasons the
student expresses for selecting a collegee-the variables he is consid-
ering.

However, articles previously mentioned by ETS, The College Board,
‘Wise , Hollingshed, Holland, Coleman, Sanford, and Iffert have illustra-

ted differences in college attendance by a number of variables. If

 

zalouis L. McQuitty, "Agreement Analysis: Classifying Persons by
Predominant Responses," British Jougggl 2; Statistical Psychology, vol 19,
(May, 1956), 16.

23Peter W. Hemingway,"Mn1tiple Agreement Analysis," (unpublished
Ph.D. Dissertation, Michigan State University, 1961), i.

24mm, 19.

 

the theory that there are a number of sub-classes in a given papulation
holds, then some of the suggested variables should aid in describing
the classes within the-college-bound papulation which have been

identified by response analysis.
Description of the Study

This project is a follow-up on the papulation surveyed by
James Coleman in his study of high school climates, Th3 Adolescent
Society. Under a grant of $21,750 from the College Entrance Examination
Board, Coleman and Rossi, of the National Opinion Research Center and
the University of Chicago, began additional research on this papulation
in an attempt to ascertain "the natural history of college choices."25
It consisted of interviews with 110 college freshmen prior to construc-
tion of a questionnaire, administration of that questionnaire to 801
freshmen, and the comparison of the data obtained with that from fall
and spring high school questionnaires completed by.l,l49 college-bound
seniors in Coleman's original sample.26

One item.in the spring high school questionnaire pertained to the
reasons.students chose the colleges they expected to attend. (All coded
items, as received from NORC, are included in Appendix.A.) The students

who actually attended the colleges they specified in high school

 

25James S. Coleman and Peter Rossi,"A.Pr0posa1 for'Research~on
the Processes of College Choice," (unpublished research preposal
submitted to the College Entrance Examination Board, New'York:
January 6, 1958), 1; included as part of NORC Survey 412.

'26Interview with-Peter Rossi, Director, National Opinion
Research Center, Chicago, January 9, 1963.

 

10

constitute the sample papulation of 277 to be used in this study. The
sample is further described in Chapter III.

In fulfilling the stated purpose of the project, multiple agreement
analysis was employed to identify categories of common patterns of
reasons for selection of colleges.' The remaining coded items in the
three questionnaires were used to describe the differences among the
categories of students. In addition, a cross-validation was made to

determine the accuracy of the descriptions.
Definition of Terms

Collegefbgggg Student. Hereafter, for the purposes of this study,
a-college-bound student-is defined as one who declared his intention to
~continue'his education in the spring high school questionnaire, indic-
ated in that questionnaire his criteria for choosing a college as well
as the name of the Specific college he had chosen, and, at the time of
the spring college questionnaire, was found to be still enrolled at that
institution. 'With these restrictions,itie§a.likely assumption-that the
Astudent's’college-selection‘criterianwere relevant to the-college he
actually attended.

_ RgggggéResponse‘Typg. Individuals who agreed on a series of
,reasons-for selecting colleges-constitute a reason-response-type. The
type.itself can be-defined as "a sub-category of N peeple that everyone
.in the sub-category is more like each-of the other'N'minus-l persons

than he is like any other person in any other sub-category."27

 

hf

27Louis L. McQuitty, "Rank Order'Typal Analysis," Educational
gag Psychological Megsurement, Vol. 23, No. 1, (1963), 55.

 

11

Limitations of the Study

The interpretations to be drawn from this study are necessarily
limited by the information contained in the three questionnaires.
Because the data available from NORC Survey 412 is a revised work deck
of IBM cards, much of the original information in the questionnaires
and in Coleman's previous study of high school climates is missing.

Other serious limitations caused by insufficient data are:

(1) There are no raw scores on the IQ quartiles used by

Coleman. The names of the specific IQ tests have
been emitted.

(2) No data on grade point average or acadenic performance

of the subjects was available, either for high school
or college students.

Because Coleman's original sample consisted of 1,149 high school
seniors from 10 rural, suburban and urban high schools, generalizations
from it may be safely made to the general papulation of 1958 high school
seniors in midwestern United States. Generalizations to today's student
papulation may be somewhat less precise in view of the increased number
of students entering college, the changing nature of the college student
papulation, and the general mobility and variability of American
society. However, these changes in the ensuing five years would not

seem so drastic as to completely invalidate the results.
Hypotheses to be Tested

Although this study is well adapted to typal theories of pattern

analysis, in the absence of any organized body of theory relating to

 

12

college selection, it must be viewed primarily as exploratory research.
The format includes one hypothesis concerning the identification of
reason-response patterns, hnd a series of sub-hypotheses questioning
whether a relationship exists between students-of one reason-response
type and characteristiosgleaned from the remaining coded items.

The major research hypothesis is:

H1: There are patterns of agreement in the reasons
students state in the selection of their colleges.

With the rejection of the null hypothesis and discovery of reason-
response patterns, additional sub-hypotheses concerning the existence
of a relationship between these patterns and identifying characteristics
were written. Rather than state all the remaining items in hypothesis
terminology, the items themselves are included so that the reader may
obtain a better understanding of the nature of the questions. (See

Appendix A.)

CHAPTER II
A REVIEW OF THE.LITERATURE
The Nature of the Review

Because .this study is primarily concerned with the reasons expres-
sed by students:for the selection of their colleges, greatest emphasis
‘will be given to the twenty-four studies done by individuals or
institutions of higher education-attempting;tocassess these reasons.
Although such investigations frequently combined.discussions on charac-
teristics of the college-going pepulation, no‘attempt will be made to
thoroughly explore all the past literature in this area. Similarly,
the "How To Choose a College" manuals will be-omitted.

For the reader's convenience, previous-research-dealing with the
.reasons for college selection are reported tepically in Table 2.1 and
in their rank order of influence,oas reported by each investigator, in

Table 2.2.

ggplanationog she gables

In Thble 2.1 is included a listing of all the reasons for choice
evaluated-in the aforaumentioned-studies, under headings whichnwure
suggested by the-sheer weight of repetition in the review of items.

A few studies will be found whose primary emphasis was not college

selection-criteria, but which made-some contribution to this area.

14

TABLE 2.1

A TOPICAL.SUMMARY OF RESEARCH ON FACTORS
INFLUENCING STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

Factors

 

Friends

 

Farental Influence

Mother's influence
stronger than father's

women more influenced
by parents than men

Higher I.Q. quartile

somewhat less in-

fluenced by parents

Parents had surprisingly

little influence

inigher I.Q. quartile
less influenced by

friends

rand Acgggintances

Hbllingshed
Katz & Allport
Reeves
Johnson
Reinhardt
Corey
Robertson
Addington
MbGranahan
Eadaway

Hollingshed

Reeves

Corey

Salem

Hollingshed
Rats & Allport
Reeves
Lovejoy
Robertson
Addingten
Salem
McGranahan
Radaway

Gorey

"26

'26
'30

"31

'35
'36
'36
'39
'49
'51

'26

'30

'43

'26
'26
'30

.'33

'36
'39
'43
'49
'51

'36

Strayer

Franks

Schloemer
Mueller

Olsen

Bests

Plante

Foskett & Martin
Beezer & Hjelm
Forrest

Robertson

Beezer & Ejelm

Strayer

Franks
Schloemer
Mueller
Rolland

Olsen

Basta

Beezer & Hjelm

'51
'53
'53
'55
'59
'60
'60
'61
'61
'61

'36

'61

'51
'53
'53
'55
'58
'59
'60
'61

15

TABLE 2.1, Cont.

‘A TOPICAL SUMMARY OF RESEARCH ON EACTORS
INFLUENCING-STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

Factors

‘—

Investigators and Publication Dates

 

Opinions of friends

more important to
women than men

'gggh,8chggl ggggonnel

Among high school
personnel, teachers
were most influential

Higher I.Q. quartile

A less influenced by
high school personnel

szflfluaraleaeeasefl;

‘Re ttve

Opinion of relatives
more important to

women than.men

Reeves '30

Rollingshed '26

Johnson‘ '31
Lovejoy~ "33
Robertson A'36
Addington '39
Salem '43

Mcoranahan '49
Lippsett‘ «HM'SO

Rollingshed '26

Robertson “'36
Hueller '55
Gorey

Hbllingshad '26

Reeves '30
-Johnsen -'3l
Robertson i'36

Katz 6: .Allport '26

Reeves '30
Johnson <'31
Lovejoy '33
Robertson "36
Reeves “'30

Hadaway
Strayer

Franks
Schloemer
Mueller

Olsen

Bests

Beeser 8: H1 elm

Addington
Lippsett G: Smith
Schleemer

MbGranahan
Strayer
Franks
Olsen
Basta

'51
'51
'53
'53
'55

m .59

'60
'61

'39
'50
'53

"49
"51
"53
1'59

'60

 

16

TABLE 2.1, Cont.

A TOPICAL SUMMARY OF RESEARCH ON FACTORS
INFLUENCING STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

Factors

Investigators and Publication Dates

 

College Alumni

Size 23 the College

Women more concerned
with small size than
men

Proximity-Location

Both sexes rank close-
ness to home as
highly desirable,
but men more than
women

Reputation 25 the College

Prestige ranked low as
an influencing factor
in 1930 and 1935

Johnson
Robertson
Addington
Hadaway

LoveJoy
MCGranahan
Cleeton
Strayer

Holland

Reeves
Johnson
Reinhardt
Corey
McGranahan
Cleeton
Hadaway
Strayer

Holland

Lippsett & Smith
Cleeton

Franks

Schloemer
chaughlin
Holland

Reinhardt

'31
'36
'39
'51

'33
'49
'51
'51

'58

'30
'31
'35
'36
'49
'51
'51
'51

'58

'50
'51
'53
'53
'54
'58

'35

Franks
Schloemer
Olsen
Basta

Franks
Holland
Forrest

Franks

Holland

Olsen

Basta

Foskett & Martin
Boozer & HJelm
Forrest

Olsen

Basta

Foskett a Martin
Forrest

'53
'53
'59
'60

'53
'58
'61

'53
'58
'59
'60
'61

-'61

'61

'59
'60
'61
'61

 

17

TABLE 2.1, Cont.

A TOPICAL SUMMARY OF RESEARCH ON FACTORS
INFLUENCING STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

 

 

 

Factors Investigators and Publication Dates
Both sexes rank aca— Holland '58
demic standing as
highly desirable, but
women more than men
Both men and women Holland '58
place very little
weight on research
publication
Religious Affiliation Reeves '30 Basta '60
Holland '58 Forrest '61
Students in denomina- Reeves '30
tional colleges
ranked church affili-
ation ahead of the
course of study offered
women are more con- Holland .'58
cerned with reli-
gious affiliation
than men
Course of study_pffered Katz & Allport '26 Franks '53
Reeves '30 Schloemer '53
Johnson '31 ucLaughlin '54
Corey '36 Holland '58
MCGranahan '49 Olsen '59
Lippsett & Smith '50 lasts '60
Cleeton '51 Foskett & Martin '61
Strayer '51 Forrest '61
Vocational Training Katz & Allport '26 Salem '43
Available Anderson '41 Plante '60
Publications of the Johnson '31 McGranahan '49
College Robertson '36 Schloemer '53
Addington '39 Basta '60

 

18

TABLE 2.1, Cont.

A TOPICAL SUMMARY OF RESEARCH ON FACTORS
INFLUENCING STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

 

Factors Investigators-and Publication Dates
Publications were not an Reinhardt '35 Strayer '51
important influence Hadaway '51
Campus Visits-~Phygica1 Lovejoy '33 Franks '53
Plant Robertson '36 Schloemer '53
Salem, '43 Holland '58
Lippsett & Smith '50 Foskett & Martin '61
Cleeton .'51
Both men and women Holland '58
placed very little
weight on physical
facilities, but men
did more than women
Job Placement After Katz & Allport '26 Schloemer '53
Graduation Lovejoy '33 Olsen '59
Social Life Katz & Allport '26 Olsen '59
Cleeton -'51 Basta '60
Schloemer -'53
College campus activi- Hadaway '51 Franks '53

ties were not an
important influence

Both men and women Holland '58
placed very little
'weight on coeduca-
tional status, but
'women did more than
.men

 

19

TABLE 2.1, Cont.

A TOPICAL SUMMARY OF RESEARCH ON FACTORS
INFLUENCING'STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

Factors Investigators and Publication Dates
C0St 23 Attending 0011689 Reeves '30‘ ‘Franks '53
Johnson '31 Schloemer '53
LoveJoy ;'33 Holland '58
Reinhardt '35 Olsen '59
Corey '36 Basta '60
Lippsett & Smith '50 Foskett & Martin '61
Cleeton '51 Beezer & HJelm '61
Hadaway '51
Men were more influenced Reeves '30
by cost than women
Costs had surprisingly Salem '43
little influence
Junior college students Strayer -'51
ranked finances as a
more important factor
than did students
attending state-sup-
ported institutions
Both men and women Holland '58
placed very little
weight on costs
Miscellaneous
"Kind of college (Teachers Colleges, Liberal Cleeton '51
Arts, Engineering School, etc.)" ~ Hadaway '51
"Type of college (Coeducational, Segregated, Cleeton '51
Public vs. Private)"
"Support of the college (Public or Private, Cleeton ~'51
Religious, etc.)"
"Objectives of the college" Cleeton '51
"History and tradition of the college" Cleeton '51

 

20

TABLE 2.1, Cont.

A TOPICAL SUMMARY'OF'RESEARCH‘ON FACTORS
'INFLUENCING STUDENTS IN THE SELECTION OF THEIR COLLEGES

(References are Listed in the Bibliography)

 

 

Factors Investigators and Publication Dates
"’lhe Basic College, making it possible to Schloemer ~'53
obtain a liberal education in the first
2 years"
"To-secure a broad education for life" Anderson '41
Plante -'60
‘"General Self-Improvement" Rate & Allport '26
"To discover a suitable occupation" 'Plante '60
"To secure training necessary to enter Plante '60
a professional school
"Reasonable course requirements" Olsen .'59
*"Entrance requirements" Basta '60
"Faculty-Student relations" Cleeton -'51
Franks 2'53
"Friendliness of the student body" Cleeton 5'51
Schloemer '53
*"Heterogeneity of the student body" Olsen -'59
‘"Football and basketball teams" Johnson o’31
-"Interest in sports activities" Cleeton 2'51
Boats \'60
-"National reputation in athletics" Schloemer '53
"Opportunity to live away from home" Olsen ~'59
""Articles about the college read in Jehnson .'31
' your-hometown-paper
"Miscellaneous" Holland '58
Basta '60

 

2a

TABLE 2.1, Cont.

A TOPICAL SUMMARY OF RESEARCH ON FACTORS

'INFLUENCING STUDENTS'IN THE SELECTION OF-THEIR COLLECES

(References are Listed in the Bibliography)

 

Factors investigatorsland Publication Dates

 

College day programs and off-campus
activities were not important
influences

Athletics were not an important
influence

Percentages of frequencies of reasons
varied among colleges

Percentages of frequencies of reasons
varied by sex

"There is no single factor that actuated

student groups in all colleges in the
same degree, or in even approximately
the same proportions, nor by the same
rank accorded these specific factors."

"No single factor was more important

than all other factors in influencing
freshmen to select the University of
Missouri."

"Students appeared to make choices in
the same consumers often, if not
usually, buy household goods; they
select colleges by means of vague
notions about reputation and values
which they seldom can document
meaningfully."

Hadaway 1'51
Schloemer -'53
Hadaway '51
Franks '53
Reeves '30
Reeves '30
Cleeton '51
Mueller '55
Holland '58

TABLE 2.2

\

AS REPORTED IN STUDIES FROM 1926 TO 1962

THE RANK ORDER OF FACTORS THAT INFLUENCED STUDENTS' CHOICE OF A COLLEGE TO ATTEND

INFLUENCING FACTORS

RESEARCH CITED

COLLEGE FACTORS

PERSONS

 

 

Q2

3931100 ONICNHIIV HO
HJI1

HHIJV

INV1J
"SIISIA

3H1 H0
H1EV1IVAV ONINIVHI °

CHHHHHO ACRES JO
NOIIVI1IHHV

3931100 3H1 HO

N011VOO1"

3931100 3H1 HO
INNHTV

1HNNOSHHJ

1HNNOSHHd 1OOHOS
SEONVINIVHOOV 9

SNOSVHH J0

HZIS

CHESHIVO SVM VIVC

 

.-

 

 

 

 

23

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ N N L o N LN 32 .L odours 3338
m N N L .L LoN LNNL To a snooze 338a
- o L N N o oNILI 33 mm. .5 53oz o oooxmom
N NiN L o N NNNN NNNL, .1», NNooooLN
L,NL.o N NL 7m L L N.N.L N N oL LL .mmi N oL NN NNNH. Nio .LLL .NL comLo .
A NL.N.N N N N o an+, oL N NL 1|Hwamu_oL 1N4, NL NNN NNNwl .LLL Lo.p comLo
NLrN LL N N N.L NJN o oL o NL oNN‘ NNNL ,LN.moLoaom oaoLLom ,
i N NL NL N LL N.L N.N N N o NL eon NNNL LNNmoLoNo omoLNmmJ
oL-N L NL NL LL N N N o.o N N N‘ N N! NL NNN NNNL NNmumoN
+ «AN N w! N NNNL NNNL LNLooLLLmoLL
qu N‘ o N L N. N o N N N L N¢.N oooL NNNL mLmoaooLooN
N o, o N LL L N N NL NL N .N NL NoN NNNL Nmeooom
L N N o o o o N NNN mmoL ommmNmmmw ,
N N IMLIL M L N o1 N N N NNN LNNL meadows
LZWNJMZ mmmmmmmmammsm m
mimaamnwfmsm mmwmmamsam w s
N vmmmmamvmmoumsmsuu a u
1 u A. ..u 0 m u m m 0 S W
mmmmma i mmmwsmm» mam v m
m .. m m m N S o . fl 1 w a M m
s N .e o a m a m N 0 mm
am Lu . _ a. mm .A mm o. n, mm a” ”H mm m"
o m mm N m m m 1 o m s m
o u m o N a m
0 m 0 0 W m
m“ Lu ni no
.oNLz N Neoao<a mouLLoo Nzommmm
mmoaooa ozLozmoomzL omaNo momoumnm

 

 

 

NNNL on oNNL some NNLopNN 2L anamomum No .
ozmaao o9 NoNLLoo some NoLoNo .Nazoopam omozuaLLzL Name Neoaoom mo_moomo LzNN may

.ueou

.~.N unm<8 .

1H”! in.

 

 

 

 

24

.afluuLoao>Ld= no aowoafioo vouuoamsauouauo opum new readouuooonhaoue>uua odLe

Nanouusuwuonuo aLsLmuu> «HI Scum duodenum we humans voooLuaavn: do no hu>uno < .ocounoo .Ammaa .oenhv

NN .Lo> .ooLoooooN No Lassen oLoLNNLLL ...oLoLN.LL> 3 «.328 o 93.86.. 63333 .N..LN

62: NL «938 3:88
No. 8.533 oNL No .ooooooo No ooooLooNN on»? N25. 4 .NNNuLNN .GNNL .36 .N .3,
.eoaueospu nommum mm.aaeunon =.Naoeosoo oaooeo cu noseaoum neoconawdn cesareeoauuonomw

 

, .omoLLoo .auoeooou‘cuouaoanae uoaoauoLvno
s. NoLoowwwo. 28: 9? NNNL o8 83, 5. $33: No No Nerd... o .NLNLLN £83 .323 .No .Lo>

 

......

.emma,eL.oxoaueoz mo NuLouo>No= oeu.woLocouua censuses nNn mo Nephew oNLaoeoNoaosv 4V .LLNuNcN
. IIIIIIIII. nl.

 

.....

.ana_sa owoaaooeheaoo uo_mmrov5ua_mo noesou.voa0aoauun= no mo hm>u5o1<

.oNNN .SeNL , .3333 Leo. oosLN «to» 35 £9.28 3 3., Loo 833 am £333 .N ooooNoLoN

 

.82 5 NLN o8 .NNNL oLNoN .oNNL. oL. ooaooooo oNN moNNL Law oNNL 825%
owoaaoo Housuasoauw< auoxoa euuoz venomous can dueeceuw «mc.L mo amoune.cuuodeowuoosu < .wwumu
.Aammfi .uaomv .Nm .Lo> .sowuoonvu :.owo-oo m wauaooeo cw nauseousnsH nouns: .docdeOH .H.3¢

.cnaa ea "wBOH ea m new .ooaauneua
6L L .ouoaeeuNszL L..mxonusuu ed M NanowusuLuaeL duo scum couscous Nn¢.¢ mo hm>u=n oHLunnoNuaonv 4

.oL-N .LNNNL .NNBNLoNL .N .Lo> .o.oL._oooooN ooouLN mm Loans. _o_N._ =.NoNoLLoo SH? .833 .2 oLLLNN

.oaaa_su Nuaauo>Ln= onsoouhm as concerns con.m mo,
ho>uso uoqflxooeo < .on .Ammma .LHLm.anuoozHquou.sozv .omuq uno.muoLLoo .Huosoom .u.szL vooao
.AmeL .oaoum dosaumouo "snow.sUZv .aovsuauu< .ouoonsum .uuoaLH4 sm,uhoam use aquLoNconu

 

 

.wNaL ed
adowusuLuasw wsNueouwuoouwov oumuoLoo uewgo nu dosflaouw new. N.wo hookup ouLQdeoLuomeu < .mom mom

.AoNaL .Nozu .mN Lo> nmmmmmm.oeo Looeom =~owu-oo 09 co aeouuom av 553: .vesuweaaaom. .m mama

 

25

.23 a.“ Assays 30:00 Seam. HERE .4 s0LLNL0um uses-Esau 3N. mo Anson 0uLaonoEAa0LL—o < Jot—n
.NNNNL .NuLuoooLoo oooLooL .ooLoooNoooLo .o.oN ooooLLooNoov =.oNoLLoo ooooN oLoooo .<.ooeNooN Lo
LLB—em cu ausovsum 08$:de £0.25 anooomm Lens: 05 0aLfin0uon cu. Serum 3. .9395 .n. Uganda:

.LNNL oL ooooLNaoo .oLoaoo
35:25.50: mo Nedu no 3003—.on Loosen emu: 03H mom we hater 030533.003. 3:8. 4 .813
. .LILM .3 33.30:: .3 L030 .33.. .033 no 593.0225 023m .3005 .45 ganja—nanny :JNONLSm
Loonom. nwLm 03H m3..— wo aecwueu009£ ungodaau can. defines—om Loosen, Liam each 05. 235039

3 33 none: LLL. songs—Loo: uuoeoa he Logos—once Stew 0 mo souuoaLmNr. .u0houum .m .3383

.LNNL oL oNoLLoo .uuoooooL oooom
oooLooL oo.ooaeoooo NNN Lo Nopoao o .NNuNN .NLNNL ..>ozv .NN .Lo> .Looosoe muoLLoo .ouoeoooa
...0w0.:oo £05009 ououm coma—Sn no Sons“— 3 3:035. Sassoon—m3” auouomm: £03025 .3903

 

233.. LL.“ defines:

“3.5.3 «0 2333.325 0.353.; mongoose 33.215 .0933 o.— LLL 00.369300 Leda d06500um mun m we

hogan 03093300,: e .3an 623300 «e DLcN0>NL5 50.33.3093 .n. LE v0enaaesa§v =.0m0.:oo
< 30er 505. :25 aus0vsum 00a0sLmLLH £0.33 aaouomm 0.3 no 390—32 5.. 5300.5 .w Anecdoxz

.oNNLAoL NwoLoooooN
no 02532; “.030:qu 0.3 on. LL0LL=L00NwL «3 mo, h0>uso 0uLeedoEe0nv 4 .awuneou .Aumaa . :53 .NN 23>
.HLLNNQLLLLLL Lona. $0.300 :.0w0.:oo uoasuwuusm o. 03er aus0v=um .33: .5925 .MJ Loco 3025.3 :LnL

.o.0.LLaqooEa0nv Leon. 0H0: oer 9.3.300 v0uro0a0o banana ooo.m 05 we Nmfim 3L5 0H0? monsoon

0am. .moz LL.“ 30.300 Loosen. ewLe 25H as; no A0559 0udmadoEo0sv nodes < .ONIS. . qwlqfl .um

.NoLLooszo oooLo LNNoNL .oaoL No NoLoNo>LoN oooom .oLooeo .sz ooooLLooNooL =.ooLoooooNo oooN< oooLN
335m .305. 0dL530n cu an0ad0m Looeom. swam 03H 93." 05 m9 bosom 4.. .eoemnouooz .m uu0oom-

E0WWOE¢NLH 0Ns£00uhaom 0.259%; on dado—onus «on «o House d £511..— .Amooa . 50.3 .9... 23>
.ooLoooooN No Looooon oLoL NL> =.oNoLLoo o No ooLoeo oLo.oL oooooLoo_LoL oooz one: .aoLoN‘.<.oLL

 

:33. 5 32.05:: 3.5390st 05 on. camham ooo.n. mo Noise. <
.LNLo £32. .885 dooofiLLLNo. NoLoogLoN .LoLLoNNoooLLc 63383 8&3 3 8:33 gouge—«NS
0w0=oo we .30» uaHLLH HEAL. cu dogma 03.33 330935 mo. sowucsum 0.5... 303022 .463

 

 

 

 

a

 

 

‘(iIILI N'FI'IOH

>lr~2§l

Hamish-56‘ ~ atrVIn

c I:
7“ IMHO

AiFIIFIV
grill- I ‘1!

 

26

.aoaa a“ :ow0uo mo muauu0>fiab 0:0 ac nuano0uw awn mo 50>u=u < .mu .Aaoma .0aahv

.N .Ho> .B0H>u0>o :.u0uuz maa< uwmaa 0noono aua0v5um 36m: .dwuunz u0u~03 wad uu0xuoh anon¢u

 

.m0ms :H waoonuu Hauauu0aaoo v0u00~00nhaaovanh «N scum

uncau0n #00500 swan mm~.N mo h0>usn 0uu0auowuu0sr v0aaua 4 .Aooma .uauuuu0aaoo mo auwuu0>aab

.aowuuuu0uuao .nqnm v0nawansm==v :.uduam ufl0na 00a0=HwaH snags wuouomm awauu0o can asoeuo0aaoo
a“ whoaa0m doosum unavaoo0mnuoom How mfluam 0nu mo maohaud< a<= .0uawam .h u0cfi0x0a¢w~

.wmmfi a“ 0w0fiaoo 0u0um afloaaHHH au0uaum

us whenma coaunosva muuua0a0~0 am can muoanHH mo >uw0u0>wap 0nu um nuanua.uoauao=v0vmuaud0ama0

mnu mo h0>u=0_0u«aaa0Huo0=v ¢ .memem .Amnma .yum0mv .oH .Ao>a.:oauao=vm “050009 mm.H0aM=OH
=.00w0HHoo ““059 wnHuu0H0m you 0d0000m .mud0usum "cowuaoswm.>uauu0auam: ..uh a0aao .0 mammNN

 

 

 

.nmwa a“ uu0caH3

magnumaoaum uau0z Accowumz «am we 0Hm500 Eovduu 0,mo z0>hsm 0HH0auoaua0sv < .mamnmam .Awmma

.wafiummv mm .Ho> .NMMMMMMNmm.mmm.mmmmmmw =.00ua0u0wwun Num can .muw>wuosvoum 0w0HHoo .zuauwasmom
0w0HHoo cu coau0H0M‘HflunH can 0uwonul0w0aaoo mo maOfiumaaanxm uaovSum: .vamafiom .A anchau

.mmaa aw nouusvouw Hoosum swan

0v0>0z mum scum 0H0 0H0: v030w>0u mu~=u0u 0:9 .n0uuum :u0un03 unwwm scum o0uuskuw Hoosun am“:

new we a0>u=m 0uacaaowu00=v v0afiua < .Aoomfi .¢Hahowuawo au0su50m wo_zuuuu0>fiab .a0fiuuuu0muan

.n.vm v0smwap=aaav :.wva>0z comp aquusmam Haauonm an“? noumum :u0uu03 u0uu0a0m aw u0um=vuuu
Hooaum swam ud000m mo n¢u0uumm dowuo0fi0m 0m0HHoo 0£u mo uguh~0n< :4: .oumum H0aaumo~

.nm¢H aw «usonnfiz mo huauu0>aab 0nu u0.a05£00uw unm.~ mo h0>h=a.0uauadoau
-85.. < .32: .2583: no 3030325 .833333 .nfim wonmflnsmfiv ...H«u=o3§ no 30305::
0nu uouaom ou n0a£noum wawaafiwom 00¢00HwaH udSu muouuah cud“ huflnudH a4: .umaH0az uu0nHHm¢H

.mnma a“ 0w0HHoo 0u0um duwfisoaz u0,:0aauoum
wauaouafi ooo.~ mo hvsun 0uwoauoaua0sv can .h0000 .30H>u0udu.q< .mNINN..Amnma .hgsnv .c .aOh
.Nau0uuwsm mdowuaaum oaansm 0u0H~oo :.0w0HHoo usow 0uoono h0g9 >33: .u0a0oHnom .A 0000.53»H

 

 

 

 

27

Although the compilation in Iable 2.2 is illuminating, caution.mu8t
*be exercised in contrasting individual entries in the table. The studies
'reviewed were extremely dissflmilar in several important respects.

There werertriking differences in the types of samples employed by
different;investigators. Of those researchers who sampled college stu-
dents in a.sing1e institution, some worked with one class, and some, like
Johnson and Reinhardt, with several classes in different years. The in-
stitutions themselves ranged from.multi-purpose universities to such
‘specialized schools as North Dakota Agricultural College and Rochester
Institute of Technology. Hollingshed, Reeves, Addington and Cleeton
collapsed data from many institutions,.while'McGranahan, Holland, Basta,
and Plante sampled high school students. Holland limited his sample to
National Merit Scholarship Winners, while Olsen concentrated on elemen-
tary‘teacher~candidates.

Differences in the initial purposes of the research further serve
to differentiate between-studies, undoubtedly accounting for different
emphases in the types of items investigated. In order to better aim public-
-relations materials, Robertson, as President of Gaucher College in 1936,
asked which persons in the students' experience had been most influen-
rtial to their'choice.25 Basta, sampling high school students in
Nevada,‘was attempting to discover why such a high proportiénuchhsscto

26
attend institutions out of state. Plante was concerned with the

 

25Robertson,"Who Influences Freshmen-to Choose CoucherT," Journal

of‘Higher Education, Vol. 7 (May, 1936), 271.

26Samuel Basta, "An Analysis of the College Selection Patterns of
'Recent High School Graduates in Selected Western States with Special
Emphasis upon Nevada," (unpublished Ed.D. Dissertation, University of
Southern California, 1960), Chapter I.

“' P?.:‘*__---'.t4 «5y

 

 

 

28

general area of post-high school plans, and sampled only five reasons
for college choice as one part of the larger design.27

The methods of obtaining data and the instruments employed also
varied widely, the number of items alone ranging from two in one study
to fourty-two in another. While Holland simply requested answers to the
question, "Why have you selected College?,"28 Katz and
Allport asked their subjects to check the three.most important of eight

items,29 and Schloemer asked that they rate fourty—two items as "among

the most important," "important enough to be mentioned," or "unimportant"
in making the decision to enroll at Michigan State College.30 Some

researchers, like Cleeton,31 used instruments that were-designed all or

in part to require preferential choice.

 

2ZAlexanderJ. P1ante,WAn Analysis of the Plans for Post Secondary
Education of 1959 Secondary School Seniors in Connecticut and Certain
Factors Which Influence Their Plans," (unpublished Ph.D. Dissertation,
University of Connecticut, 1960), Chapter I.

28John L. Holland, "Student Explanations of College Choices and
Their Relation to College Papularity, College Productivity, and Sex
Differences," College and University, Vol. 33, No. 3, (Spring, 1958),
317. 7

29Daniel Katz and Floyd H. Allport, Students Attitudes,(New'York:
Craftsman Press, 1931); cited in Bennett, M. E., College'ggg Life,
(New York: McGraw Hill, 1933), 56.

30Clarence L.-Schloemer, "Why They Choose Your College," College
Public Relations Quarterly, Vol. 4, (July, 1953), 27.

31Kenneth N. Cleeton, "An Analysis of the Factors Which Influence
Students When They Choose A College," (unpublished Ph.D. Dissertation,
University of Colorado, 1951), Chapter III.

 

 

29

Imbalance in Table 2.2,occurs-because of the difference .in methods of
reporting conclusions. While Addington collapsed his data from fourteen
Virginia colleges into one rank order of six influential factors,
Foskett and Martin divided one class of 588 University of Oregon fresh-
men into three types, and reported results separately for each. A
glance at the chart will reveal three entries for this study: one
reporting-on the "pred13posed" group (P), who had chosen the Univer-
sity of Oregon before their senior high school year; one on the
"thinkers" (T), who had always planned on college but didn't decide on
Oregon until during their senior year or later; and one on the "late
starters" (L), who did not plan on college until during or after their
senior year.32 Similarly, Holland and Olsen have been given double
entries because of the division of their samples in the reporting.

The entry for Schloemer deserves Special explanation. While his
main emphasis in sampling 1,000 Michigan State College students was to
evaluate fourty-two college factors, he did administer a separate
questionnaire-evaluating the influence of five categories of persons.33
Because these-were not collapsed, and by coincidence fit conveniently

in the right and left halves.of Table 2.2 reapectively, they have been

entered side by side but separately, as reported.

 

32John Foskett and'Walter'Martin, "How Students Choose Their Ahma
Mater," Overview, Vol. 2, (June, 1961), 23.

33Clarence-L. Schloemer, "Why They Choose Your College," College
Public Relations Quarterly, Vol. 4, (July, 1953), 21.

 

30

Frequently the items themselves posed problems of classification.
Olsen reported that the number one ranking reason for students at
Eastern Illinois State College was, "This institution is well known for

the calibre of its:teacher education program"34--simultaneously reflect-

" "course of study offered," and

ing the influence of "reputation,
"vocational training available." It was entered in all three-cate-
gories in the table.

Where more than one number occurs in each space, it indicates thatan
influence was revealed more than once by the instrument employed.
Olsen's number one.ranking item above was followed by (2) "All students
at this institution are provided with a well-rounded.general education
background," and (5) "the reputation of this institution as a place to
get a good educationiswideSpread."35 Both were-credited, with number
(1), to the "reputation" column.

For the lack of Space, Table 2.2,omits some of the more unique items
which defied classification and which.the investigator reported to be

relatively unimportant. For'Holland's 1957 survey of males, for

example, twelve of the fifteen reasons investigated.were reported.
Criticism of the Literature

In interpreting the results of these diverse investigations, one

must keep in mind the obvious pitfalls of mush-of the-reporting.

 

A34Hans 0. Olsen, "Elementary Education: Students'Reasons for
Selecting Their Colleges," Journal gpreacher'Educatiog, Vol. 10,
(September, 1959), 316.

351bid., 317.

31

(1) Researchers in twenty of the twenty-four studies questioned
students already enrolled in college regarding-their selection moti-
Vations. Only Holland, Plante, Basta and McCranahan surveyed high
school students. However, the-purposes of the remaining.studies may
have differed from these, the basic validity of asking students already
influenced by college life to recall their initial motivations is
questionable. I

(2) Although most of the studies-apparently involved the question-
naire method of research, researchers infourteen of the twenty-four did
not describe.their instruments and made.no.mention of prevalidating
them. Therefore, the questionnaires used may not have represented an
exhaustive, or even a complete listing-of the reasons actually employed
by the students. They may, in fact, have represented only those reasons
the-author wished to sample. (This is evident in evaluating "reputation
of the college" as an influential factor, and will be discussed later.)
Among the studies weakened in this-regard, eleven of which were pub-
lished before 1943, are those by Bollingshed, Rats and Allport, Reeves,
Johnson, Lovejoy, Corey, Reinhardt, Robertson, Addington, Anderson,
-Sa1em,~Lippsett, and Smith, Hadaway, and Foskett and Martin. For all but
four-~those by~Katz and Allport, Reeves, Robertson and Selene-no
explanation-was offered as to how the-final rank order was achieved.

(3) A third area of error involved carelessness in the use of
statistics, _Katz and Allport, Reeves, Robertson-and Salem erroneously
equated percentage with rank in assuming that those reasons acknowledged
by the-largest proportion-of students were the "most important." To

accurately assess relative importance requires the use of a rank order

 

32

technique demanding perferential choice on the part of the student.
Since these studies lacked such techniques, the relative order of
importance of the individual items sampled cannot be inferred. However,
they do indicate the prevalence of an influence among college students.

Four studies which compared two samples did not employ statistics
in testing the differences between samples. These included the investi-
gations by Reeves, Johnson, Reinhardt and Addington.

One investigator was guilty of an invalid statistical manipulation.
When comparing the differences in reasons for selection between students
at two institutions, Olsen used rank order questionnaires, assumed equal
intervals between ranks, and applied the parametric "t" test.

In only seven of the remaining studies--those by Cleeton, Holland,
Franks, Mueller, Schloemer, Basta and Plante-ewere the assumptions and
conclusions-clearly stated and the statistics correctly employed. Of
these, all but'Holland's and Schloemer's represented doctoral disserta-
tion projects. The investigations of McGranahan and Strayer, reported
as masters theses, were not readily obtainable and so must be-excluded

from judgment.

Trends Revealed by the Literature

-Bearing in mind the weaknesses_in research design described above,
and the imprOpriety of comparing such divergent studies, one is never-
theless justified in scrutinizing their conclusions side by side, if
only because the researchers themselves reported them in rank order.

It must be remembered, however, that the number of times an

influence appears in Table 2.2 is partly a function of how often it was

 

33

included in the choice by the investigators. Where separate results
were reported for a divided sample, the study was counted more than
once. Also, the rank of 5th in 5 must not be equated with 5th in 12 or
5th in 42. Despite the difficulty of making precise observations from
the data in Table 2.2, the following table, clarifying the compilations,
may help to identify trends.

This researcher was reluctant to collapse the results of studies
over the-entire thirty-six year period, particularly in view of the
lack of statistical saphistication of the early studies. Since a
natural chronological division seemed to occur between those studies
prior to 1943 and those after, they have been divided on that basis
for the purpose of comparison.

Among the early studies, parents appeared strongly influential in
-college choice, friends and acquaintances somewhat less, with high
school personnel, relatives and college personnel following in descend-
ing order. Indeed, college personnel were credited with little influ-
ence. The influence of alumni is difficult to ascertain; they ranked
third out of six items in the only study which included them.36

II and

Among the college factors, "vocational training available
"proximity-location" emerged the strongest, with "course of study" and
"cost of attending" following. However, because in studies with broad
inclusive categories of choice, "course of study" could conceivably

have been chosen by students with vocational aims, its true position

of importance must remain uncertain.

 

36L. F. Addington, "Choosing a College in Virginia," Virginia
Journal 2f Education, Vol. 32, (June, 1939), 406.

THE FREQUENCYWITH WHICH INFLUENTIAL FACTORS

TABLE 2.3

IN COLLEGE SELECTION WERE RANKED

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Studies Before Studies Since Total 2.1:: §_1_l
1945 1945 Studies
0) m m
5 B fg’ '8 8 5 gt 8 5 8'3
up: u UN u .3 an u egg
H g H E-| 3 H u H H m H u H 6-:
m 0 mun ‘H on O mm '44“ m 0 mm 'H g
o o I o o o I o o o I o
13".?" .33 5" .51" .3 3" 5" .3
“'8 up, 03 E—c'o [-'u'o 03 [-13 Hg 03
°§°f€ 7.3. °§ we. °§°§:.
Factors 0.,“ 0-0-1 46' H 00% O- M 3H 0- 604 3H
2 z‘ E z z H z' z E-c
A T?
1? rs ' 8
Parents 6 9 ’ 9 3 9 12 ‘9 18 21
. FriendL etc. 2 7 .8 2 6 jg]; 4 13 19 ‘
H.S. Personnel 1 6 8 0 7 ll 1 13 19 r
‘ College Personnel 0 1 6 O 2 .4 0 3 10
Relatives 0 2 5 O 33 6 0 5 ll
' College Alumni 0 1___ 1 O 3 6 O 4 7
College Factors
Size .of College 1 l l i 3 .4 6 :4 L 7 ' ‘
Proximity-Location _'3 4 4 3 9 _16 6 13 20 ‘
‘ k putatiog o o o J 13 _;7 19 1_3_ 17 19 ,
Religious Affiliation .0 0 1 O 2 _f_§ O 2_ 4 ‘
Course of Stu_dly 1 4 4 6 -9 1; 7 L3__ 16 r
. Vocational Trainin .3 3 3 3 3 .3 6 6 6
‘ Colleg Publications 0 1 2 O l g 0 L 4
' Visits-Physical Plgt l 2 '2 1 _3 8 L .5 10 l
‘ =10b Placement O 2 2 O O 2 O 2_ 4 ’
‘ Social Life 0 o _; I o o 4 o o 2.4
Financial
. Cost of Attending 2 5 6 92—4 9 12 4 14 18
*Miscellangous
g_ .4 4 4 12 27 6 16 ‘31

 

 

 

35

Similarly, it is difficult to draw conclusions about those items
which only appeared in one or two studies. "Size of the college" is
interesting because it ranked second in the only study which included

37
it. 7When "religious affiliation" and "social life" appeared in

single studies, they ranked eighth38 and sixth39 respectively. Also of
interest is the fact that "reputation of the college" was not considered
important enough to sample.

In retrospect it would appear that the emphasis of the early
researchers was on persons as influencing factors in the college
aselection process. Six categories of persons were included in studies
thirty-seven times, whereas eleven separate college factors, including
cost, appeared only twenty-six times.

While the post-war studies are not strikingly more numerous than
the earlier ones (thirteen published studies in the later seventeen-
year period, as compared with eleven studies in the earlier nineteen-
year period), one is struck by the numerical increase in items sampled,
and the increased attention to "college" factors. The six categories
of influential persons appeared in studies fifty times, while the

‘eleven "college" factors, including;cost, were ranked eighty-seven

times!

 

37Clarence E. LoveJoy, So You' re Going To mllege, (New York:
Simon and Schuster, 1940), 90.

 

3821113 n. Reeves, "Which College?," Journal 01 Higher Education,
Vbl. 3, (February, 1932), 13.

39Daniel Kat: andPloyd H. Allport, Students' Attitudes, (New York:
Craftsman Press, 1931); cited by Bennett, M. E., College and Life,
(New Ybrk: McGraw Hill, 1933), 56.

 

 

36

In addition, the "Miscellaneous" column indicates that at least
twenty-seven descriptions of reasons proved so unique as to defy
.classification, compared to only four in the earlier studies. If the
-unique reasons which the investigators reported were less important,
had been_inc1uded in the total, the "miscellaneous" figure would be
considerably higher. In part it would appear that the increasing
number of items per study included many peripheral items not previously
considered central to the decision-making process.

In this connection, however, it is interesting to discover which
items, ranked one to five by the investigators, were.relegated to
“Miscellaneous." "The kind of college (teachers college, liberal arts,
engineering, etc.)," was ranked first in 1951 by Cleeton,40 whose sixth-
.ranking item, "type of college (i.e., coeducational, segregated
institutions vs. public institution, etc.)," graphically illustrates
the confusion possible when one attempts to classify such broadly- .
-Characterized items. "Kind of college (teachers college)" ranked third
by Hadaway in '51,41 "friendly attitude of faculty and students" ranked

third by Franks in '53,42 and "the friendliness of the student body,"

 

4oKenneth N. Cleeton, "An Analysis of the Factors'Which Influence
Students When They Choose A College," (unpublished Ph. D. Dissertation,
University of Colorado, 1951), xii.

41C. W. Hadaway, "Factors Influencing Students to Enroll at
Indiana State Teachers College," Teachers College Journal, Vol. 22,
(November, 1951), 23.

42Lawrence T. Franks, "A Study to Determine the Major’Factors
Which Influence Students to Enroll at Stephen A. Austin State College,"
(unpublished Ed.D. Dissertation, Indiana University, 1959), 58.

 

37

fourth by Schloemer in-F53. ,In addition, Schloemer's Michigan State
‘College subjects ranked "the basic college, making it possible to obtain
a liberal education in the first two years," fifth.43 Holland discovered
that true miscellaneous factors-comprised the second largest category
of answers for both males and females-ina'57.44 Olsen reported that
"heterogeneity of the student body" ranked third highest among elemen-
tary education students at the UniVersity of Illinois, fourth among
those at Eastern Illinois State College. In addition, "the opportunity
‘to live away from home" ranked fifth in the University of Illinois
sample.45 In 1960 Plante reported "to secure general knowledge, dis-
cover a suitable occupation" and "secure training necessary to enter a
professional school" second, third and fifth, respectively.46

Among people classified as to influence in the later studies,
parents remained in first place, although they did not make as strong

a showing as in the earlier studies. Friends and acquaintances

 

,___,_

43Clarence-L. Schloemer, "Why They Choose Your College," College

Public Relations W, Vol. 4, (July, 1953), 27.

44John]... Holland, "Student Explanations of College Choice and

Their Relation to College Popularity, College Productivity, and Sex
Differences," College‘ggg University, vol. 33, No. 3, (Spring, 1958),
314. ' °

“Hens c. Olsen, Jr., "Elementary Education: Students Reasons for
Selecting Their Colleges," Journa1.g§‘Teacher Education, Vol. 10,
(September, 1959), 316.

46Alexander J. Plante, "An.Analysis of the-Plans for Post-

Secondary Education of 1959 Secondary School Seniors in Connecticut and
Certain Factors Which Influence Their Plans," (unpublished Ph.D.
Dissertation, University of Connecticut, 1960), 55.

 

38

followed closely, and, with high school personnel which appeared third,
retained their earlier order of influence. College personnel, relatives
and alumni, investigated less often, appear to rank last and approxi-
mately equally. College personnel emerge as somewhat stronger in
influence than they appeared in the earlier studies.

Among the college factors, "reputation of the college" appeared by
far the most influential, ranking first or second in thirteen out of,
nineteen studies, and in the tap five, seventeen out of nineteen times.
"Reputation" was one area which was not deemed sufficiently important to
investigate before the war! Its' nearest competitor for first and
second votes, "course of study offered," showed up six times in twelve
studies. However, "vocational training available" did a repeat per-
formance over the earlier studies by once again coming up first in the
three studies which included it, studies of fourty-seven, sixteen and
five items reapectively. Combining these latter two categories, it
would appear-that students since the war at least, and possibly before,
were indicating a search for fairly specific academic goals in their
Selection of a college.

An item which.may be as influential as "course of study offered"
was asked by investigators considerably less. In six studies "size of
the college" ranked in the tap five items four times, and in the tap
first or second, three times. "Size.of the college," like "vocational
training available" in recent studies, may well be an item unduly

neglected by both earlier and later investigators. It is interesting

39

that, in the one-study which questioned high school students in this

'regard, size.ranked second in importance out of five items!47

Somewhat less impressive as an item was "cost of attending college,"
which got "honorable mention" by ranking in the tap five nine out of
twelve times; however, it polled only two first or second votes.

Indeed, cost would appear slightly less crucial than in earlier studies,
which may be a reflection of the increasing availability of financial
aid.

"Proximity-location" trailed still farther behind, in sixteen
studies ranking in the tap five only nine times, in the t0p one or two,
three times.

Of three studies which sampled "religious affiliation," two ranked
it fifth. "Publications of the college" ranked less often; "social
life," sampled four times, and "job placement after graduation,"
sampled twice, proved insignificant.

In conclusion, a comparison of the earlier and later studies in the
.influence of persons shows parents, friends and acquaintances, and high
school personnel retaining their positions of primary influence in that
order. ‘Parents appear to have declined in influence somewhat in the
later years, although this could be partly accounted for by the ins
“creased Opportunity for choice in-the later questionnaires.

Among college factors in the later studies, "reputation of the

college" is the surprising runaway in influence, despite its total

neglect before the war. This may well be a reflection of the increasing

 

“Robert s. McGranahan, "A Study of the 1949 Iowa High School
Seniors to Determine Their Future Plans After Graduation," (unpublished
M.A. Thesis, State University of Iowa, 1949),* cited by Mueller,

33. fl” 17-20.

 

 

 

competitivenesa.of the American college scene. "Proximity-location,"

one of the factors most frequently ranked high in the earlier studies,
dr0ps to approximately sixth place in relation to other variables.

This too may be a reflection of deve10pments in American society--
namely, improvements in transportation and increased mobility of indivi-
‘duals.

While the-early studies slighted college factors so as to make
comparison between them and personal influences difficult, it is clear
from the later studies that "reputation of the college" and "course of
study offered," at least, are ranked as influential more often than
parents. "Vocational training available" and "size of the college"
might well have been also, had the incidence of their occurrence in
sampling instruments been higher. Parental influence and "cost of
attending college" would appear to be about equal in strength as

determinants in the process of college Selection.
Summary

Twenty-four studies by individuals or institutions of higher educa-
tion attempting to assess influential factors in college choice, were
,reviewed. RepreSenting a period of 36 years, from 1926 to 1962, they
showed considerable variation in types of samples, purposes of the
research and methods of obtaining data. A high proportion of the
studies were weakened by carelessness in the use of statistics or
failure to prevalidate or describe the instrument used. In addition,
only four studies actually questioned high school students prior to

enrollment in college.

41

The emphasis of researchers before 1945 was-on persons as influ-
‘ential factors, and parents emerged the strongest of six categories of
peOple. When later studies increasingly included college factors in the
choice, "the reputation of the-college" became by far the single most
frequently ranked item.

,A thorough search failed to disclose any prior studies attempting
to differentiate between students on the basis of patterns in their

college selection criteria.

 

 

 

CHAPTER III

DESIGN AND METHODOLOGY
Description of the Sample

The sample for the study was drawn from a project originally conducted
by Coleman. In the following section Coleman's sample and the one used
for the purposes of this research are defined.

Colemgn's Original Sggplg

As was previously stated in Chapter I, the data for the study had
its origin in Coleman's analysis of high school climates entitled Th;
Adolescent Society. In that extensive work of research, Coleman and
his staff analyzed pretested questionnaires and other data from all of
the 10,000 students in ten northern Illinois high schools in the fall of
1957 and spring of 1958. Sections of these questionnaires were blocked
out for completion by college-bound seniors only. There were 1,149
seniors who’completed these sections in both the fall and spring
,questionnaires.

Whether or not generalizations can be made from this sample depends
upon the acceptance of Coleman's original method of selecting high
schools. In the introduction to Th; Adolescent Society, Coleman offers
the following justification:

The schools are located in northern Illinois, but are
not intended to be 'representative' of this section. To

the contrary, they were Selected with diversity in mind
. . . The results of this study, then, are intended to

 

43

apply. . .to all schools encompassed within the range of
community composition exhibited by these schools.1

In a later paper, Coleman gave the following description of the
communities*chosen:

1. Size 100-400: Two schools, serving prosperous
farming areas, and including students from a broad socio-
'economic range and from both town and country.

2. Size 400-1,000: Four schools in all. Three are
prosperous schools, including mostly town students of
diverse sociodeconomic background, but some farm children
as well. One of these has a tradition of sending students
on to good colleges; graduates of the others more often
attend nearby colleges and teachers' colleges. The fourth
school in this size group is a Catholic parochial school
serving working class students in Chicago. Deepite the
socio-economic-level, a relatively high prOportion of
students go on to college.

3. Size l,000—2,000: Four schools. Two are suburban,
and two are_in independent towns, 50,000-100,000 in size.
One of the towns is predominantly lower middle class; one
is predominantly upper middle-class. One of the suburban
schools is predominantly lower middle; one is upper class
and upper middle class.

The ____Fo11ow-2p. Samoa 122 219st.

The descriptions cited above were included in a prOposal to the
College Entrance Examination Board by Coleman and Rossi, of the National
Opinion Research Center. The prOposal, a request for funds to do a
follow-up study on these seniors, in an attempt to get at the "natural

3

history" of college selection, dwas accepted. After interviewing llO

 

.lJames 8. Coleman, Th; AdoleScent Society, (New York: Free Press
of Glencoe, 1961), ix.

2Coleman and Rossi, "A PrOposal for Research on the Processes of
College Choice," (unpublished prOposal submitted to the College
Entrance Examination Board, January 6, 1958), 5-6.

31bid., 1.

 

high school seniors to glean the "flavor" of college decision variables,

Coleman and Rossi constructed a questionnaire and administered it to 581
students of the original sample who were enrolled in colleges in the
spring of 1959. This new data, together with items from the fall and
Spring questionnaires of the original 1,149 college-bound high school
seniors, was punched onto IBM cards by the NORC staff. (NORC survey
412).

In the fall of 1962 and January, 1963, negotiations with
Dr. Peter Rossi, Director of the National Opinion Research Center, for
the use of this data for the present research were completed. Because
the IBM cards delivered constituted collapsed work decks, not all of the
information in the fall, 1957, spring, 1958 and Spring, 1959 question-
naires was available. Furthermore, not all of the decoding information
was available. A complete cOpy of the items received is included in
Appendix A.
THE Presegt figmplg

Because this research was primarily concerned with the reasons
students selected the colleges they would attend, it was necessary to
reduce Coleman's original and follow-up samples for those subjects whose
answers were incomplete. To be included in the present sample a student
had to have (1) completed all three questionnaires, (2) stated his
reasons for selecting the college he would attend, (3) been admitted to
the college he referred to when listing his reasons for choice, and
(4) been in attendance in that college at the time of the Spring college
questionnaire. These requirements reduced the number of students
elgible for inclusion in this study. The refinement process is reported

in Table 3.1.

1,149

801

-220

 

45

TABLE 3.1
THE REFINEMENT OF THE NORC FOLLOW-UP SAMPLE

high school seniors completed both the fall and
spring high school questionnaires.

who did not complete the spring college question-
naire.

college freshmen completed the spring.college
questionnaire.

who were not in the original high schools sampled,
and therefore did not fill out the high school
questionnaires.

who completed all the high school and college
questionnaires.

who had not selected a Specific college at the time
of the Spring high school questionnaire.

who did not answer the question in the spring high
school questionnaire pertaining to their reasons
for choice.

who were not enrolled in the same college they had
named in the high school questionnaire.

whose colleges were not specifically coded (i.e.,
they were coded only as "nursing schools,"
"business schools," or "other.")

who were enrolled in the same colleges which they
had stated reasons for selecting.

 

46

The resulting sample of 276 students was further divided into two
groups to provide for validation and cross-validation. This was
accomplished by sorting the 276 students' IBM.cards, in the sequence
order of the original code numbers, into ten piles corre3ponding to
the high schools they had graduated from. Code numbers were randomly
assigned to each pile of cards. Even-numbered cards constituted the
validation sample, odd-numbered cards the cross-validation group. In
Table 3.1 a comparison of the groups with Coleman's original pOpulation
‘may be found.

Inapection of this table reveals that the validation and cross-
‘validation samples are not distributed among the schools in a.manner
similar to the original p0pulation. A Chi-square Test of Independence
between Coleman's sample and the present sample of 276 resulted in a
chi-square.of 20.72 which was significant at the .001 level, thus re-
jecting the null hypothesis of no difference in the distributions of
the two grOUps. This was to be eXpected in view of the fact that the
students involved in the present research represented only the college-
bound group, whereas Coleman's study included the entire student papu-
lation. Indeed, the combined validation and cross-validation groups
bear out Coleman's statement that some schools send proportionately
‘more students to college than do others. This fact should not, however,
affect the generalizations to be made from this study. If the results
of Coleman's study are applicable to the high school papulation in
general, then-the results of this study should be-equally applicable
to the college-bound members of the pcpulation, at least in the Midwest,

since types of colleges available for choice--and hence reasons-~may

vary by region.

N . m "MAE (H.

 

 

 

 

 

47

me
on

ma

m

a
,ogmawm

we
mm

ON

w

m
mammmm.

sonoeeaae> soaoeeaee>

immouu

one.a
one.a
oee.a

omo.a

Amaomv one
eon
cos
eon
one

oma

coo.aa
ooo.oon
ooo.m~

ooo.e

Aowmofinov
oee.m
coo.e
ooo.a
ooo.e

ooo.a

 

Aoameam m.swEmaoUv cowumasmom

seam seesaw

huaosaaoo

mucwfiom o>auoooxm
.383:
mustang:
camasoz

aaeanoonemv.
swam e.etoe .um

cosmonnh scone
o>ouw oases
ssouaam
oaaa>uostz

samvauwm

 

Henson

ZOHH<ADmom Qz¢vmnmz<m A<ZHOHMO m.z<ZHAOU Qz<
mum<Mmum mHmH zH numb mmfithm mmH mo muoomum mUHm_wm ZOHBDmHMHmHQ NUZMdemm NIH

~.n_mom<e

system
sues Human
.33 Seem

custom

kuao swung

"pesos Hausa

Nuwsnaaoo
mo.oo%H

 

48

Design of the Study

The design of the research included four stages, two in the vali-
dation process and two in cross-validation. The first was a multiple
agreement analysis of the reasons indicated for selecting their
colleges by 138 students randomly selected from the total sample of
276. This was employed prflmarily to classify groups of students with
similar reason-reSponse patterns. (I.e., re5ponse patterns A, B, C,
or D, where Type A's order of reasons might be costs, closeness to
home, course of study offered, and chances of getting admitted.)

The second stage involved identification of these groups according
to similarities of re3ponses to the 153 available items in the three
questionnaires. These items included information usually available to
the college admissions counselor (i.e., sex of student, educational
level of parents, socio—economic class, I.Q. quartile of student, type
of high school attended, study habits, social activities, method of
financing college education, and the role of significant others in the
collegevdecision.)

The third stage was a cross-validation of the first stage, again
'employing multiple agreement analysis to classify the remaining 138
students into groups of students with similar reasons for selecting
their colleges. It was intended that, if the groups resulting from
this cross-validation reasonably approximated the validation groups,
the research hypothesis that there are patterns of agreement in college
selection criteria would be accepted, and the fourth stage would then

be completed. Because of the exPloratory character of this study, the

 

49

nature Qof the data itself, derived from the ensuing investigation,
determined what would constitute a reasonable approximation.

The fourth stage was a repetitionof the second stage, an attempt
to discover the distinguishing characteristics of each group, using
only those questionnaire items which had significantly differentiated
in the validation stage. Stage four involved a testing of hypotheses

deve10ped in stage two.
Methodology

Stages 9% and _Tl_1_r_e_e:

Multiple agreement analysis is a method deve10ped by Hemingway "for
the objective classification of objects on the basis of their common
characteristics or patterns."4 It was largely derived from McQuitty's
1956 article an agreement analysis and was influenced by later publi-
cation by McQuitty.5 The advantage of this method over other methods
,of pattern analysis is that it is designed to classify subjects into
thevminimum number of classes while using the maximum amount of infor-
mation. (Other methods yield all possible patterns.)

Multiple agreement analysis accomplishes its purpose by supplying

a strict criterion for the termination of classes based on the

 

l"Peter W. Hemingway, "Multiple Agreement Analysis," (unpublished
Ph.D. Dissertation, Michigan State University, 1961), 1.

5Louis L. ‘McQuitty, "Agreement Analysis: Classifying Persons
By Predominant Reaponses," British Journal 9; Statistical Psychology,
Vol. 19, No. 1, (May, 1956); and Hemingway, Ibid., 103.

"vs-g;

 

 

W1

is .
the

for

re
K-l

for

ad

afi
as:
1131]
we-
pr
th
of

c1

50

maximation of information contained in each class, where "information"
is defined as the product of the class size (number of members) times
the pattern size (number of common reasons). This procedure, developed
for cdmputer use, is basically an iterative process of reductive matrix
partitioning. "Beginning with a matrix of n persons recorded as either
possessing or not possessing each of r characteristics, successive sub-
matrices are extracted. These sub-matrices are of maximum product size,
each having identical rows (reasons) for all class subjects."6 In
Appendix B of this report the basic steps of multiple agreement analysis
-are illustrated, using a hypothetical example. (A MISTIC program,
K-lOdM, is available at the Michigan State University Computer Center
for this technique.)

In.Hemingway's original work on the 83rd U.S. Senate, he advocated
a.doub1e entry method of IBM card punching for computer analysis of his
program. That is, the negative votes were punched together with the
affirmative ones. A similar approach was used in this project. The
assumption-was made that the reasons a student did not check were as
important as the ones he did check. Therefore, the reasons omitted
were punched into the IBM cards following the reasons selected. This
proceedure yielded a matrix of l6-re8ponses by 138 subjects, rather
than just the 8 stated reasons by 138 subjects. Only the major classes
of reason-response patterns were analyzed in this project. (A major

class consisted of 15 or more members.) While this approach reduces

 

-6Peter'W. Hemingway, "Multiple.Agreement.Analysis," (unpublished
Ph.D. Dissertation, Michigan State University, 1961), i-ii.

 

51

the size of the sample, generalizations from categories containing
relatively few individuals would be.of little-practical value.
Question number 174 of Coleman's original spring high school
questionnaire was used as the control item for the multiple agreement
analysis. The question reads as follows:
What were the most important consideratiOns you
had in mind when you were detiding which school to

attend? (Check all the things that were important
for'you.)

1. Tuition costs and living expenses

.2. Closeness to home

-3. The high educational standards

4. Course of study offered

5. Chances of a scholarship

6. The religious character of the school

7. Your chances of getting admitted

8. PeOple you know who went to those schools.
Stage Two:

The second stage of the design concerned the identification of the

distinguishable characteristics of the groups resulting from the initial
multiple agreement analysis in stage one. These characteristics were

derived from the.responses to the 153 items available.from the three
aquestionnaires. The statistical treatment of the-data involved testing
'whether any one group differed significantly from the other groups_in
its responses to each of the questionnaire items, or "A" versus "not A"
testing. Because multiple agreement analysis yields nominal sealing of
the.reason-response groups, the-chi-square.test for two independent

samples was an appropriatestatistic.7

 

Sciences, (New York: McGraw Hill, 1956), 175-179.

 

52

A table was constructed for each item showing the number of persons
in each group that responded to each alternative of the item. Each
table was inspected to determine whether the responses of any one group
appeared to be distributed differently from the responses of the remain-
ing groups. Observable differences were tested with the chi-square
statistic.

The null hypothesis was that the group in question did not differ
from the remaining groups with respect to the item under~investigation,
or, statistically stated, that the group in question belonged in the
same pOpulation as the remaining groups. If this null hypothesis was
rejected, then the item content was used as an identifying characteris-
tic of that group. Because of the exploratory nature of this research,
the rejection region for stage two was set at the..20 level of signifi-
cance. In the validation stage, then, the chance of being in error
‘twenty percent of the time was taken in accepting the content of an
item as distinguishing one group from the rest. (This error was par-
tially overcome by using a more stringient .10 alpha level in the cross-
validation.)

An illustration of this inspection and chi-square method follows.

Question number 53 of the fall high school questionnaire asked:

-- Where were you born? ‘

7. .In this town or county (the county of the
high school)

8. Outside this county but in Illinois

9. Outside Illinois but in the U.S.
.10. Outside the U.S.8

 

8Coleman and R0881,-22,‘cit., NORC Survey 412.

 

53

Assume that four groups resulted from the initial multiple agreement
analysis (A, B, C and D), plus a group whose reasons for college
selection were individually unique and could not be classified (N).

The responses for each group were recorded in a table similar to

 

 

Table 3.3.
TABLE. 3.3
RESPONSES ‘TO ITEM 53 or m FALL QUESTIONNAIRE
Not Avail-
Groups 7 8 .9 10 able Tbtal
A ,5 7 3 w 1 16
B 15 7 3 25
c 13 7 _ 3 1 24
o 19 11 3 7 1 34
N '23 13 3 39
' Totals 75 45 15 3 138

The resulting tables for each item were inspected to see whether
the responses on any group(s) appeared.to be different from the responses
'of the remaining groups. This was accomplished by partitioning the
response tables into dichotomous categories on some logical basis
according to the content of the item. ,Ilble 3.4 is illustrative of
this partitioning of the responses of students to-item 53 of the fall

high school questionnaire.

 

54

TABLE 3.4

DIRECTION‘OF‘RESPONSES'TO‘ITEM'53'OF'THE"FA11-QUESTIONNAIRE

 

7 n 1 s, 9, 10
A 5 < 10
s 15 > 10
c 13 > 10
n 19 > 14
‘N '23— :p 16

 

 

It can be seen in the above table that the frequency of responses
of group A were in an opposite direction from the responses of groups
B,:¢, D, and N. That is, most of the students in group A chose-answers
8, 9, or 10, while the majority of persons in the other groups chose
answer‘7.

For all iteerwhere‘a similar reversal in the direction of response
frequency appeared, the chi-square test was employed to determine
whether there existed a better than chance relationship between the
*group and its-unique-response. ln-all cases, these were 2 x 2 chi-
square-tables with one degree of freedom, where the group,in question
-was partitioned against those students who were not in that group, and
the group?S'predominant responses were dichotomized against the remain-
ing responses. Table 3.4 is presented to illustrate the chi-square

~divisions used in the analysis of item F-53.

 

 

 

gs;

 

 

55

TABLE 3.5

CHI-SQUARE TABLE FOR ITEM 53 OF THE FAlloQUESTIONNAIRE

 

 

 

 

 

 

 

 

 

A Not A Totals
F0 5 70‘ 75
7 , Chi-Square
2.438*
Fe 8.33 66.67
F0 10 50 6O
8.9.10 p (.20

Fe 6.67 53.33

Totals: 15 120 135

* Yates Correction Applied

The chi-square of 2.438 in the above table was significant at the
.20 level. This would indicate that a.factor which might distinguish
Type A students from checotherstlstthatemoretbfuthemswbrenbonnsdutside
the town or county of the school district., Further interpretation of
the unique character of the Type A group would depend on which other
items also proved significant as differentiators, but the fact that
TypeuA students came from more mobile families would be part of that
interpretation.

Similar tables and chi-square statistics were computed for every
item where a difference in direction or quantity of response was
indicated by inspection of the data. These tables appear in Appendix
A. Only those items whose chi-square values were significant after

cross-validation are included in the body of the report.

 

56

The fourth stage followed the cross-validation multiple agreement
analysis (stage three), using the remaining 138 students not employed
in the validation stages. Those items which were significant at the
.20 level in stage two were subjected to chi-square analysis in stage
four, with the cross-validation level of significance set at .10. The
content of the surviving items was used to-describe characteristics
of the groups to which the items applied.

Reliability Estimates

 

In order to obtain an estimate of the reliability of the items
used in the project, two reliability coefficients were computed using
Hoyt's Analysis of Variance technique.9 Using two different samples
of 50 students each, reliability checks were made on the eight reasons
for college selection used by Coleman in item 174 of the spring high
school questionnaire, and on the 16 items which held up after cross-

validation and were used to identify the differences between the

reason-response groups. The students' responses used in this relia-
bility analysis were randomly selected from the total validation and
cross-validation samples, stratified according to their respective

reason-response group.

 

90. J. Hoyt, "Test Reliabilitvastimated by Analysis of variance,"
Psychometrika, vol. 6, (1941), 153-160.

,.

'_'.-‘

 

 

57

Summary

The sample employed in this research was part of a larger sample
of high school students used by Coleman and Rossi in their preposed
study on the "natural history" of college choice. (NORC Survey 412.)
All high school seniors in the NORC sample who had selected a Specific
college by the Spring of their senior year, had indicated their reasons
for selecting that college, and who were enrolled in that college the
following Spring, constituted the sample of 276 used in this project.
By a stratified random selection of students from each high school, the
sample was further divided into two equal groups of 138 students each
for purposes of validation and cross~validation.

The design of the research included four stages. Stage one
employed multiple agreement analysis of the re3ponses of 138 students
in order to classify these students into groups with similar reasons
for selecting their colleges. Stage two was concerned with identifying
the distinguishing characteristics of each group on the basis of avail-
able items from the three questionnaires compiled by NORC Survey 412.
Tables were~constructed for each group's reSponses to each item, and
these tables were then inSpected to determine the apparent direction of
unique reSponses of any one group. The chi-square statistic with an
alpha level of .20 was used to determine whether these directions were
significantly different from those of the remaining sample.

Stage three again employed multiple agreement analysis, using the
second sample of 138 students to cross~validate the groups obtained by
the first sample. Stage four similarly cross-validated the results of

stage two, and the distinguishing characteristics of each group were

 

 

58

identified. However, in this stage the rejection region was set at
.10 and the chi-square analysis was employed only with those items
found to be significant with the validation sample.

Reliability estimates for the eight reasons for college selection
and the items which-remained Significant after cross-validation were

aISO‘computed.

 

 

 

~ " “3'57..."

 

CHAPTER IV

ANALYSIS AND INTERPRETATION

As stated in the previous chapter 'on design, the analysis pro-

’ceeded through four stages.
Stage One: Multiple Agreement Analysis of the Validation Sample

Most of the difficulty encountered in the analysis of the data
occurred in the first stage of the design. Because multiple agreement
analysis is a complicated process of matrix reductions, an electronic
computer was necessary in the initial phase. While Michigan State
University has four varieties of computers, only one had been program-
med to use Hemingway's multiple agreement analysis: the MISTIC-computer
on which Hemingway's original work was done. However, at the time of
this analysis Michigan State was in the process of installing a newer,
larger computer and was phasing out MISTIC. For this reason, the
programs that MISTIC was unable to handle, due to internal equipment
breakdowns, were not being rewritten to circumvent the electronic
difficulties, nor was the computer itself being repaired. Multiple
agreement analysis was one of these programs.

Hemingway's original analysis, of the voting patterns.of the 83rd
U.S. Senate, took twentyieight hours of computer time.1 When the-data

for this project was ready for MISTIC, the programmers could not

 

1Peter W. Hemingway, "Multiple Agreement Analysis," (unpublished
Ph.D. Dissertation, Michigan State University, 1961), 69.

 

6O

guarantee valid results if the computer was required to run for more
than two hours. An attempted analysis with a matrix of 138 students

‘by 8 reasons supported this statement; theresults were checked manually
and found to contain numerous errors.

Three possible alternative solutions were considered: (1) to
program the data for a different computer; (2) to employ another method
of pattern analysis; or (3) to change the size of the input matrix.

The first solution was not feasible, partly because no other available
computer could handle this program with a sample size larger than '40,
and partly because no computer authority could be found at Michigan
State who was. familiar enough with multiple agreement analysis, or
could supply the time necessary to assist in the writing of the program.

The second alternative also proved fruitless. Several statistical
techniques were attempted with the-data, including Guttman's Scalogram
Analysis,2 Coomb's Parrallelogram Analysis,3 Lingoe's Multiple Scalogram
Analysis,4 McQuitty's Typal Analysis,"5 and McQuitty's Rank Order Typal

Analysis.6 Either similar computer problems were encountered, or

 

28amuel A. Stouffer e_t. 31., Studies 1.3 Social Psychology, Volume
5, Measurement and Prediction, (Princeton, New Jersey: Princeton
University Press, 1950).

‘3Clyde'H. Coombs, "The Theory of Data," (unpublished manuscript,
University of Michigan, 1962).

4.1 ames C. Lingoes, "Multiple Scalogram Analysis: A Generalization
of Guttman's Scale Analysis," (unpublished Ph.D. Dissertation, Michigan
State University, 1960).

5Louis I... McQuitty, "Typal Analysis," Educagional 33d Psychological
Measurgent, Vol. 21, (1961), 1677-696-

6McQuitty, "Rank Order Typal Analysis," Educational and Psycho-
logical Measgement, Vol. .23, No. l, (1963), 55-61.

hm.

 

 

 

the

COKE

thi
stt
in

a c
For
off
alo
gro
did

onl

ana

fro

to
no

sta'

one:

cell

for

equaj

rathe

 

61

the analysis did not divide the students into groups according to their
commonality of reasons for selecting colleges.

Upon investigation, it was concluded that the primary reason for
this lack of results was the basic nature of the data matrix. When 138
students have only 8 reasons to choose from, there will be much overlap
in the reasons selected. Thus, pure types, where some students selected
a combination of reasons that no others chose, were-virtually impossible.
For example, all but 29 students indicated that the course of study
offered by the college was influential in their choice. This one item
alone made it impossible to separate 109 of the 138 students into pure
groups. Because other reasons indicated by the remaining 29 students
did coincide with some of theirs, no pattern analytic method yielding
only pure groups could be employed.

Therefore, it was decided to continue to try multiple agreement
analysis until a matrix could be found which would provide valid results
from the MISTIC program, and to reduce and alter the shape of the matrix
to accomplish this objective. Hemingway, in his analysis of "yes" and
"no" votes to determine voting patterns, had concluded that he was
statistically more successful when including both "yes" and "no" votes
in the original matrix, than he was when simply tabulating affirmative
ones. While it was recognized that failure to check a reason for
college choice was not quite the exact antithesis of checking it, in
the same way that a "no" vote iS'CO a "yes," it was decided to assume,
for the purpose of the analysis, that both types of reaponses were
equally relevant. This approach yielded a matrix of sixteen items,

rather than the original eight.

 

~3Eis;EiIEIIIIIIIIIIIIIIIII!___e
I’SEEZEIIEIIIIII!!~

 

EH

be

the

col‘
mu11
(723
c011
in t
81‘0L'
sele

agre

more
If 0:
not a

Cauti.

 

62

The problem of reducing the samples to be submitted to MISTIC
remained. It was decided that the patterns of reSponses would be com-
piled manually, and that only those patterns including two or more
individuals would be programmed. This compilation yielded 71 different
patterns, 31 of which contained two or more individuals and accounted
for 98 of the total 138 students. After the analysis of a matrix of
these.31 patterns by 16 items was completed, the remaining.40 students
were placed in their appropriate groups manually. To qualify for
entrance into a group, a student's reasons for college selection had to
be absolutely identical with those of the group.

In Table 4.1 the findings of the multiple agreement analysis, and
the preportion of students placed in the resulting groups,are summarized.

Every person in groups A through D "agreed" on-every item in the
column under the appr0priate letter designation. It appears from the
multiple agreement, then, that almost three-fourths of the students
(72%) could be classified into groups according to their reasons for
college choice. The members of each group were more like each other
in their college selection.motives than like any person outside their
group. About one-fourth of the students, (28%) however, apparently
selected their schools for their own unique reasons and were not in
agreement with any one group.

Inspection of Table 4.1 reveals that several of the reasons typify
more than one group. Thus, the groups are not unique in the sense that
if-one type chose variable(s) X, Y, and/or Z, those variables could
not appear as classification criteria for any other type. However,

caution must be exercised in assuming that those reasons which Span

9.
11.
12.
13.
14.
15.

16.

TABLE 4.1

63

 

MULTIPLE AGREEMENT ANALYSIS OF THE VALIDATION SAMPLE'S
REASONS FOR COLLEGE CHOICE

Reasons

Tuition and Living Costs

Closeness to Home

High Educational Standards

Course of Study Offered

Chances of a Scholarship

Religious Character of the
School

Chances of Getting Admitted

Pe0ple you knew who went to
those schools

NOT tuition and costs

NOT closeness to home

NOT high educational
standards

NOT Course of Study offered

NOT chances of a Scholarship

NOT Religious Character of
the School

NOT Chances of Getting
Admitted

NOT pe0ple you knew who went
to those schools

Groups

Total Number of Persons per Group

Percent of 138 per Group

*Unclassified

Note:

Groups 2; Students
with Similar Reasons

 

 

A B c D N*
x x
x
x x x
x
x
x
x
x x x x
x x x
x
x
16 25 24 34 39
12% 18% 17% 25% 28%

 

An X indicates total agreement.

 

64

more than.one group can be interpreted as significant omissions for the
groups which did not include them. The antitheses in rows 9-6 were only
considered classifiable variables for a group if all the included
individuals never chose the alternative. For example, "course of
study offered" (item 4) was a reason checked by all students in groups
A, B, and D, but not by all the members of C. However, the antithesis
of that item, "NOT course of study offered" (item 12), does not appear
in the index of Type C reasons. Consider "tuition and living costs"
(item 1). All but Type A students checked this reason. Its Opposite,
"NOT tuition and living costs" (item 9) was indicated by all students
in the Type A group. Therefore, while the course.of study offered is
not an identifying reason for Type C, the fact that Type A is not
considering tuition and living costs $3 a unique characteristic of

that group.

As a result of the first validation analysis, the following obser-
'vations are offered regarding the patterns of reasons which differen-
-tiated among students:

.Type A students were-characterized by selecting their

colleges for their high educational standards and the

course of study which they offered. These students were

apparently not concerned with the expense involved, the

proximity of the college to their home, their chances of

receiving a scholarship, or the religious character of

the school. In brief, they appeared to be interested in

obtaining the best education available in their field of

study regardless of the cost, location or denomination

of the college.

Type B students were also interested in the school's

ability to offer them a particular course of study, but

they wanted it at a price they could afford. Other

peeple who went to the schools they chose were appar-

‘ently of no consequential influence, nor were the

denominations of their colleges or their chances of
getting admitted. These students wanted an education

 

 

 

65

in their chosen field within a certain price range, and
would take their chances of getting into those schools
which met their needs.

Type C students, too, were concerned with finances,
but appeared to be less willing to gamble on their
chances of admission. While they were.not concerned
with scholarship possibilities or religious association,
they did apply to colleges which they thought would
accept them.

Type D students were the only ones uniquely concerned
with the religious denomination of the schools they would
attend. They agreed with other groups, however, in
‘wanting their schools to provide them with an education
in their field of endeavor, and were not concerned with
scholarship possibilities.

Type N is not a homogenous classification in the same
sense that the other types are. It is merely a grouping
of all those students whose reasons for selecting their

colleges did not agree with any of the foregoing
:classifications.

Stage Two: Identifying Characteristics of the Validating.Groups

The second stage in the design involved the identification of the
distinguishing characteristics of each of the reason-reSponse types.
To that end, responses on the remaining available items from the
three questionnaires were collapsed to dichotomous classifications,
making it possible to use 2 X 2 chi-square tables with one degree of
freedom.

The chi-square tables didhotomized the reopenses of one group
(i.e., "Type A") against the responses of subjects not in that group
(i.e., "Not Type A.") This partitioning.was then compared to similar
two-fold divisions of the item under analysis (i.e., reSponses 1-3

versuS‘reSponses-4-6.)

 

66

Of the 153 questionnaire items, only 48 proved significant at the
..20 level, in isolating the responses of any one group from the total
validation sample. Because with alpha set at .20 it could ordinarily
be expected that 31 items would be significant due to chance factors
alone, the 48 items found represented 17 more than chance variations.

It would be both cumbersome and Optimistic to ask the reader to
follow the-chi-square analyses of the-48 items. For those interested,
the complete proceedure is available in Appendix A.

Significant Differences Among Reason-Response Types
Within the Validation Sample

Only the results of those items which proved significant in the
validation chi-square analysis will be reported in this.section.
Egg Contegt Associated 113; Em A Students

Clearly the most recognizable were the sixteen students with Type A
reasons for cellege choice--those who selected their colleges because
of high educational standards, apparently indifferent to the expense
involved or the proximity to their home. Their reaponses-to 18 of the
153 items on the questionnaires were significantly different at the .20
level; the item analysis is reported in Table 4.2. On each item a
significantly greater preportion of Type A students responded in the
adirection indicated than did students in groups B, C, D'or N.
1593 Content Associated gig Tm _B_ Students

Students with Type B reasons for selecting their colleges were the
vleast-clearly distinguishable. ‘Perhaps this finding can be accounted
for by the fact that the only "reasons" unique to them were negative

ones, indicating a lack of reaponse, rather than any positive

67

TABLE 4.2

SIGNIFICANT DIFFERENCES IN THE RESPONSES OF TYPE A STUDENTS*

Itan Nunber Item Content Chi-Square Vglue
Fall Items
Column 1 Graduated from a suburban high school 8.977
53 Born outside the county of their high 2.438
school
56 Father was a college graduate ~6.000
57 Mother had some college education 3.380
64 Is not Catholic 2.450
64 Is probably Jewish 7.541
65 Attends church less than every week 3.440
during high school
Spring Items
106 Likes'himself the way he is-dwould not 2.386
want to change and be someone-else
l36b Applied t0'3 orMmore colleges ‘8.055
Column 77 Is in the tap I.Q. level 4.680
Column 78 Father has a white collar occupation 15.223
and student is in tap I.Q. quartile
College Items
7 Prefers an ivy league or eastern women's 23.031
college of high prestige
32 Friends were very pleased with his choice 17.138
of a college
43 Felt that his high school prepared him -3.111
for college better than most students
'52 Planned to go on to graduate school 2.929
after college
129 Prefers a job "which pays an extremely 4.526
‘good income, if you make the grade,
but in which you lose almost-every-
thing if you don't make it."
146 Enjoys classical music more than other ‘9.902
types of music
167 Soziorwmore of their best friends are -3.169

at the college they attend

*Chi-square at alpha-of..20-equals 1.642

 

 

HI

 

68

characteristics. They were not concerned with their chances of getting
admitted, nor were they influenced by students or alumni of the college
they selected. In Table 4.3 items which served to isolate group B are

summarized.

TABLE 4.3

SIGNIFICANT DIFFERENCES IN THE RESPONSES OF TYPE B STUDENTS

 

 

 

Item Number Item Content Chi-Square Value
Code Column 1 Graduated from small city high schools 3.138
Spring 174 See themselves as not being close to the 5.958
center of activities in high school
C- 32 Their closest friends were less than 6.231

"very pleased" with their choices
of colleges

C-142 They do not see themselves as part of 4.076
the leading crowd in college

C-l48 They attended one half or more of the 3.191
home basketball games in college

0-149 They look forward to college primarily 4.561
for the social life and activities
there

C-173 They find it difficult to find someone 4.400

to discuss a personal problem with

It Content Associated with Type E Students

 

It will be recalled from Table 4.1 that type C students, while
agreeing with other groups on three reasons for selecting their
colleges, were the only group whose members were all concerned with
their chances of getting admitted to the college of their choice.
The distinguishing characteristics of group C in the validation

sample are reported in Table 4.4.

"I

69

TABLE 4.4

SIGNIFICANT DIFFERENCES IN THE RESPONSES OF TYPE C STUDENTS

 

Item Number Item Content _Chl-SgLare V_a_1ue
‘Sb129 . Have read less than all of the-college 5.379

catalogues for the colleges they
were interested in

S-135 Family pays for less than half of 1.823
their yearly college expenses

S-136‘ They pay for more than half of their ‘3.208

' own college expenses

S-l43 Carried a part time job during high 3.352
School

C- 7 Prefetho attend teachers' colleges 2.439
or colleges in nearby Illinois areas

C- 37 They were personally less than "very 2.484
pleased" with their choice of a
.college

C--44 Found the hardest thing to get used to 4.244
in college was being away from home

0-108 Family paid for less than half of their 1.739
second year in college

C-110 Did not enjoy college more than high 5.752
school

£1232 Content Associated 1it_h .'Typg _12 Students

These were the only students who stated the religious character
of the-college was an important factor. In Table 4.5 is listed the
items-which significantly differentiated‘ D types from all other types.
IL”?! Coptept Associated 11:21. Tm N Stpgpnts

These 39 students comprised the largest single-group in-the
validation sample. Representing:28% of the total, these-were.the
students-whose reasons-for college selection could not be classified
into one of the other four existing groups. They apparently selected
their colleges for reasons unique to themselves and, while indicating
many similarities to the other reason-reaponse types, were.not-wholly

in agreement with any of them.

 

 

 

IH

gui
How
stu
hav
30m.
Stu:

tut;

 

70

TABLE-4.5

SIGNIFICANT DIFFERENCES IN THE RESPONSES OF TYPE D STUDENTS

 

Item Number Item Content §_l'1_i-S guare Value
Fall 64 They are probably not Protestants 3.107
129 Prefer to date star athletes rather 5.586

than the best scholars or best
looking students

.131 Prefer to be remembered as a leader 6.553
in student activities
Spring l36b Applied to 3 or*more colleges 3.358
Col. 71 Graduated from small town high schools 2.398
College 7 Prefer to attend a college near their 6.652
home

118 They rank "living up to religious beliefs" 29.589
as the most important ideal to strive

for
130 Attend church every week during college 5.314
131 They find themselves "daydreaming often" 9.234

inrcollege

It may, therefore, be confusing to attempt to identify the distin-
*guishing characteristics of these students as a homogeneous group.
However, this attempt was made in the desire to determine whether
students with.individually different reasons for college selection
have common characteristics. As might be-expected, the attempt proved
somewhat fruitless. The content of the items found to identify these
students is reported in the following table, merely because it consti-

tuted part of the total analysis.

 

71

TABLE 4.6

SIGNIFICANT DIFFERENCES IN THE RESPONSES
OF THE UNCLASSIFIED STUDENTS, "TYPE" N

 

 

 

Item Number Item Content ,gpysflre Value
Cod-e Col. 5 Includes more boys than girls 8.402
Fall 114 Prefers to attend a large university 3.272

129 Prefers to date "star athlete" or ‘2.877

"tap student" as Opposed to the
"best looking"

Spring 127 Tend to select-college majors in 4.739
Specific vocational fields
139 Feels that the general reputation 13.058

of the college he selected is
better than the others he-was

considering
College 108 Family pays for less than half of 2.049
second year in college
160 More likely to live at home during 13.870
-college

In reading the preceding tables it must be remembered that they
include items which proved significant for the validation sample
alone. (Therefore, subjective impreSSions about the students in these
groups are unwarranted. A more reliable estimate of the characteristics

‘of-these-groupszcan be found in the discussion of the cross-validation

sample.

Item Content Which Did Not Discriminate-Between
Reason-ReSponse Types
Although the primary purpose of the project was to identify the
differences between students who chose their colleges for different
-combinations:of reasons, the items-which did not separate the groups
are also interesting. The remaining pool constitutes 105 of the 153

items; that is, for 60% of the items all students had approximately

-. " me

 

 

8]

fr

 

78

the same responses. These items are of value because they serve as
indicators of areas in which college-bound students tend to agree.

Only the content of these tiems will be reported in this section;
complete tables showing the distribution of responses for each are
included in Appendix A. Similarly, percentage figures were not
included, since the decision on where to-divide each table was made
relative to the purposes of this research. These may not be the most
appropriate divisions for generalizations on college-going trends;
readers interested in researching such trends are invited to make
their own-divisions in the tables in Appendix A.

Table 4.7, then, merely indicates how the majority of students
responded to 105 items. Although a complete discussion of the content
of Table 4.7 is not within the scope of this report, some of the more
obvious conclusions follow. By the spring of their freshman year of
college, most students report adjusting rather well to their new
environment. They state they are pleased with their college selection,
find college life exciting, like the people they meet there, and,
except for a decrease in church attendance and an increase in the
amount of homework done, they do not change their attitudes appreciably

from what they were during their last year in high school.

 

 

 

 

 

 

#

#
#

# 117-20

#

Spring #‘52-6

 

62
64

65
113

129

(and College

124-8)
Spring

# 106
# 112
# 125
#‘129
# 130
# 131

# 132

 

73

TABLE 4.7

ITEM CONTENT ASSOCIATED WITH THE MAJORITY OF
STUDENTS IN THE VALIDATION SAMPLE (N-l38)

Item Content (Most College Bound Students)

 

 

 

Prefer to attend a high school the same size as the
one they now attend

Spend less than three hours per day on high school
homework

Were born in the same county where their high school
is located

Have fathers who did not complete college although
half of them did graduate from high school

Parents are not divorced

Half of the students are Protestant, 28% are Catholic
and 22% are Jewish

Attend church every week during high school

Prefer to go to a college "away from home"

See the "preparation for making a living" as the most
important aspect of a college education

Prefer to-date the "best looking" student of the
Opposite sex rather than the "star athlete" (or
cheerleader) or "best student"

Rank the occupation-of "business executive" as more
desirable than "writer" or "scientist”

Find that there are many things they would like to
change about themselves, but not completely

Report that their parents would be very proud of
them if they had been asked to be a biology
laboratory assistant

Prefer to attend a large college or university

Have seen the catalogues of all the colleges they
arerconsidering

Have visited at least half of the-campuses of the
-colleges they are considering

Rank the persons they talked to about-college in the
following order (of number of students checking
.each person) (1) Parents, (2) friends in high
sch001, (3) guidance counselor, (4) high school
teachers

Parents agree with their children on the course of
study to be undertaken in college, which college
to attend, whether to live at home or not, and
how the expenses-will be met

 

74

TABLE 4.7, Cont.

ITEM CONTENT ASSOCIATED WITH THE MAJORITY OF

Item Number

Spring # 135(&
College 108)

Spring # 136(&
College # 109)

# 13Gb
# 136C

# 136d

#.138
# 139

# 142

#‘l74

Col. 77
C01. 78
Col. 65

Col. 71
College # 32-6

STUDENTS IN THE VALIDATION SAMPLE (N=138)

Item Content (Most College Bound Students)

 

 

Family will pay for at least 75% of their childrenk
college education

Students will pay for at least 25% of their college
education

Apply to fewer than 3 colleges

Are admitted to the college of their first or second
choice

Do not apply for scholarships

Attend the college of their first choice

In considering the college they selected in com-
parison with the others they considered, agreed
that the college they selected (I) offered a
better course of study in the field they desired,
(2) was about equal to the others considered in
its general reputation, (3) had no Opinion of
the relative quality of athletic teams among the
schools

Had the following rank order of reasons for selecting
their colleges:

Number checking each
reaponse out of 138

students

1. The course of study offered 109
2. Tuition and living costs 91
3. High educational standards 89
4. Chances of getting admitted 56
'5. Closeness to home 52
6. Peeple they know who went to 41

those schools
7. Religious character of the 33

school

8. Chances of a scholarship 22

See themselves as being in center one third of the
activities in high school

Are in the bottom three I.Q. quartiles

Fathers are in white collar occupations

Students who applied for scholarship, received offers
from the college of their first choice

Graduate from "large" high schools

Mothers, closest friends, teachers and fathers were
very pleased with the college the student decided
to attend

 

 

 

-* Pl

 

Item Number

College # 37
# 44-8

# 50

# 52
#'53

# 68
# 72

#110
#111

#114
_#117
#118
#122
#123
#129

#130
#131-2

 

75

TABLE 4.7, Cont.

ITEM CONTENT ASSOCIATED WITH THE MAJORITY OF

STUDENTS IN THE VALIDATION SAMPLE (N=138)

Item Content (Most College Bound Students)

Were very pleased with the college they were attending

(1) Agree that "college is pretty much the same as
high school except that the subjects are different
and the courses are harder." (2) Disagree that "the
hard thing to get used to in college is being away
from home." (3) Agree that they "found the first
year to be very exciting because the instructors
Opened whole new worlds of ideas and knowledge."
(4) Disagree that in high school they had a lot of
friends but were not able to make many friends in
'college. (5) Agree that college work was much more
difficult and that they had to work much harder than in
high school

Had a particular occupation or career that they wanted
to prepare for in college

Are planning to graduate from college

Do not plan to dr0p out of college they are now attend-
ing

Are not receiving any scholarship aid

Have not worked at a paying job during their freshman
year

Enjoy college more than high school

Prefer a college the same size as the one they are
now attending

Have found courses in college that they would like
to "follow up in later life"

Spend at least 3 hours per day in homework during
the first year in college

Ranked "learning as much as possible in school" as
the primary "thing to strive for in college

Date at least twice a week in college

Do not "go steady" with a member of Opposite sex

Would select "a job which pays a moderate income but
whiCh you are secure of keeping."

Attend church less than every week during college.

Agree that (l) the real qualities of a person come
out in a group. (2) Would like to be someone
different than themselves. (3) Enjoy social
gatherings just to be with peeple. (4) "There are
a few who control things in this school, and the
rest of us are out in the cold." (5) They are
not doing so well at school. (6) "If a fellow
wants to be a part of the leading crowd around here,
he sometimes has to go against his principles."

 

15:

TABLE 407, Cont.

ITEM CONTENT ASSOCIATED WITH THE MAJORITY OF
STUDENTS IN THE VALIDATION SAMPLE (N-138)

Lgem Number

College #142
#146
#147
#148
#157-9
#160
#167-72

#173-4

Itgg Cogtent sugst College Bound Students)

Do not see themselves as part of the leading crowd

Enjoys other than classical music

Go to the movies less than once a week

Attend less than half of the home basketball games

Find the other students and college teachers friendly

Live in a university dormitory

Have the majority of their current friends at the
[college they, themselves, attend

Have no trouble finding someone to discuss a class
assignment with. Usually this person is a student
in that class.

“_—

+0

 

 

pa

re

38

da-
are
on]
wit
tui
n

tu

are

in:

 

76'

Stage Three: Multiple Agreement Analysis of the Cross-Validation Sample

Following the same procedure as that used with the validation
sample, a multiple agreement analysis was computed with the second, or
cross-validation sample. As in the first analysis, 71 different comp
binations of reasons for college choice were discovered. These were
again collapsed, so that only those combinations including two or more
students were programmed into the computer. Thirty such patterns
emerged, making the total matrix one of 16 reasons by 30 response
patterns. Students with individual responseLpatterns were placed in the
resulting groups manually, provided that their reasons were in perfect
agreement with other members of a group.

The results of the multiple agreement analysis of the cross-vali-
dation sample was reported in Table 4.8. To facilitate comparison, they
are listed with the results obtained from the validation sample. In
only two cells of the table did the cross-validation sample disagree
with the validation group. These were in the dropping of the "NOT
'tuition and living costs" reason by Type A students, and the addition of
'"tuition and living costs" by Type D. For ready identification, these
are marked with a double asterisk in the table.

The changes cannot be considered extremely important, because,

in each case, the opposite reason failed to appear. That is,.while

 

77

TABLE 4.8

MULTIPLE AGREEMENT ANALYSIS OF THE REASONS FOR COLLEGE
CHOICE OF THE VALIDATION AND CROSS-VALIDATION SAMPLES

Legend: V = Validation analysis

CV = CrosSSValidation analysis

 

<£>
2>

Reason-Reaponse Types

B C C D
CV V CV V

<33

DN*N*
cvvcv

 

1. Tuition and living
,costs

2. Closeness to home

3. High-educational

4

x
N

standards
. Course of study

offered

5. Chances of a scholar-
ship

6. Religious character
of the school

7. Chances of getting
admitted

8. People you knew who
went there

9. NOT tuition and costs XT*

10. NOT closeness to X X
home

11. NOT high educational
standards

12. NOT course of study
offered '

13. NOT chances of a X X
scholarship

14. NOT religious charac- X X
ter of the school

15. NOT chances of admission

16. NOT peOple you knew who
'went there

N
N

V CV
Persons per type 16 17
Percent of 138 pertype 12 12

N >4

MN N

X
X
X
V CV V CV V

25 30 ,24 22 34
18 22 17 16 ‘25

X**

CV V CV
32 39 37
23 28 27

 

*Unclassified Students

alidation and CrossAValidation Results do not Agree
Note: The X's Indicate Total Agreement

 

 

(7“

 

78

Type A drOpped reason number nine, it did not add its antithesis,
reason number one, and, conversely in Type D, the addition of reason
number one was not preceded, in the validation sample, by the
appearance of reason nine.

It was concluded that, in the pepulation studied, patterns of
reasons for college-choice ggg exist which could be used to classify
students. Thus, the null hypothesis of no such reasons was rejected,
and the alternate research hypothesis accepted. The reason-response
types resulting from the cross-validation sample-were essentially
the same as those for the validation group. Because a basic descrip-
tion of these types was provided in the-discussion of stage one, it
will not be repeated. However, a more complete description-of the
students comprising the four reason-response types will be presented
in the following section.

Stage Four: Identifying Characteristics of
the Cross-Validation Sample

Finally, the 48 items that had significantly differentiated between
one or more groups in the validation sample were subjected to an identi~
cal analysis with the cross-validation sample. The IBM cards containing
responses for the 138 students in this group were sorted according to
the responses to these particular items. The division of the responses
for each item was identical to those employed in-the-validation analysis.

A more stringent level of significance was used in the-cross-
-validation, setting alpha at .10 and, in a12 X 2 table with one degree
of freedom, rejecting those chi-square values under 2.706. The com-

bined probabilities of the rejection regions of the validation sample

 

 

 

5‘}

t0

CI‘I

 

79

(.20) and the cross-validation sample (.10) equal a .02 probability
of committing a Type I error: accepting an item as a significant
discriminator, when, in fact, it should have been rejected. By chance
fluctuation it would be expected that 3 of the 153 original item would
be significant at the .02 level. However, the fact that 16 items (or
one-third of the 48 validation items) held up after cross~validation
indicates that 16 items represent a differentiation between the groups
of students by factors greater than chance.

Of these sixteen items, nine separated students with Type A
reasons from those students with reasons other than Type A; two
identified Type B; two isolated Type C; two characterized Type D;

and only one-was significant for the heterogeneous "Type" N.

Chi-Square Analysis of the Cross-Validation Sample

In Tables 4.9 through 4.13 the distribution of responses and chi-
square values are reported for only those sixteen items which served
to differentiate significantly in-the-cross-validation analysis.

It appears, from Table 4.9, that those students with Type A
criteria for college selection had a higher probability of identifying
.with the following characteristics than did students of other reason-
response types. Type A students in the main (1) were graduates of a
higher socio-economic level high school, (2) were born outside the
county of that school, (3) had mothers with some college education,
(4) were more likely to be Jewish than were students in other groups,
(5) applied to more than two colleges, (6) had friends who agreed

with their college decisions, (7) felt that their high schools prepared

80

 

A.-mc......as.

 

 

 

 

sooav .os.~s ow m cs.v. .mm.m an s «masses as use nuance
Aa-aVaoausuaso
as «a me as «season «so. was nuance am-a
ArnoV:muanom.o>Hunooxm:
mo.V awn...» cm s S.V ammo mm m menu. nonuo 69G oouosoouw
Amvaoonom swan :mugwaom H .aoo
mm ca an am o>flusooxms scum oouososuu coco
. . Amwhuosoo Hoonoo
csmv som.~ ma 0 cauv. see.~ an a sane commune anon uoz
Ao-mvauasoo
we as on on moose. ewes anemone anon mn-a
m oumsvm < < m ouosvm < < usousou souH .oz
-20 on? on? -20 «may «a? _ as“:
uoz uoz
sowuoowwe>wmmouu segumoa~o>

 

z+n+o+m monks..< sexy uoz, Amvuopsss uncommon souH u A V

ooaaae<.so«uoouuoo noun» a *

”snowed

 

 

mmsmzem Anna 1 2 .cs. n.aeaa<v zOHH<nHa<>.mmoeu nz< Ammo u z .oN. u <ema<v
zoma<an<> may meow zH .< mass z<ee sumac mzmmee<a nze meouaaoo quaomsmm_moa

mzom<mm mo mZMMHH<m ¢ MPNH zmmBHmm an<HHZMMmmmHG MAHZ<UHMquHm mUHm3.mZMHH ho HZMHZOU

a.¢ mAm<H

 

Am-ov

momv s¢m~.m me n dodv koo.w Nu m mowoaaoo N sszu mama 0» ooaaaad
Ae-~V
mm No me as mowomsoo once so n on sadness :<= nonm-m
Aa-w.n-ov
moo.uv swm.na cum am Hoomv «no.mm aaa m moonHou omonu osouuo uoc moon
Aa-svamommou
.msosoz snoummu usoam>wnuo
a s 0H o no :oswoou h>H= so monouu< nno
Am-mv
musoosum Season we sonuuomoua
mo.v. aam.e com am Honv aem.n mm m Heewan s confines uoc moon

Anymomhu vogue
menu musoosuu season we

 

1
.u es s ON a soauuoaoua scams: a masseusH so-m
m seesaw 4 < m seesaw m < uncucoo souH .oz
- Eu 25 «may - as on? $5.. suuH
uoz uoz
83332,. $85 533:;

 

z+n+o+m mommH n < coke uoz Amvuonsos uncommon souH n A v moaammm seguoouuoo moumw u k "rammed

 

 

mmumz<m Ammo u z .os. M seesaw zOHH<eHm<>ummoeo nz< Anna 1 z .om. u <eme<v
zOHeenHe<> ems mace 2H .4 mass 24mg mango messages oz< announce ezHaomsum mom
mzom<mm mo mzmmee<a a mass zmmzemm emeeHazmmmaaHe usezeonquHm eoHes mzmeH so azmezoo

.uaoo .m.e mmm<a

 

 

 

82

 

3 .V «N: am e

Amuovaoonom
cumsoouw on so cm as wowssoaa uoz

AmV Hoonow

3V «mag em a

 

 

 

mm m mm m ouosoouw cu no em ou wsassoam unuo
Am-aV
:musoosum umos.so3u,uouuop: mesa
S.v .35 S n 3V is...” s... e a... consummate moose. SE 3am
onzmuooosum umos_sonu acumen:
2 S R S as. aoflmumaasa 323 SB 3...... $10
Awumvsomoso.owoaaoo sues
Nov .35 S m S.V .3: an a .eomsfla .53., est 2.3. Sasha
Aquomono,owoadoo.
an 3 a... 3 staffing? be... 3.... Sufism 2.6
m cumsvm < < m cumsvm 4. < unoucoo souH. .oz
.30 as; as; -30 «a? 2:9 as:
uoz uoz
seguoowfim>1emouo cofiuooaao>

 

z+a+o+m mon%H_n < cozy uoz vauonsss uncommou,souH u A v oouaan<.soauconuoo some» n % homewon

 

 

mg... 82 n z .3. H 45.5 zoflfifiéémog 92 82 u z .8. a seat
zofiefigs as ES 3 .4 mg EB swab manages ea $3.58. ezflofimm mom

mzom¢mm mo mZMMHH<m 4 mmwa zmm3amm QMH<HHzmmmmmHn_NAHZ<UHMHZUHm mUHmz mZHHH mo HZMHZOU

.28 .a... fig

c F q\ r3,.-Nn<t—.

 

 

 

 

 

83

 

Ase
owoaaoo mo sewage can suds

 

 

 

~o.uV ae.n as as ~o.uv am~.o Me a =eomuoma aua>s any: message
Am-nvowoamoo so
cowoso.man nua3.:oooooam
we as mm. on auo>= amen mass one: assuage ~m-o
Aa.o-ovmooeua and: Anna Hanan
no.v 3a 2 2 S.v .25 i S a 5:3 aafio sous sausage
Am-aVsooeuu ewes a .moo
on as an, em euao,msuau.u scum conquests «coo
m museum m m m ouosom m m usouooo souH .oz
.20 some mar... .30 «can 25. so“:
uoz uoz
soauueasmsummouo aoauaeamm>

 

z+n+0+€ momma u.m onha uoz Amvuonsfis uncommon souH n A v vowamg< souuoouuoo mouow I %

"osowoq

 

 

meamzem Ammo u z .os. u emaaev onaeaHa<>ummqmo nz< Anna 1 z .ou. a <mma<v
zoaaeaaa<> may egos 2H .m ease zeea.mmmao mzmmya<m nz< muumauoo ozaaommmm mom
mzom<mm so mzmmaaem a mass zmmzamm nua<HazemmaaHn saez<onHzeHm mons mange so azmyzoo

o~.¢.mnm<a

it "ta—1

 

 

~84

 

£3333. means no“ 253

 

 

3V is... S 5 av .32” .3 w Lama a 32s 3 names no: 38
Amnmuowoaaoo weaken
3. 2 3 S no“ «States a 32 3 33%» «SA
Aau~.ovo>ono omens comm nonuo
S.v End 2 2 2V as; a 2 «mafia. econ"... 3 has: 93:
Adveeouo housed
no meosmHAH as momoaaoo
2 a 2 a .2333 33% 3. 39:: «so: To
m museum 0 u m museum 0 o usousoo suuH .oz
.30 25 25 -20 25 25 53H
32 32
souuooaflm>ummouo wmmwmoMAmM

 

z+g+m+< wonky I10 mama uoz Amvuonsss uncommon souH a A v vowamm< souuoouuoo nous» n,«. "osomoq

 

 

mega 32 u z .3. m «EQV zofleflqg Home 82 $3 a z .2. u .353
zofléfig as Eon 2H .0 were fies mess mzmmEE as mmumfloo uzflofimm mom

mzom<mm mo zmmya<m o mmwa zmmzhmm QMH<HHzmthmHn MABZ<UHMquHm mUHm3.mzmHH ho.HZMHzoo

AA.¢.mAm<H

 

 

Annnvomoaaoo wswuso

Ncmv. oo.n No NA no.uv an.m on ma xoos muo>o mousse vacuum uos moon
Aevowoaaoo .
No ON mm NH wswuso.xoo3 kno>o mousse monouud onNuo

Am\mm .e\ouvumuam
:Noonom aw oHnNewon we more
we measured: axons one umoa uN
axons .uoom sH .umufiw :. . .
do.uv o~.~ am oN Hoo.uv *mn.mN we we snowmen macawaaou . . .: uos moon
AH\oNv:u0m o>Nnum cu wagon:
uceuuomsa mass was no :mwoaaoo

85‘

 

 

 

no ON 0 on waoamumos as on a: scopes: magma was-o
m seesaw a n m museum n n usousoo souH .oz
.20 case 25 .20 25.. case 83
uoz uoz
soNumvNNm>1mmouo dowuooaam>

 

z+o+m+¢ momma u a sexy uoz Amvuonsss uncommon souH n A v oofiamm< cowuoouuoo moon? n a "common

 

 

mmaaz<m Anna 1 z .om. ”.emmaav onaenHa<> mmomo nee Anna 1 z .oN. n semaev
one<nHm<> use neon zH .9 mass z<ee «mayo mzmmea<m nz< muwumsoo ozHeomsem mom
mzom<mm mo mzmmaeam a mass zmmzaem an<HezmmeaeHn asez<onquHm monz mzeeH mo azmazoo

NH.¢ mqm<a

 

 

Aa-uvumamaoo

 

 

 

 

8V SN. 8 R, EV 3.” on 3 means. «so: an «>3 bassoon
no cm as., on Aavamosaoo,maauae_oaoa us mo>as coauo
M m seesaw z z m cums—om z z - acousoo 63H .02
name 093. 093. 1.30 can. 093.. . Sou...“
uoz uoz
couuoozoawnewouo scaumozmdw

 

n+0._.m+< 39A“. I 2 can. uoz Amvuoosos announce. an”; n A v oozes? coauoonuoo some? a a .“osowoa

 

 

mmamzam Ammo u z .om. m <mma<v zoeaenHaeswmemo nz< Anna n z .ou. u.ammm<v
oneaaHa<>,mme anon zH .z.mmwa z<me amuse meanness nz< mmumemou.czHaomnmm mom
mzom<mm mo meanness 2 mass zmmzamm ama<He2mmmaaHa asyz<oHequHm mUHms mzeyH so azuazoo

mg .q MAB;

 

 

 

87

them.better than most students, (8) were more likely to attend an "ivy
league" or "seven sisters" institution, and (9) planned to'continue
their education into graduate school.

‘Students with Type B-criteria were more likely to be graduates of
high schools incities of 25,000 to 100,000 inhabitants. Their closest
friends did not enthusiastically support the Type B student's college
selection; the largest preportion of friends either thought the Type

' or "didn't really care one way or the

B's could have "done better,’
other."

Students with Type C reasons apparently were characterized by
financial motivations to a larger extent than.others. These students
were more likely to attend the less expensive teachers' colleges, and
fully expected to'earn part of their college expenses. In the valida-
tion sample, these were the only students whose parents were.not~con-
tributing more than 75 percent of their college~expenses;:however,
this finding did not hold up inrcrossdvalidation.

Religious factors were more influential for Type D students than
for any other group. In college they attended church weekly, and were
inore likely to rank-"living.up to my religious beliefs" as a more
important goal than "pleasing my parents," "learning.as much-as
possible in school," or "being accepted and liked by others."

For the heterogeneous "Type" N group, with no agreement on a
sequence of.reasons,the only item-which proved significant for both

samples was a question asking where the student lived during.college.

While the majority of this group, like the majority of all college

freshmen sampled (82%), were not home residents, the preportion that

were was significantly higher than the prOportion for other groups.

. y."

T'N“ V—y

 

....__

 

 

88

Reliability Estimates

In order to check on the internal consistency of the eight reasons
for college selection used in the multiple agreement analyses and the
16 items which remained significant after cross-validation, Hoytis
Analysis of Variance reliability coefficients were computed. Two
separate stratified random sub-samples of 50 students each were~drawn
from the combined validation and cross-validation samples for these
reliability calculations. For each sub-sample, 25 students in the
validation sample and 25 students in the cross-validation sample were
selected by randomly choosing approximately one-third of the students
in each of the reason-response classifications.

The reliability coefficients for the eight items used in the
multiple agreement analyses was .63, while the reliability of the 16
items which held up after crossivalidation was only .32. The differ-
ences in these two estimates of item stability might help to explain
why the crossdvalidation of the multiple agreement analysis (stage-3)
agreed, in all but two cases, with the initial attempt to classify
students according to their college selection reasons (stage 1); yet,
the cross-validation of the distinguishing characteristics of the
students in each reason-response classification (stage 4) did not
substantiate the original descriptions.of these groups (stage 2).
Because the purpose of the 16 items was to identify the differences
between the reason-response groups, where-each group was-distinguished
by its responses to different items, the low degree of internal con-
sistency might help to explain the lack of unidimensionality of these

16 items. This, in turn, would support the somewhat sketchy description

 

89

of the differences between the students in the various reason-response
groups because these descriptions were dependent upon a relatively

small number of unstable items.
Summary

Multiple agreement analysis was-employed to separate each of two
sub-samples of 138 students into groups exhibiting a commonality of
reasons. When poiitive reaponses were analyzed together with the lack
of reaponse to eight "reasons for choice," 72 percent of the students
in the validation sample, and 73 percent in-the crossvvalidation, were
~classified in four categories of reason-response types: Types A, B, C
and D. The.remaining students, whose-unique combinations of reasons
did not coincide with any of the established patterns, were designated
as "Type" N.

Chi-square analysis was made, setting alpha at .20, of the respon-
sea of students in the validation sample to 153 other available items.
Forty‘eight significantly differentiated the students in at least one
reason-response category from the remaining students. ‘When these 48
items were subjected to another chi-square analysis with the cross-
'validation group, setting alpha at .10, only 16 of the.items signifi-
~cantly differentiated reason-response.types. Thus, sixteen character-
istics were discovered which served to "identify" or describe the four
”major-reason-response types isolated, plus the unclassified students.

A summary of the college selection patterns and identifying

'characteristics for these five groups is included in Chapter V.

 

‘ry-rflf

 

 

90

It must be remembered that a majority of the students in the
validation sample answered 105 of the 153 available items in prac-
tically the same manner. Despite the fact that for purposes of
scrutiny they could be separated according to patterns in their college
selection criteria, these students exhibited homogeneity on at least

three-fifths of the items employed in this research.

CHAPTER‘V
SUMMARY AND CONCLUSIONS
Purpose

It was the purpose of this research to determine whether common
agreement existed in students' college selection criteria, and
whether students could be separated into readily-identifiable-cate-

gories on the basis of such agreement.
The Sample

The sample-employed in this researdh“was a small portion of the
high school papulation surveyed in 1957 and 1958 by James Coleman,
and reported in his book, Th; Adolescent Society. Of the approxi-
mately 10,000 original students from ten northern Illinois high schools,
581 completed a questionnaire in-the Spring of their college freshman
year, as part of a follow-up study by Coleman and Peter'Rossi, of the
National Opinion.Research Center in Chicago. (Survey 412.) From
among these 581, the author extracted.276 who had completed fall and
spring questionnaires-their senior year in high school, indicating
their-choice of a specific college and their reasons for choice, and
whO‘were.enrolled at that particular college when completing the third
questionnaire the following spring. For the purposes of validation
and crossdvalidation,~thersample-was randomly divided into two groups

of 138 students each, along the lines of high school representation.

 

92

Design of the Study

The-design of the project involved four stages--two for validation
and two for cross~validation.

Stage 93;: To discover whether common patterns of reasons exis-
ted in students' selection of-colleges, Hemingway's multiple agreement
analysis was employed. This is a technique which classifies reaponses
so that the reSpondents in any one group agree with each-other more
than they agree with the responses of subjects not in that group. It
also yields the items which all members.of that group agreed upon.

The reasons which 138 students checked or failed to check were sub-
jected to this analysis, and resulted in the classification of patterns
of reasons into four groups accounting for 72 percent of the sample.

Stage Egg: To identify the distinguishing characteristics of
students in each of the groups isolated in stage one, a chi-square
analysis was made for 153 available itens from the three questionnaires.
With alpha set at .20 for the validation sample, 48 of the items yielded
significantly different responses for at least one group of students.

Stage Three: A second multiple agreement analysis was run on the

 

morass-validation sample. Except for a-variation of one reason for each
>of two groups, this confirmed the findings in stage one by yielding the
same basic reason-response types, comprising 73 percent of the sample.
figagg E225: Chi-square analysis-was again employed with the
_cross-validation sample, setting the rejection region at the .10 level.

0f the 48 items isolated in stage two, 16 were found to be significant.

'93

ReasonéResponse Types Identified

The following distinguishing characteristics were identified for
students in each of the groups displaying similar patterns of college
selection‘criteria.

TXEEHA: The first, the smallest group of students, was unique in
their disregard for the proximity of the-college to their home, and
-their insistence on its high-educational standards. -No other group
of students completely agreed on these two reasons. Like most students,
Type A felt that the course of study offered was_important, but were
windifferent to the religious character of the college or their chances
for a scholarship. Although the validation group expressed a unanimous
indifference to tuition and living costs, this finding was not borne out
in the cross-validation.

When the~distinguishing:characteristics of group A are examined,

a logical connection to the reasons expressed is found. Students with
Type A reasons came from high schools in higher socio-economic suburbs
and felt better prepared for-college than most other students. Their
parents tended to be more mobile geographically, were more likely to

be Jewish, and the mothers had probably had some college education.

They had applied to more than two'colleges and their friends appeared
‘to strongly support their choice. Type A students tended_to=congregate
at "ivy league" or the prestige eastern women's colleges, and planned

to continue.their education into graduate school. In short, these were
wealthier students who wanted quality education at prestige institutions

and were apparently willing to pay for it.

 

“

‘94

Type g; The second group-type were the only students whose
college choice was not influenced by their chances of gaining admission,
nor by the peOple they knew'who went to their colleges. They agreed
with other students in disregarding the religious character of the
institutions they were considering, but were insistent that their ’
school should have the course of study they wanted atta price within
their means. They came from high schools in small cities, but went
to a variety of colleges. Their friends thought that they could have
made a better choice than they did. In conclusion, these students
'emerged as independent thinkers who sought a college they could afford
that would provide them with their chosen course of study.

2222 Q: IMore so than the students-in the other groups, Type C's
were concerned about their chances of getting admitted to the colleges
of their choice. However, they were.not interested in their chances
of a scholarship, nor in the religious character of their schools.
They were students who would hold part-time jobs during college and
were-considering the expenses involved. Perhaps this partially ex-
plains their tendency to enroll in the lower-cost teachers' colleges
near their hometowns. A case might be made that these students were
<primarily interested in getting a college education which they could
realistically finance. The course of study offered, while possibly
important, was not as vital for them as for the other groups.

Typgflg: The most significant characteristic of students inthis
group was their religious enthusiasm. They alone attended.church
‘every week during:college, felt that living up to their religious

beliefs was the most important thing to strive for, and included the

95

religious character of the college as a reason for selecting it. They
came from.a wide variety of backgrounds, attended many types of colleges,
and, like most other students, were not concerned with their chances of
a scholarship. They did insist that the colleges provide them with
their intended course of study.

$222.33 Since the last type were the students whose responses
were not in total agreement with any of the above patterns, the hetero-
geneity of their reasons for choice is an obvious conclusion. Appar-i
ently they chose their colleges for reasons unique to themselves or
particular to the colleges they attended. They included the»largest
single percentage of students (28 percent in the validation sample
and 27 percent in the cross-validation), and had no meaningful identi-
fying characteristics. The only significant item in both analyses
reported that slightly more of the Type N students lived at home
during college than did students in the other groups. However, since
72 percent of Type N students did not live at home, compared to 82
percent of each sample, this item cannot be construed to indicate that
living at home is a characteristic of students with unique patterns of

college selection criteria.
Conclusions

The major hypothesis that students could be classified-according
to their reasons for college selection was supported. The minor
hypotheses-«that the content of the questionnaire items would identify
the characteristics of students in these.classifications--were par-

tially supported. Because of the limited number of items found to

' Sir

 

96

hold after cross-validation, it was not possible to crystallize the
identifying:characteristics of the students in-each.type of reason-

response pattern to the-extent originally anticipated.
A Comparison with Previous Literature

The percentage of students who checked each of Coleman's eight
"reasonsfor college choice" is interesting_in light of the review
of the results of previous studies in Chapter II. These percentages

are reported in Table 5.1.
TABLE 5.1

PERCENTAGE ORDER OF REASONS FOR SELECTING COLLEGES
OF 276 HIGH SCHOOL SENIORS

 

 

 

Total Percent

Item Sample of 276
Course of study offered 215 78%
Tuition and living costs 197 71%
AHigh educational standards 178 64%
Chances of getting admitted 102 37%
Closeness to home 4100 136%
PeOple you know who went there 77 28%
Religious character of the school 72 “26%

Chances of a scholarship 54 20%

 

 

 

97

Students from this study agreed withoothers sampled since 1945 in
ranking "course of study offered" high. The frequency of checks for
'"high educational standards" would also seem consistent, although
revealing a basic area of confusion in many studies. Items were often
phrased in such a way as to blur the distinction between academic
standards and the student’s concept of the reputation of the college.
Here the addition of "high" to "educational standards" could well have
called into play the latter. A clearer division between "academic
standards" and "reputation and prestige" is advocated for future
instruments.

The surprise item of the top three is "tuition and living costs,"
which would seem to challenge the findings of previous investigators.
Its appearance raises two questions. (1) Had Coleman included other
factors which appeared stronger than costs as a result of the compila-

tion in chapter II--i.e., "reputation, fivocational training available"
and'"size of the college,"--wou1d cost have drawn as much response?

(2) Is the importance of cost one of the factors which would be most
minimized by the practice of sampling students already a year along

in college? An obstacle hurdled rarely looms as large in retrospect,
and it is known that many for whom cost is most crucial never appear

an campus.

1G. Stice,.W.G. Mollenkppf and W.S. Torgerson, Background Factors
,Relating to College Plans and College Enrollment Among Public High
School. '__ ass, (Princeton: EducationairmastingServiceyVISS7I; a
Appendic C, Table C-24.

 

 

98

Of the four researchers besides Coleman who investigated high
school student's selection reasons, neither Plante nor McGranahan
*chose to include cost as a factor for-investigation. As one might
*expect, cost ranked only 11th and 12th in.Holland's study of male and
female National'Merit ScholarShip winners. But Basta, studying a group
not disprOportionately laden with scholarships, found that cost ranked
first in importance!

COmbining the findings of this project with Basta's, one is
inclined to question the results of other studies which minimized cost
as a variable. At least, cost as a factor deserves full representation
in future sampling instruments.

Two of Coleman's items--"chances of getting admitted" and "chance
of a scholarship"-~were sufficiently unique as to make any com-
parison with predecessors impossible. And his inclusion of "pe0ple
you knew'who*went there" among "college" factors makes its importance
-difficult to assess. (See number (4) under "Recommendations for
FutureQuestionnaires" later in this chapter.)

The present study did confirm the impression that "closeness to
home" and "religious character of the school," in that descending
order, were of minimal importance.

Elsewhere the students had an Opportunity to rank influential
persons, and they placed parents, friends, guidance counselors and
teachers in that order. In this they agreed with most of their
predecessors since 1926, but gave the fledgling guidance profession

the edge over teachers in influence.

 

99

In several other areas this study contradicted or supported
previous.research findings. It did not support the hypothesis of
sex differences in the selection of college advanced by Reeves in
19302 and Holland in 1957.3

The-effect of affluence as a factor-in college selection has
‘been demonstrated by Coleman,4iMedsker,5Wise,6 Beezer and.Hjelm,7

Plante,8 and The Educational Testing Service.9 In the present study,

 

2Ellis M. Reeves, "Which CollegeZ" Journal 9; Higher Education,
Vol. 3, (Feb. 1932), 7-14.

3John L. Holland, "Student Explanations of college Choice and
Their Relation-to College Papularity, College Productivity and Sex
Differences," College Egg Univgrsity, Vol. 33, No. 3 (Spring, 1958),
313-320.

‘"James 5. Coleman, The Adolescent Society, (New York: Free
Press of Glencoe, 1961), 261-274.

5Leland L. Medsker, Factors Related 59 College Attendance 33g
Performance 2; High School Graduates, Paper presented at the
Annual Meeting of the American Educational Research Association,
Chicago, (Feb. 14, 1963.)

Students Toda , (Washington, D.C.: American Council on Education,
1958) , 6'22 o

, '7Robert H. Beezer and Howard F. Hjelm,«Fa§tors Related 22
College Attendance, (Washington, D.C.: U.S. Department of'Health,
Education, and Welfare, Office of Education, 1961), 34‘37.

aAlekander'J. Plante, "An Analysis of the Plans for Post-
Secondary Education of 1959 Secondary school Seniors in Connecticut
and Certain Factors'Which Influence.Their Plans," (unpublished
Ph.D. Dissertation, University of Connecticut, 1960), 142-149.

9G. Stice, W. G. MbllenkOpf, and W. S. Torgerson, Background
Factors 32g Collegg Going Plans Ameng High Aptitude Public High
School Seniors, (Princeton: Educational Testing Service, 1956),
4-7; and Background Factogg Related 52 College Plans Egg Collegg
Enrollment Among‘High School Students, (Princeton: Educational
Testing Service, 1957), iidvi.

 

100

when similar-types of institutions were combined and scrutinized, the
.most obvious differentiating characteristic was the socio-economic
status of 7the students enrolled. For example, students who could
afford to pay for what they perceived as a quality education repeat-
edly tended to select a few high prestige institutions.

Of all the socio-economic indices, only the level of education
of ‘theJmother was a significant differentiator between reason-response
types. Its importance confirms the findings of Hollingshed10
Robertson,11(30tt11eb,12 Kah113 and the current paper by Medsker in
1963.14

Finally, peer group influence has received much emphasis in
current studies of high school climates and college-goingetrends,
as well as in the studies reviewed in Chapter II. Besidessocio-

economic indications, the onlyother item which distinguished between

 

1allay. B. Hollingshed, "Why Do Persons Go To College?," School
Egg Society, Vol. 23, (May, 1926), 565-568.

lJ-Robertson, "Who Influences Freshmen 'to Choose Goucher?,"
JouEal 9; Higher Education, Vol., 7, (May, 1936), 271-272.

12David Gottlieb, "Social Class and the College .Going
Experience," Adolescent Behavior in Urban Areas: A Bibliographic

(Jon Reeves, (Michigan State University, April, 1962), IV-33--
Ill-44.

13.] oseph A. Kahl, "Educational and Occupational Aspirations
.of 'Common Man' Boys," Harvgrd Educational Review, Vol. 23,
(Smer,1953), 186-203.

‘u'Leland L. Medsker, Factors Related 53 Collegg Attendance
_agl_d_ Performance 2f High School graduates, Paper presented at the
Annual Meeting-of the American Educational Research Association,
Chicago, (Feb. 14, 1963.)

101

more than one—group of students was the extent to which.the students'

friends agreed with the selection made.
Limitations of the Conclusions

Although Coleman maintains that his original sample of 10,000
high school students from ten northern Illinois high schools was
representative of American students in general, it is doubtful
whether the-conclusions above on the college-choice aspects.of his
data are applicable beyond the Midwest. The [concentration of
colleges _inethe Chicago areaand the "Big 10" universities in this
section of the nation could conceivably have affected the patterns
of reasons discerned among Illinois students. Indeed, the large
inumber-of students in every reason-response group selecting "Big
10" universities was striking. One cannot readily assume that the
results would have been the same had_this study been done elsewhere
in the United States.

In addition, two basic limitations in the nature and treatment
of this data may have affected the-conclusions:

(l) The fact that the groups of students-who were 80*clear1y
separated by their reasons for college‘selection, were not as clearly
identified in other reapects, may be the product of insufficient data.
Of the 522 items originally included in the three-questionnaires, only
153 were made-available. This was-the unforeseen-result of applying
for the collapsed IBM work decks, in which many items had been deleted.
Had all the original information been applied, the differences between

the reason-response.types might have been more fully explained.

102

(2) During the time of the analysis, when.Michigan State Univer-
sity was phasing out its MISTIC computer in favor of a new one, long
programs run on MISTIC yielded notoriously Spurious results. The
original validation data for the study was programmed into MISTIC six
times before results free of error were obtained. Although manual
analyses assured that the results were as reliable as could be
-expected, minute errors resulting from MISTIC and the 082-10 IBM Card
Sorting Machine of the College of Education allow some small room for

question.
Implications for Further Research

Six general and five Specific recommendations for further research
are-offered.
General Recommendations

(1) Subsequent investigators attempting to replicate this study,
or-explore similar facets of it, would be wise to secure samples
independent of Coleman's original data. .Aside from the obvious
‘handicaps of working with someone else's data, the design of the
questionnaires could be improved for-college selection research.
Coleman was primarily interested in continuing.his study of the
leffect of high school climates into the reahnof-college.selection.
This is quite different from exploring the criteria for-college
selection .2?! g.

(2) Because the-college selection decision is made during;high
.school, it is recommended that samples be-drawn from that level--

.ideally, Spring of the senior year--regardless of the difficulties

103

involved. It is noteworthy that, of 581 college freshmen who completed
the-NORC questionnaire, 475 indicated that they were "very pleased"
‘with their selection and could not have made a better choice. If this
tendency to adjust to their environment should serve to warp their
"recall on their initial motivations far choice, than even the-director
of admissions wishing to discover why students choose hig particular
institution, should sample students prior to their arrival on campus.

In this connection, there appears need for a study to sample one
papulation in the spring of the senior year in high school, and again
in the Spring of their freshman year in college, to see whether
maturation and adjustment to environment do affect students' recall
as to their initial reasons for choice. In addition, a question such
as NORC Survey 412's, "If you had to choose all over again, what
reasons would you give the most weight or'importance to?" might prove
revealing for many reasons, including subsequent refinement of the
sampling instrument.

(3) It was initially suspected that, if one could discover patterns
of reasons for college choice, one might eventually find correlations
between students' patterns of reasons and the types-of colleges they
chose. Although this was strongly suggested in the analysis of reason-
response types A, C and D, it is doubtful whether 276‘students-from
-ten different high schools enrolled in more than 50 different colleges
can provide anything_but generalizations in this direction. Besides
the complexity of the decision-making process itself, the variations

in high school climates (demonstrated to influence college

104

decisionsl5), and in college climates as well,16 further complicate
the problem.

It is therefore felt that a study attempting to generalize to
national trends would have to involve a huge initial sample of high
school students, particularly in view of the attrition rate between
those who qualify as "college-bound" in the spring of their senior
»year, and those who actually enroll. This goal, if capable of attain-
ment, probably belongs in the realm of sponsored research.

Nevertheless, studies of value can be made on a smaller scale,
if samples are limited to a few Specific high schools or to high
school students accepted for admission by a selected number of colleges.

(4).An-exploration of this field has resulted in convincing evi-
‘dence that the factors Operating as-determinants in college choice
,involve more than one dimension (i.e., social class background,
students' ideals and expectations, vocational choice, peer group
influence, parental influence, etc.) In addition, Holland, who called
for'"more adequate criteria for evaluating colleges," and "more
~counse1ing information about colleges" for students, concluded of

his National Merit scholarship winners that they "appeared to make

 

15Philip Outright, "Students' Decisions to Attend College,"
Journal of Educational Sociology, Vol. 33, (February, 1960), 292-299;
'and.James S. Coleman, The Adolescent Society, (Nevaork: Free Press
of Glencoe, 1961), 266-274.

16David Gottlieb, Social Class, College Students gn_d gs Ideal
Qollegg Professor, Paper presented at the Annual Meeting-of the Ohio
Valley Sociological Society, (April, 196]); Martin Trow, _Thg Entering
Collegg Student, Paper presented at the Annual Meeting of the
~American.Educationa1 Research Association, Chicago (February 14, 1963);
and Alexander W. Astin, A Method.g_i_:'_ Omaring 5h; Effects ‘9_f_ Colleges
23 Student Performance, A paper presented at the Annual Meeting of the
American.Educational Research Association, Chicago, (February 14, 1963.)

105

choices in the same way consumers often, if not usually, buy household
goods; they select-colleges by means of vague notions about reputations
and values which they seldom can document meaningfully."17

It would seem, therefore, that attempts to assess reasons for
‘college-choice closely approach attitude measurement, and might benefit
‘considerably by the use of techniques in the latter field. To date,
use of this approach has not been reported. Three steps are recomr
‘mended to deve10p a reliable instrument forrmeasurement: (1) compile
a thorough list of possible reasons, through reference to the previous
literature and interviews with high school students; (2) facet analyze
the.reasons obtained; and (3) test each facet, using Guttman or
similar scaling techniques to determine-unidimensionality~of the item.
Then, and only then, can one be sure what a-response to an item
indicates.

The confusion resulting from the lack of precision in items here-
tofore was discussed in Chapter II.

(5) In-the validation sample of this research, some.of the 48
items which proved significant at the .20 level in chi-square analysis
were strongly suggestive of personality syndromes wdthin the reason-
response grOUps. It would seem desirable to administer such instru-
vments as the Edwards Personal Preference Inventory, the Allport-
Vernon-Lindsey Scale of Values, and the Rokeach Dogmatism Scale

simultaneously with the questionnaire on reasons for‘college choice.

 

17John L. Holland, "Student Explanations of College Choice and
Their Relation to College-Papularity, College Productivity and Sex
Differences," Collegg and University, Vol. 23, No. 3, (Spring, 1958),
319.

 

 

 

106

(6) In addition, data normally available in high school cumulative
,records should be included as identifying information: interest inven-
tories, aptitude (including I.Q.) and achievement tests, adjusted

grade point averages and class ranks. One would suspect that

interest inventories particularly might have.relevance in distinguishing
between patterns of reasons, although in the past these sources'have
been completely disregarded.

Specific Recommendations ._f_o; Elm Questionnaires

(1) To assess the importance of an item, as Opposed to simply
its prevalence, instruments employed should involve rank order testing
requiring preferential choice. For simple studies this may be adequate;
where_thoroughness and reliability are important, rank order and the
testing of items for unidimensionality (discussed previously in this
section) are recommended.

(2) A larger“matrix than Coleman's-eight would be desirable in
searching for patterns. Presumably this would evolve naturally in
prevalidating the instrument, since, judging from the review'of the
literature, Coleman made-at least three notable omissions: ."avail-

" "size of the college" and "reputation

lability of vocational training,
or-prestige." (In this connection, a clear distinction should be made
between the academic standards of the college~and its:reputation and
prestige, both of which were commonly confused in the studies
-reviewed.) In addition, the availability of loans at a college is

a relatively new factor which would seem worthy of analysis.

(3) It is recommended that influential persons, financial con-

siderations and what have been called herein."college" factors, should

107

be tested for separately, rather than in one rank order. The discovery
of other reasons might suggest other divisions. This latter point is
simply a reflection of the belief that college choice is not a one-
dimensional matter.

(4) The questionnaire should require students to indicate which
reasons were unimportant to them, as well as those important.

(5) As far as possible, objective items or statements should be
used. While researchers like Olsen18 and Schloemer19 have admirably
tried to recapture the flavor of students' own statements in such
items as, "The educational program at MlS.C. is far-sighted and
progressive,20 the emotional identification which such an item
necessitates on the part of the student checking it, complicates the

problem of analyzing reaponses.

This research supported the hypothesis that patterns of reasons
for college selection can be isolated. Although the identification
of characteristics of the various reason-reaponse types was less
successful, it is haped that the thorough review of the literature
in Chapter'II and the suggestions for further research advanced here

may generate more conclusive investigations.

 

18Hans C. Olsen, "Elementary Education: Student Reasons for
Selecting Their Colleges," Journal 9; Teacher'Education, Vol. 10,
(September, 1959), 315-318. ‘

19Clarence C. Schloemer, "Why They Choose Your College,"
College Public Relations Quarterly, Vol. 4, (July, 1953), 22-28.

2°1bid., 27.

 

BIBLIOGRAPHY

Addington, L. F., "Choosing a College in Virginia," Virginia Journal
2; Education, _Vol. 32, (June, 1939), 405-6.

Anderson, G. L., "The Reactionsof Minnesota College Freshmen To Their
First Year of College Experienc-e," Studies _i_n Higher Education,
Biennial Report of the Committee on Educational Research,
Minneapolis: University of Minnesota, 1941.

Astin, Alexander J., "A Method of Cornparing the Effects of Colleges
on Student Performance," A paper presented at the annual meeting
of the American Educational Research Association, Chicago:
February 14, 1963.

. , "An Empirical Characterization of Higher Educational
Institutions," Journal 2f Educationagl Psychology, Vol. 5,
(1962), 5.

, "Influences on Students' Motivation to Seek Advanced
Training," Journal 2; Educational Psychology, Vol. .53, (1962), 6.

Barday, D. , "Thinking Ahead To College," Nit! York Times Magazine,
(April 3, 1955) , 46.

Basta, Samuel, "An Analysis of the College Selection Patterns of
“Recent High School Graduates in Selected Western States with
Special Emphasis upon Nevade," Unpublished Ed.D. Dissertation,
University of Southern California, 1960.

Beezer, Robert H. and Howard F. Hjelm, Factors Related 53 Collegg
Attendance, Washington, D.C.: U.S. Department of Health,
Education, and Welfare, Office .of Education, 1961.

Bennet, M. E., College fl Life, New York: McGraw Hill, A 1933.

Boas, R. P., "Considerations on the Selection of a College," School
and Home, Vol. 16, (April, 1935), 668-71.

Bowles, Frank H. How 3 Get Into College, New York: E. B. Dutton
and Co., 1958.

Bowling, William 6., "An Appraisal of College and University
Publications used for Promotional Purposes," Collegg and
University, Vol. 29, (October, 1953), 1.

Clark, W. H., "Education by Facade .or How to Choose a College,"
American Association if. University Professors Bulletin,
Vol. -38, (December, 1952), 528-68.

 

109

Cleeton, Kenneth, N., "An Analysis ‘of thevFactors Which Influence
Students When They Choose a College," Unpublished Ph.D.
Dissertation, University of Colorado, 1951.

Coleman, James 8., The Adolescent Society, New York: Free Press of
Glencoe, 1961.

Committee on Discrimination in College Admission, 93 Getting Into
College, Washington, D.C.: American Council on Education, 1949.

Coombs, Clyde ‘H. "The Theory of Data," Unpublished manuscript,
University of Michigan, 1962.

Corey, S. M., "Students' University Choices," Journal _o_f Higher
Education, Vol. 7, (April, 1936), 207-11.

Cutright, Philip, "Students' Decisions to Attend College," Journal
pg Educatipnal Sociology, Vol. ‘33, (February, 1960), 292-99.

Dillenbeck, Douglas D., "Facing CollegeAdmissions Pressures
Professionally," The; School Counselor, Vol. 10, (March, 1963), 3. I

Douvan, Elizabeth and Carol Kaye, "Motivational Factors in College
Entrance," The American Colle e, ed. Nevitt Sanford, New York.
John Wiley and Sons, 1962.

Eddy, E. D., Jr., "Selecting a College," School an_d_ Society, Vol. 89,
(February, 11, 1961), 46.

 

Edwards, Allen L. Techniques o_f Attidude Scale Construction, New York:
Appleton-Century-Crofts, 1957.

Ellison, E. J. and Clarence E. 'Lovejoy, "How to Choose a College,"
Saturday Evening Post, Vol. 228, (November, 19, 1955), 42-3.

New York: The Dryden Press, 1953.

Fine, Benjamine, How 2 he Accepted _i_pvthe College pg Your Choice,
New'York: Papular Library, 1961.

Forrest, A. , "Counseling Talented Students :on College Choice,"
Personnel and Guidance Journal, Vol. 40, (September, 1961), 42-7.

Foskett, John and Walter Martin, "How Students Choose Their Alms
Mater," Overview, Vol.2, (June, 1961), 23.

Franks, Lawrence '13., "A Study to Determine _the Major Factors Which
Influence Students to Enroll at Stephen A Austin State College,"
Unpublished Ed.D. Dissertation, University of Indiana, 1959.

110

Gottlieb, David and Jon Reeves, Adolescent Behavior ih Urban Areas:
_A_ Bibliographic Review and Discussion p_f_ the Literature,
Michigan State University, College of Education, April, 1962.

, "Social Class, College Students and the Ideal College
Professor," Paper presented at the annual meeting of the Ohio
Valley Sociological Society, April 21, 1961.

Hadaway, C;. W., "Factors Influencing Students to Enroll at Indiana
_ State Teachers College," Teachers College Journal, Vol. 22,
(November, 1951), 22-3.

Haveman, E. and Patricia West, They Went _t_o_ Collegg, New York:
Harcourt Brace, 1961.

Havighurst, R. J., American Higher Education e_iph the 1960's, Columbus,
Ohio: Ohio State University Press, 1960.

Hemingway, Peter W., "Multiple Agreement Analysis," Unpublished
Ph.D. Dissertation, Michigan State University, 1961.

Holland, John L., "Determinants of College Choice," College ing
Universipy, Vol.35, No. 1, (February, 1959), .11-12.

 

, "Parental E'Xpectation and Attitudes About Colleges,"
9m §_n_d University, Vol. 34, (Winter, 1959), 2.

, "Student Explanations of College Choice and Their Relation
to College Papularity, College Productivity, and Sex Differences,
Collegg‘ and Universit , Vol. ‘33, No. 3, (Spring, 1958), 313-320.

Hollingshed, Byron S. m Should §_9_ 12 College: New York: Columbia
University Press, 1952.

Iffert, R. E., Retehtion ad Withdrawal _o_f_ Collegp Students,
(Washington, D.C.: U.S. DepaMent of Health, Education and
Welfare, Office of Education, 1958.

Johnson, T. 'W., "Major Inducements in Choosing a College," Education,
Vol. 52, (September, 1961), 27-8.

Kahl, J. A., "Educational and Occupational Aspirations of 'Comon
Man' Boys," Harvgrd Educational Review, (1953), 23.

Katz, Daniel and Floyd H. Allport, Collegp and Life, New York:
McGraw-Hill, 1933.

, Students', Attitudes, New York: Craftsman Press, 1931.
Kerr, William D., "Student Perceptions of Counselor Rolein the

College Decision," Personnel ing Guidance 'J rnal, Vol. 13,
(December, 1962), 4.

 

 

 

111

Lingoes, James 0., "Multiple Scalogram Analysis: A Generalization of
Guttman's Scale Analyses," Unpublished Ph.D. Dissertation,
Michigan State University, 1960.

Lincoln, R. L., "College Bul<g'e#,!" College and University, Vol. 3,
(Spring, 1961), 338-42.

Lippsett, L. and L. F. Smith, "Why Students Choose A Particular
College," College and University, Vol. 27, (January,.1952),
264-690

Lovejoy, Clarence E., fig You're Going pp College, New York: Simon
and Schuster, 1940.

, Lovejoy's College Guide, New York: Simon and Schuster, 1962.

McGranahan, Robert S., "A Study of the 1949 Iowa High School Seniors
to Determine Their Future Plans After Graduation," Unpublished
MA.A. Thesis, State University of Iowa, 1949.

McLaughlan, John M., "Extent to Which Pittsburgh High School Seniors
Possess Essential Information About the Colleges They Plan to
Attend," Unpublished Ph.D. Dissertation, University of Pittsburgh,
1954.

McQuitty, Louis L., "Elementary Linkage Analysis for Isolating Orthogonal
and Oblique Types and Typal Relevancies," Educational and Psycho-
logical Measurement, Vol. 17, 1957, 207-229.

, "Hierarchial Syndrome Analysis," Educational ing Psycho-
logical Measurement, Vol. 20, 1960, 293-304.

_____;_, "Typal Analysis, Educational _apg Psychological Measurement,
Vol. 21, 1961, 677-696.

, "Rank Order Typal Analysis," Educational ing Psychological
Measuremept, Vol. 23, No. .1, 1963, 55-60.

Medsker, Leland L., "Factors Related to College Attendance and
Performance .of High School Graduates," Paper presented at the
annual meeting of the American Educational Research Association,
Chicago, February, 1963.

Mueller, Hilbert, "An Inquiry into Factors that Influence Beginning
Freshmen to Select the University of Missouri ," Unpublished
Ph.D. Dissertation, University of Missouri, 1955.

Newsom, C. 'V., "Higher Education and Student Needs," chool gn_<_1
Societ , V01. -89, (March 25, 1961), 144-45.

Olsen, Hans 0., "Elementary Education: Students Reasons for Selectin
Their Colleges," Journal p; Teacher Education, Vol. 10, '
(September, 1959), 315-18.

 

112

Payne, Stanley L., The Art g Aghing Questions, Princeton, New Jersey:
Princeton University Press, 1951.

Plante, Alexander J., "An Analysis of the Plans for Post-Secondary
Education .of 1959 Secondary School Seniors in Connecticut and
Certain Factors Which Influence Their Plans," Unpublished Ph.D.
Dissertation, University of Connecticut, 1960.

Reeves, Ellis M., "Which College3"AJourna1 _9_f_ Higher Education, Vol. 3,
(February, 1932), 7-14.

Reinhardt, E., "Reasons Given By Freshmen for Their Choice of a
College," School _a_nc_1_ Society, Vol. 47, (April, 1938), 511-12.

Riesman, David, "Varieties of Excellence," National Association of
W_pm_e__n Dean___s and Counselors Jou___rn__a1, Vol. 21, (March, 1958),—
99- 114—_—

Robertson, "Who Influences Freshmen .to Choose Goucher?," Journal 9_f_
Higher Education, Vol. 7, (May, 1936), 271-72.

R0per, Elmo, "College Ambitions and Parental Planning," Public Opinion
Quarterly, Vol.25, (Sumner, 1961), 159-66.

, Factors Affecting the Admission .2; High _Sghool Seniors "_Tp
Collegg, Washington, D.C.: American Council on Education, 1949.

 

Salem, 0. A., "The Need for Guidance in the Choice .of a College,"
Virginia Journal 9_f Education, Vol. 36, (January, 1943), 174-75.

Sanford, Nevitt, The American Collegg, New York: John Wiley and
Sons, 1962.

Saylor, Robert H., "The Association Between Selected Pre-College
Experiences and Subsequent Academic Achievements of Freshmen
Who Entered the Pennsylvania State University in September
1957," Unpublished Ph.D. Dissertation, Pennsylvania State
University, 1959.

Schloemer, Clarence 0., "Why TheyChoose Your College," Colleg_e_
Public Relggions Quarterly, Vol. ‘4, (July, 1953), 22-28.

 

Schosteck',‘ Robert, Th___e_' Colleg_e_ Finder, Washington, D. C.:‘ B'nai
B' rith, 1955.;— .. ,.

Sellitz, Jahoda, Deutch, and Cook, Resegr rch Methods _i_n _S____ocial
Realhtions, New York: Holt, Rinehart and Winston, 1959.

Siegel, Stanley, Nonparametric Statistics for the Behavorial
Sciences, New York: McGraw Hill, 1956.

 

 

113

Standing, Robert G. and Clyde A. Parker, "The College Characteristics
Index as aneasure of Entering Students' Preconceptions of
College Life," Unpublished manuscript, courtesy of Dr. Buford
Stefflre, Michigan State University, May, 1963.

Stice, G., Mbllenkon, W. G., and Torgerson, W. 8., Background
Factors ghg College Going Plans Among High Aptitude Public
High‘School Seniors, Princeton, New'Jersey: Educational
Testing Service, 1956.

.________,‘hpphgppph§_Factors Relating hp College Plans and College
Enrollment Among Public High School Students, Princeton,

New'Jersey: Educational Testing Service, 1957.

Stouffer, Samuel A., pg. gl., Studies 22 Social Psychology, Volume
£X,‘Measurement and Prediction, Princeton, New'Jersey:
Princeton University Press, 1950.

Thistlethwaite, D. L., "Fields of Study and Deve10pment of Motivation
to Seek Advanced Training," Journal pfiiEducational Psychology,
Vol. 53, 1962, 53-64.

, "scholarships and the College-Going Behavior of Talented
Students," College and University, Vol. 34, 1958, 65-73.

Townsend, Agatha, College Freshmen Speak Out, New York: Harper
and Brothers, 1956.

Werner, V. E., "High School Counseling for Private Junior College
Admissions," Junior College Journal, Vol. 32, (February, 1962),
352-54.

White, Clyde R., "The Demands for Higher Education in the Cleveland
Area," College app University, Vol. 25, (April, 1950), 432-36.

Wise, W. Max, They Come for the Best of Reasons: College Students

Toda , Washington, D.C.: American Council on Education, 1958.

Wolff, W. H.,"Better Counseling and Selection for College," Journal
.piiEducational Sociology, Vol. 29, (January, 1956), 216-19.

Woellner, R. 0., "Counseling on College Entrance," School Report,
Vol. 58, (October, 1950), 381-82.

 

 

"~1_

 

 

 

 

114

APPENDIX.A

Items Received from the National Opinion Research Center,
Survey 412, together with Tables of Responses

al_

.a.‘ .F' '4:

 

.ucmoamwsmwm
mums muasmou mfimhamc< cowuoowam> ecu Scans a“ msouw omosu new case mace ems mfimhamsd coauooflam>ummouo "maoz

 

 

 

 

 

 

He no mHmuoH wma ma mm mm mHouoH
m A on. 2 mm m e on 2
U 2 A mm a «m m w mm a
1 o A we o «N m n ma 0
o A S a mu m a 3 m
o A 3 4 3 o S 4
miw a m agency a m: m .mabouo
maoauooufin , momsommmm .
mMAm<H ZOHH<QHA<M
nomuoa m
a.mum\¢m uoHHmsm w
beam was» A

message 30% cases guess .oso pumped
o co no .Hoonom HoHHmSm m on no .oso menu we when one Meagan Honuocm ou mswom mo mowoso usoz can 50% MH «A m

 

moHQEmm scammeHHo>umaouo use monumeaao> now
mfimmflma< oneowmufino com .mocosvoum seasoned .oufimccoaumoso

MMH<ZZOHHmmDO AOOflUm MUHE Afldm

 

.00..I\~.~.usv-.an uaoV. .-sa-.

 

 

 

 

 

 

.eoms mos :owuuouuoo .mouo» uocu eofiwwcwems

 

 

 

 

 

 

 

we.a he NN m-a
.a.a , Na an N-e

a Na 73% z we 2 Easy as N Na 2 S O. m-. a assess
3 A N 2 an m 2 NH e m N z

D. u S a on a S 2 a N H a.

S v S o «N e S N m e a o

in; as m a; S A 3 a N e e a e N a a a
as R 2 To S v a a 3 e m m a
a Na fig 4 To a a; Eases N a a e m N H amaleam

meofiuooufio noncommom

 

 

mMAm<H onfiddHA<>

hop m euro: ouos no n

how m mason N moons

% .N-H\Hm mom a muse: wH moons
Mme o “son A unocm

hum m noon w,u=onm

hoe m woo: w menu mmoH

econ umosam no coo:

Hde'mxou

masonom oewmuoo xuoamsos wagon cause 30% on «owmuo>m oau co .osfiu cuss so: NH m

 

mofimsom noauovflfim>nmmouo can doaumwefim> new
mfimhflmc< oueovmnfiso one .hososoonm uncommom .ouaocsowumood

mmH<zonemmod aoomom :3: Add

 

 

117

 

 

 

 

oH. me a a
a x < eoz e
see.N om oH a-a
ONnv on a N
a

 

oHese sass m-aau

 

mmam<fi ZOHH<GHH¢>nwmoMU

 

 

oe ma aHsooe
as aw mN z
«H A aa a
oH _A nH o
oH an mH a
oH “v m a
a-m a a
mCOHuooHHn

 

mMHm<H ZOHH<QHH<>

 

 

 

 

x.o-n\um moumum eouasp one :H use aHocHHHH

wMH H «N on an mHouoH
mm m w HN 2
an H o n wH a
mu m w m u
on N n mu m
NH H 0H 0 <
mHouoH N o m w n mmsouo
noncommom
me m mH no me mHmuoH
mm m mH MN 2
on H m HH mH a
cm H m 5 MH 0
mm m m mH m
0H H m m m <
meaoH x o m w n mmoouw
noncommom
moumum woan: on» ochuso o
oonuso m
mHoaHHHH 5H use zunsoo mHnu oonuoo w
mucooo no neon mHHu cH n

wcuon 50%

one: some:

 

monsmm oOHumcHHm>ummouo_com mOHuoeHHm> now
mfithm=< oumovm-Hnu use .mosoououm uncommom .ouHoocoHumosd

mMH<ZZOHHmHDU AOOmUm mDHm 444m

:- a- - I,\-.a~ss - n L. me?- . .59

 

 

 

 

a!“

 

.- . C: .. .2.< .iL ..~:Cmv.lfi~_u 1:.» p>UCev3cs$LHnH

‘nAvAI-un-‘u

Hal feelers...“

Itwv. I.

I l\.-.

a «UL h. maze-0 N J mu...“- :0

 

 

 

 

 

 

 

 

 

me Hm mHouOH «NH m 0H 5N NN NN MN NH NH mHmuoH

oH Au NH z mm m m N CH N N a e 2

HH A NN n #m H m w e n o m e a

w An 9H 0 «N m n m o m N o

N-m o in oN m mN m H N m m m N m

onw . oH “V o 4 pH e o H H N N <
0.» a Nun mHmuoH N o m m N c n e m mmsomm

acoHuoouHo . mwmmmmmwm
muHm<H ZOHH<QHH<> .

coax u.coe o

omoHHoU nouns Hoocom Hmconmomoum so Hoosom cumsoonw couscous m

oonHoo ooeaHeHe a

x.o|m\m¢ omoHHoo osom N

Hoonoa ach ooeaHaHa e

Hoonom anH osom m

Hoocom ommum eosoHsHm o

Hoonom spasm osom m

No>m£ uoauow pooh ch :oHumuseo Hmsuom nuns so:

on m

 

 

moHasom soHumpHHm>ummouo woo coHuowHHo> How
mHmszc< cannon-Hen ecu .mocosooum uncommon .ouHmosoHumona

MMH<ZZOHHmMDd Hoomow mUH: Ha<m

 

119

 

 

 

 

me H NH ON NN mm NN NH 0 mHmuoH

Nm m N o e m m N z

Nm H m e e m e m N a

NN N N N o o m o

seNo. Nw oH Num on N m e mH m N m

.m.c em N Ouw NH m q e m N H <
a x < uoz e aHsooe x o o a N e m e m aaaoeo

mHan mumndmaHao monsommom

 

mMHm<H ZOHH<QHH<>:mmomU

sosx u.coe
oonHoo momma Hooaom Hmconmmwoum so Hoocom mumscmum couscous
omeHoo coLmHaHm

x.o:m\mq monHoo msom
Hooaoa amHa ooeaHaHH

Hoonom LwHa oEom

Hoosom oomuw oozchHw

Hoosum momuw osom

MN'YLfiONwGO

No>m£ Moscow H30N eHc COHumosoo Hmsuow Lose Bow

om

 

mmHQEmm GOHHmeHHm>smmoeo vow coHomeHHm> How
mHmNHmc< mumsvmaHao one .Nuooovoum omcommom .ouHmccoHumoso

mmHmzonemmod Hoomom :oH: Ha<a

 

 

*wm.m mN o o m
oH.uv Ne oH a N
a Nx uoz <

 

 

 

 

 

 

 

 

 

Nm mm mHouoH me H H q oN HN mm HH «H n oHouoa
NH an mN z aN H m OH «H m o 2
NH ANN n on H m m H N m N a
3 A H o eN H e m OH H N H o
N An wH m mN N m N nH N H N m
0H UV 0 < oH H o m m H N <
muN a cum mHouoH N o m m N o m d m mmsouu
mGOHuuoan noncommom
muHmcH onH<nHH<>
sons u.soo o
omoHHoo Houmm Hooaom Hononmomoum no Hoonom mumsomuw couscous m
omoHHoo soaaHon a
x.OIM\N¢ oonHoo osom N
Hoeeoa amHa soaaHaHe e
Hoonom smHH osom m
Hoonom oomum oonHon q
Hoonom steam osoo m

No>ms cosmos “ooh ch soHuouseo Hmsuow some sex

Nm m

 

monEom ooHumeHHm>ummouo can cOHumcHHm> now
mHmNHmo< mumsvm-Hso ecu .Nuoosooum uncommom .ouHmccoHumosa

HMH<ZZOHHmNDd Hoomom muHm HH¢h

 

 

 

 

 

In..|.-\.

.I-oaI-u.

 

 

1L
2
1L

 

 

 

mnH m oN Na Na NH eH o aHoaoe
Nu a HH a N m H
Nm o 3 H m
NN , N N a e m
aee.NH om m e-m on H m N eH e H m
Hoo.uv Ho oH o-N NH N a N N H
a» aHaooe NH o o a N a m e m eeaosu

NM 4 uoz <

 

oHese oaomomreeo

 

amHaae onaaaHHas-mmomu

coax u.coo

oonHoo Houwo Hoonom Hmsonmowoum no Hoonum oumnomuw couscous
omoHHoo poschHm

Necrmee oonHoo osom
Hooaoa ewHe osaaHaHa

Hoonom stn osoo

Hoonom opouw ooanon

Hoosom ovouw osom

No>o£ Honuos Moo» oHp GOHumusvo Hmsuom nous sow

fix'I'UWVONCDQ‘O

Nm m

 

eonsmm :OHuooHHm>ummouo can aOHumcHHm> you
mHmNHmo< oumovmuHco ecu .Nososooum oasoamom .ouHmcoOHumood

mmHHszoHemmod .838 no? Hana

 

 

a" 'I

 

17rd; _

“lg-f ‘1 ..' "

 

c I {‘0‘ I‘ll:-

UI 0- - (7595A v a .n In}... .‘-

ll IllIUILQIIlEiId I:-

.I‘-I U-m1v10a

I.‘ Ii I‘liiii-l I‘ All

noise..-

.n..l\ll

 

122

 

 

mnH N. mmH. m mHmuoa
mm H Nm H z
«m H mm a
«N NN N 0
MN nN m
0H 0H <

 

 

sHsaoa HH a N H -amaoae
noncommom
manna onHflnHHHH

on, N

h.N|H\mq
ooh H

Nmousuoeom no coupo>Ho mucouom.u=oN ou<

No b

 

aoHcsmm soEoeHHoNTmmouo Hess. soHumpHHg How
32323 333.20 Hess .Ngooooum uncommom .ouHmosoHumosd

NMHdZZOHHmuDd Hoomum mOH: HH<H

 

 

 

 

[Viahbl U I

Nem.N om w NuH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No.V 8. N N
a NM 4 use a.
oHNsH. oaaaoNHHHm
mml as 338
HN V NH lie
8 V NH N
a, A NH 0
HH.N No. oN N-N «NoN.N NN NH N-H N an NH N
oH.V. NN NH H oN.V NN H N N V N a
a Na Nooz N a Na <eoz < NN NH w
Candy UHQflMWIHfiD . . QHQMH OHMDWmIHSU maowuumuwn—

NoH NN aHaooN Na NM, aHaoeH NNH N NN NN as sHoaoe
B NN V N» z NN v «H 2 an . N 3 NH 2
l .N V N N 2 V NH N 3. N N NHNH N

HN v N 0 NH v e o 1N N N NH 0

NN v N N NH V N N NN H N N NH N

N H N 4 NH V H a NH. N H N <

NH mlN NH N N aHsooN, N N N NA H Nassau

mmOH u 00 HHG mac.“ u 00 H.2— mflwfiomm 0M—
NaHNfi. zofiafifis
Norse N
N.N-HNoN oHHeaooo. N

uaouuououm H.

NconHH u pooh NH.umH3. nozm

NoHHHst. 533327305 coo coHuNeHHNNr .Ho
NHNNHNc< oumovmuHco one .Nocooooum omsoamom .ouHmscOHuNood

NNHazonNNNHHo .838 Non Hana

 

 

 

 

 

.‘q-wr

  

 

 

aufiv

.nevso-nu

V‘s.

nous.

 

 

124

sHa.¢ ooH HH NuH
mo.uv eH o m

 

o NM. <uoz a.

 
 

manmh.oums.mwwno

wH.H mm mH muN
.mp: No NH H

 

a Na <Hoz <

 

 

 

 

 
 

 

oHase oaoaoN-NNm. NNH. H oN NN NN aHsooe
NN HV N o HN z
NN, N a NH N
NN N N NH 0
aNe.N NN NH N-H oN N NN NV
.a .e NN H N NH N H NH. 4
e Nun 4 ooz < NHoHoe o N N H amaoao
oHan one: arena Noncommom
NNHNNN NNHNNNHH<>-NNon»
eaHson. N,
N.N-HNNN oHHoNeoo N.

ucmuoououm H

NsonHHoH HooN NH umHB co m

 

Nonsom soHumvHHm>uNNouo coo coHumvHHm> sow
NHNNHNo< opmowmuHeu coo .Nocooooum omcoammm .muHmssoHuNood

. Obdflaflfi CONJQCNNG>IWEOHU USE CONUUUHHT> HON
342%4 Nor—1V $w-~3-.-UWlfia—O Urn-w akfiUnuUuauuhruhm Concave-muss: aso.H.—.~H...~:..CHUta-do
in Us a [\YLQLI- - .s 0.7... -\¢

.onvli-mu.l‘h nuns-1. .I-II\...N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

me m NN em mN NHmuoH
Nm H. o OH ON 2
Nm H a NN n
NN H m w m o
smHo. Nm N Nun on H m m «H m
.N..s mo OH 1N NH m N 2 <
o. NNH 4 uoz < NHmuoH N o m e museum
oHnoH ouosdmano _ Noncommom
NNHNNH onNNNHH<>-NNONu
an mN NHouoH NNH N m NN NN mo mHmuoH
m A HN 2 mm H H m w HN z
s. 0H An «N a an m m «N a
We. HH A H, 0 «N m m H o
sNN.N Ne oH Nun a An mH m mN H N m mH m
ONJUV MN, 0 Cum oH UV N 4. NH N m m. o <.
e Nx < oez < N-N \m o aHaoeN N. N N N e .Neaoso
oHcmH mumsomuHcU moowuooan Noncommom
NNHNNH onNNNHH<>
uo>ocl N
cocoa a mono menu mNoH N
K.Nn¢\om cocoa o mosHu m on H m
zoos Nuo>o e
NHOHSSU moouuo.sON om coumo son no M
\II

 

 

 

Nmesmm ooHumcHHN>nNNouo New ooHompHHm>.nom
NHNNHNN< oueoomuHeu New .Nososooum Nacoamom .NHHNacoeuNoso

NNHazonNNNNN HooNoN NNHN HH<N

 

«.cv-orxn..o.:-q o......Q.-‘ .-.-.o-.H.‘N o-N-Noe .u...I\.~

 

 

 

 

 

126

 

 

 

 

 

 

NN NN N HN HN mHouoa cm H NN NH m HH mHmuoH
NN o H N NH z HH N N N H 2
HH 0 H q n NN H N N N m a
HH N H m N 0 NH N H N N 0
NH w e m NH, N N H N N
CH m N N < e H e H <
mHouOH N o m N N mmsouo mHmuoH N o N N N mmnouu
mNomumomcommom
mNHN<H ZOHH<QHH¢5
umHucoHom 0Hsouo noeomou Hoocom o
Num:0HmmHs Hopes m
x.o|N\Nm ouoHnum moosmw NHHmnoHums owns: w
uoHHm uon umHuum Ho mmouuom N

mNom NHNHU

Non ou uses uses 30% oHsoz HOHHB accuses ooh mwnHHu omoau wo has on pHsoo 30% MH

we m

 

monsom :OHumeHHw>ummouo can noHuoeHHm> pom
NHNNHNHE 333-30 ES .5523; NNHSNNBH .NHHNHHHSHNNNHHo

NNHazonNNNHHo Hosea HHNHN HHNN

also. (75.1.... QI~.l§.-s .nnl-.I-..vcu sumptu-

 

 

 

I.

.

 

l.‘f\.

NNH N do. me mHouOH

 

 

 

 

mN H NN NH z
«N N mH 0H m
#N N NH 0
NN.N He NN w mN NH NH m
OH.UV em mH. N OH H N. m < .
o NNN z uoz z NHouoa N N N Nassau
oHomH mumsdmano Noncommom

 

NNHNNN onaaNHHes

 

Nufiouo>Hcs no oonHoo omnoH N
m.NnN\NH oonHoo HHusm N

Nomooao ooh pHsoB soHsk .NuHmuo>Hso Ho
monHoo owumH m can oonHoo HHmsm voow m on wcHow assauoa omooao on on: com omoHHoo ou wsHom one? 50% NH «HH m.

 

NNH N NH NNH.aHaooN
7 mm m sum 2
m" NN N N oN N.
eN N. HN o
NN N oN N
NH H NH 4

 

MHouoH o o m monouo

Noncommom

 

mNAm<H.ZOHH¢dHH<>

osos um o>HH N
o.NIN\NH oonHoo on head m

Nose: no o>HH vHooo ooh muons oonHoo
m.ou no oonHoo ou Noam ow eunuch soN pHsos asoHcoum on ones modes can oonHoo ou wsHow ones so% NH. NHH m

 

NNHNst.coHHNNHHN>-NNoHo New eoHuNeHHN> NNH
NHNNHNHE oumovmaHeb HEN .Nuooowoum oNooaNom. souHNosoHuNood

NNHszoHNNNNN HooNuN NNHN HHNN

 

EWHQH—Hfim COfiuEUfiHnwxwlmmOhU U26. CCquwT+NG<H L21;
nudnaxnd 33" LuHQH-HUMHIHNSU HVCE oxnUnHLiEULVuunn .UMIJCCQmuqun «Unni uw—HCOH‘JEQWHO

Halvfii,<7h7.-IvI.I,-Lm.‘.-‘i wicu‘duowv‘L InuvI-l ...-<..I

 

 

 

 

 

 

128

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

accoHpm HwHoom mmmvH wcH>HH z
unuvH HuHoom avamHum wcH>HH a
numvw HwHuom avauHum waN>HH o
mamVH HmHoom avamHuw wcH>NH m
HwHuom avamHum muocH + wcH>HH <
smudbh qusa. vcoumm\ umuHm «macho
uwmwxcmmlwd Mummysn
m mH N¢ «m HH. m an oq MN oH m mH H «H NN c ed N am mN meuoa
N NH .NH c 0H NH q o m N N mH mH. H NH. m z
NNequ.NmHNHo meomqumcmN .—
N c 0 HH N N w m N H m H NH N m m n q o
qu‘aH NH¢¢¢ HNNmH NNNH m
H N N H m e e o N H N H m ¢ w <
N, c N‘ N a N! N N N H NJ N N N H Nl e N N NV museum
nvcmHum HmHoom mmwxwm mumVH
mmmsomnum_
mnnm<a ona<nmm4s
N.¢uH\a mumouwuaH as manna 0:3 avawHuw 3w:
N.¢nH\m «NHH HmHoom cam mmHuH>Huum mamamu
N.¢uH\N mcH>HH m wcmea now GOHuuuuaoua 5
N.¢uH\o mmuvH 3w: mo cOHustaHum usu

Hq ou H scum samav
HHHB wcHsoHHOm onu mo scum xcHnu com on ucmuuomaH 3o: .wmeHou :H mummm unoN ou vuw3uow.md«xooq,

N30N cu up

ONHINHH m

 

NHNNHma< mumaumuHsu cam .Noamsvoum wmaoammm .ouHmcaoHummso

NuHNamm aoHumcHHm>-Nmouo can coHumvHHm> you

ESNQEfim COHUGUHAG>I3QO¥U Ufifi COAUQUNHQ> hDM
fluishd 351V rwrflz-HVWIHF—uv :35. .KUF.-nw:~umv.u..fl .vncr-CA—Znoz a.i.H.fl-~.~.0.Hdm~..uv30

I‘II‘I f\’§u...~I‘-I..I.u‘u .ouluvcu‘vi. auva-u Nu.l’\.AI

 

 

 

N

 

N ‘

w.

 

A.maauH Suon cu mmmaonmuu :mwN: was HHw .uuomwuwnu .uonHoo cu om
ou wchcan wuoB uuonoum mwnu aw .wnsouw HHw :H .muaovnum HH< .uuHmaaowummsd Hoonum swam wcHhmm mnu co
EwuN HmuwuawuH cm mo mumaoamuu :mwh: any auwz cucwnaou mums umwsh .kuou muua mumaommou :mom: unu NHsov

NHoonom st; uuumm omeHou ou om ou wqucmHm sch mu< mHH m

 

 

 

mMH 00 wN mHmUOH
9 Nm GH ma 2
n NN NH NH a
NN N ,MH 0
NN.H H¢ mH m on NH mH m
.m.a oN NH N NH N oH <

m NM 2 mum z mHmuOH m m N mmsouo

mmmaommom

 

 

mNHde ZOHH¢QHH<>ImmOMU

Nuwmuo>was Ho umeHou owumH m
a.wuN\mH ow0HHoo HHuBm N

Nwmooso no» vH503 sown? .Nuwmuu>aca Ho
uwMHHoo uwumH m can mmmHHoo HHmEm voom m ou waHow sousuwn mmoosu cu was was umeHoo ou wawom mama so» «H ¢HH m

 

\‘

mmHaamm.=oHumvHHa>ummouo new :oHumcHHm> you
NHNNHNHE Swag-Eu was $2553; oncommmm .wuHmHEoHumosc

NNHszOHHmmpc Hoomom mon HH<N

 

ax—Cpuxm Funuu‘flwFiT—nnkfllfllnur‘thr.‘ "Kr-a.- (In.n.t¢u-.. .1FI1-v . L
H: #33 H” 21.-6‘ ava- ..w:_umwl._‘.—U Hut-4 nxAnJHNUNHquyHLH 33:02:“qu- flhvhflavcnuafi d «um-:0

.l.'I I‘<§.I7-K'I" {.I.bu~l .hllIIIlV..lI,I II\II.I n'.-’.\,.\I

 

 

 

 

"3"" 1"

 

 

130

.ww.N Nm mH muN mm.m no. NH anw

 

 

 

 

 

 

 

 

 

 

 

 

 

NH.V NN NH N NN.V NN NH N
fiNx z 32 z N Na Nuoz N
NHNNN NNNNNN-NNN . NHNNN «NNNNN-NNN

HN NN NHNNNN NN NN mHmuoe NNH. NH. HN NN. NN. NHNuoN
NH V NH 2 NN V NH z NN N NH N NH 2
N A NN N NH A NH N N N N N NH N
NH V HH 0 HH V NH o «N N NH H NH. 0
NH A NH N NH V N N NN H NH N N N
N A HH < NH V N < NH N N N <
N-N N N N-N N N NHNNoN N N N N museum

chHuumuNn maowuumuwn mmmaommom

 

NNNNNNH onNNNLmN<>

wawxooH umun.

m
o.muN\oH ufimcSum umon m
ANHHNw u uwwmoHumosov‘ oumHsum.umum N

Nnufiz use
ow umnump som.vHso3_mco soap: .waHu :N AHuva Non wawxooH umwn «nu Ho .mmmHu :N uamvSum ummn may NH
onsVAHuva Non N no .AumvmeummSU a.mHuva muanum umum m umnuwm nuwa use ow ou mucmnu m can 30% omomadm mNH m

 

NNHNaNm NNHNNNNHN>-NNNN0 New NNNNNNHHN> mom
NHNNHNN< oNNNNm‘Nno Nam .Nocmsumum NNNNNNNN amuHmaaoNuNNsa

at".II‘VII’I‘VI.I.O~.(|-~u .IICAUIIAIIH U-uIIQ- «I‘Il\ul

 

 

 

 

 

 

 

000. Nm HN wuN

 

 

 

 

 

 

.Né NN NH N
\m NM. 2 Noz z
NHNNN 98%;.6 NNH m NN NN mm .338.
NN N NH N NH 2.
1 NN NH N NH N
m... NN H NH N N N
NN H NH N N N
NNNN. NN NN N-N NH N N N <
.3. NF NH N U a
N Na. N 32 N mH33 N N N N 3:95
.anmH mum=dm:N:U mmmdommum

 

mflAm<H chH¢GHA<>ummOMU

wcwxooH ummn m

a.muN\oH uamcsum ummn m
AmHuNw u uwcmmHumw£UV oumqum Hana N

mnuws uso
ow nonumu 30h vHaoa mac sown: .wmmHo :N NHHNMV Non wdHJOOH umon mnu no .mmmHo a“ unmvaum uwmn onu NH
0&3 NHunv Non a NO .AumvmoHuumno u NHHNmV muwHaum “mum m HmauHm saws use ow ou manage m an: 90% mmoaaam NNH m

 

NNHNNNN.NNNNNNHHN>-NNNN0 NNN NNHNNNNHN>.NNN
NHNNHNN< NNNNNNINNU Nam .NNNNNNNNN NNNNNNNN .NNNNNNNHNNNNQ

NNH<ZZOHHmMDd.Aoo=Um mUHm 444m

 

u .u- Ilr\ 1!... net... {(1.1 P CNN 11“.”.

-.. . . . . .
ENBNANNNCAV nah-.wHLUmuIfiN—U 151.. .xfiU—Nfla—UShh TNDCOANmmfiz .«ehfiswrN:OfiUIU:Nv

4.!qul.‘7-r~av-N- nu-...-u¥u .uAvAi-NN-IL nuwu-un .-.-I\..~

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH N NN NN NN NHNNNN
NN N NN NH 2
NN H N NH NH N
NN N HH N N
NoN. NN NH N-N on N N NH N
.N.: NN NH N NH H N N N N
n NN n uoz n mHmuoH N o m N mmsouo
NNNNNN zNHNNNHN<>-NNNNN
NN NN mHNNoN NNH N NN NN NN NHNNON
2 NN V NH 2 NN NH NH NH 2
nu HH An HN N NN N N HN N N
NH n NH N NN N NH N N
.NN.N NN HH N-N NH ”V N N NN H N N HH N
NN.uv. HN HN N NH “v N 4 NH N N N N
N NN Nxmmm N N-N N N NHNNNN N N N N mmsmmm
anmH mumdmmnflnu mcowuuwufln mamaommwm

mNHde ZOHH<dHA¢5

Hstaom umoa o
x.o-¢\oH mowuw>fluum ca uwvmuH m
NNNNNNN NNNNHHNNN N

Nun on uan 50% anoB wao cogs? .on A mwdHSu mwunu wzu we use How Hoosow um wwuonawswn up uHsou :0» «H HNH h

 

\‘

NNHNst NNHNNNHHN>-NNNNU Naw NNHNNNNHN> you
NHNNHNN< NNNNNm-H:o Nam .Noaosvmum NNNNNNNN .NNHNNNNHNNosc

NNH<zzNHNNNNN NNNNNN NNHN NNNN

 

udmuxn _. NNC< SHEIHVWINN—NU _u: «a nNAUCfiZUQLGN 32:32.32: - 3... N 2.2:..N ,Nl.....c
..-.I in I I‘vflurhh u... ,,- mhnoi. uNi .hnv-vunvu‘... .116. - N :L I 1.~A~!..

 

 

 

 

 

 

 

 

 

 

 

 

Inmmml[ N NH NN NN NH HN NHmuoN NNH N NN NN NN NN NN NHNNNN NNH N NN NN NH NH N NHNNON
NN N N N N NH 2 NN N NH NH N N z NN NH N N N N z
NN H N N N N NH N NN H N N N N N N NN H NH N N N H N
NN N N N NH N NN N N N N N N NN NH N N N N N
NN N N N N N N NN N N N N N N NN NH N N H H N
NH H N N N N < NH H H N N N N < NH H N N N N N

NHMuOH N m d. m N H wmmmuu mHmuoH a m N m N H mnsouo mHMuoH NH m N n N H mmnouo

m>HuDowNM mmvdwmsm uwfiucmwom HUuHMB.
mwmcommwm

 

mmAm<H ZOHH<§HA<5

N.muH\mm u>wu=uuxm mmwaHmsn
Nan..H\mm , . umHusuNom
N.mnH\Nm umNHmsusofl no uwuwua

Aumwnwwn u H mmnH xammv .NuNHNnmuHmov uHmnu mo mauou :H mnONuwasouo m>Hw wcwonHom anu xamm omuNm m

 

NNHNeNm NNENEHNNTNNNNU HEN NNHNNNHHNN’ you
NHNNHNEN 335.20 NNN .NNNNNNNNN NNNNNNNN nNNfiNNHHNHuNNNa

NNHszNHNNN N NNNNNN NNH: NZHNNN

 

\<¢\ llllll I«l llvllllll IIIIIII““LV IQHIICII ,lIz-Vzt‘.’l¢

 

 

 

 

 

134

NNH H no oN N mHmuOH

 

 

 

 

 

 

 

 

 

 

Nm NH ON 2
Nm NH mH HH
NN w NH H o
*OON.H no m. w on NH NH H m
.m.: «m HH‘ m NH H HH m <
n‘ NM < uoz < mHmuOH o N m N unsouu
mHnma onwaumano noncommmm
N53 onNNflJNNNNoNu
NNH H «o NN H meuoa
mm. mH ON 2
«m HH MN m,
_NN H 0H NH 0
*Nwm.N No m w mN mH HH H m
NH.V NN HH N NH HH N N
a NN < uoz < mHMuoH o a m N museum
mHan mum=UmnNnu mamaommom
muHm4H onH<aHH<>
.mmnmnu anoa H mHuuHH Num> mH mumnu “08mm «surname huw> hmum on waH v.H m
.NamumHmEoo uoa usa .wwcmnu cu meH c.H «mafia» hams mum «mesa m
o.auN\oN .NHoumHmEoo «mango cu meH c.H mam H Nm3_mnu wHoNNE meH u.cov H N

NMHMNusoN usonm mwaaHmwm.uzo> ou ummmoHu mwsoo sows: Nuommuwo wnu sumac ooH m.

 

mmHnEwm aOHumuHHm>uNmouu can aowuvaHm> you
NHNNHmc< onwavmaHso cam .Nuamucmum mmcoamwm .muHmacoHummso

adqczzaqannny Abanun :34: uzqzhn

.‘I ...I <l4.ltallII.l.!.cIIlll llll .‘I,

 

 

 

"

 

 

 

wnH m e H «N \me meuOH

an N H o «N z

«n N H o mN a

«N H N HN o

mN N N c nH. m

6.. NH N NH <

1...
mHmuOH x m m N o m mmsouo
numaommmm
wnnmdn ZOHH<QHH<>

oumo u.angos NmSu m

oumu u.:vH=oB menace .vaoua wa cHsos uwsumw w

a.mum\mH uumo u.avH:o3 nusuwm .vsoum ma cHsoa umnuoa N

vaoun uHuuNH « Huuw unwws Nufiu c

ma mo unoun Nno> an vHaos Auon m

ArNMOHon :H ucmumwmmm «nu ma ou mouwm ou cwvwowv um: DON
«magnum 302: "muwmndowummsu chHMHpo msu so GOHummsv mSOHbuum unu ou mawuuumomv Noumo Nunu u.apoo3 no
.aoN mo vnoun mucwumm Mach uxma vHsos umnu wuwnuwaom ma uw vHsoB .50» cu vwnmmaw; away OMHH wawnuosom NH NHH m

 

mansum :oHumvNHm>-mmouo wcm soHumvNHm> pom
mHmNHm=< ouwsvaHau van .Noaosuoum uncommom .wuncNOHumusa

géogmupd .883 :3: quHHmm

 

“use: uuuoz

a~¢
N:

N .N Nae "Huauoz‘ououm ”NHH
N .o "aouHuuo
. m .m N .N .o "chomauum
63335 H258 2: you 3532:: N .N .N .N .N "Noon. 395:.
"x who? 23 23:0 nuaovsun nH :33 who: "mHuouHuu n .m .w .m .H "NNOGHHHH mo .225
1
A.n=oum one Nan new Hanna uawcaaov ozv .mnaouw vwaHnEoo ma munaomnou no “make xamm
NNHNHNHfiboum .N
amsounu uuw ou cum: .w
HuzuuuHHuuaH .N
annnoam .o
coauMOOH «UH: .m «awn uuuoz IIII N.OvH\Nn
msomeHmu .q Huauoz uuwum mwouwHHH IIII N.o-H\Hn
can we 30H .N NouHumo |.| N.N-HNNN
u>wucumxu .N vuoucmum IIIl Nae-H\mN
naumu uHuoHnum .H enuua mwoanHH IIII N.OIH\wN
NH .53 3:8 .N NNoNHHHH No in: I NATHNNN

.uwuHHoo was» we xaHsu sch can? 90%
cu musuoo uusu who? uuuwm.0£u uswwu map so umHH wan Eoum JOHQ .ume may as mowuHHoo usu no name pom MNHIQHH m

xIIIIIIIIIIIIIIIIIIIIIIIII

 

NoHNaum.:oHquHHu>-mmouo can :oHumvHHm> you
NHNNHwn< unusvmgwsu can .Nuaosvuum umaomumm .ounauoN

NNNNNNNNEN SE. 5; 35%.

amend

 

NNH NN NN .Huuoa

 

 

 

an HN NH 2
an NH 0H 9
. «N NH HH 0
‘nHw Ho HH w . nN HH NH m
.apa Nn NH N oH a. N 4
4'44le mag N N. N ass
«Hp-H mun -Hnu. m0¢¢omuum‘
afluuzu,chHaaHH4>N
Nuwuuo>aca we uw¢HHou umuuH‘ a
A.m-N\mN owuHHou HHnan N

. Nouaonu 90% vHaoB nous: .Nuuuuo>«aa
no uwuHHou omHuH.q vac uonHou HHuau voow a couauua ouco:u on can can .owuHHoo cu wagow «nu: :oN «H mNH m.

 

 

13?

NNH a. ONH‘oHauoH
mm m on 2
#m H. mm a.
QN H. MN 0
MN N NN n
mH H mH d

 

 

oHuuoa o. c n mmmmmm.

noncomwod

 

mMHm49_ZGHH¢GHH<5

use: an u>aH, o
o.onN\mN uonHoo ou Nwau n

N050: um 0>NH,anou.50N chug: omuHHao
m ou no omoHHou ou.Nwan om Husuuu :ON vHsoa,.aanoun a: muua.Noaaa raw .omoHHao,ou_waNow onus :ON.MH «NH m,

. 1“

NoHNaom aoHuuBngNNBo can 83322,. new

NNNNHua< uuuaumuwgo.vnw .Noauavoum oncomnum .ouaqauoaunosd

lllr L’Dll '

llnx ‘n‘lc.~|ll.l1uaIN-‘c -.I|I¢- I‘v Ou!1III uu’qdltvIQ\

 

 

 

 

~15“

 

 

 

 

 

 

 

 

 

 

 

No NN NHauon. NNH N mN o m HN wH mmw Nm \wfl mHauoa
MN V nH m. mm] H m m N m. N N 3 N z
m 1n NN a Nm H. m H N N N N m, N a.
. oH An NH 0 NN a m N N H m. o
NNN.N NN NH N.N..H .55 NH. V NH, N NN. N H H N N H N H N
3 mo. mm MN Nnm.uunm N A m < NH N H H H N N c H <
.L N Nm z.uoz z‘ N-N N NNN-H NHNNNN N! N N N N N N ‘m H NNNNNN
anwH mumsuwanm maOHuooan mamaommmm
NNNNNN zomHNNHNNN
:owuwosvo w awuwoum mmmchsn m N
3mHuuua .NuumHuaovuuun .mcHoNvoaumum N amuwoum ouaowum w m
N.msH\mH AmHun u mowaoaoom meonv musuH30Huwm o Sauwoum muuw HananNH m‘ N
AmHuNw.u waHmuaav. wdwuomcwwam w. H

.. N2: 825

N33» 8». H3322: NNH NK

H.mHmamm Hmmom can amawHoo oau aH vmnsHoaH an ou mEouH mama» sue: on o>HumauHmmw umuosmcu
m>w£ umaa mauHmaaoHumond HHmm so NHH EmuH mmmv NHoosum swan nouwm mmmHHoo ou ow ou,waN:=me :oN ou<_ 0N~_m.

 

 

 

Nona m :oHumvHHu>-NNouo_czm NNHNNNHHN>_NNM
NHNNHwa< uuwaamuasu can “mucusvoum uncommom .uuHNNNOHummsa

NNNNNNNN §§ NE NE”

 

VIII-<2751v-.-.ud.wv§-Hv .-AI-I-Hun.~ u-Nddl-n Nun/I.Q..INN

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hm mN mHauoH. mnHA NH NN Nn mm_ uHuuoa
NH An NN 2 mm m NH NH 0H 2
NH, An NH 9 Nm m HH mH. m 9
NH u, NH 0 NN n m m N 0
3 A 3 m nN N n HH N m
w. I m < oH H N o N <
«N N. N..H Nina... N N N H NNN.NIINN
maowuuouHa mumaommoa‘
mMHNNH chHNGHHNS
NnH\mm
.ou voHHmmm.mumeHou mo Nmeu ucmHQNMHv mo wonssz mm..Hoo NNH m
NNH m“ Nm 0H. m“ mH. NH HH NN NN mHuuoa
o, NN N m. n N H7 N; N m. n z
nu Nm H N m, n. m e N m N a
NN N H. N N N m. m o
NHw. Nm mN Nun on m N H m N m. N m m
.Né NN NH N.N..‘.HN NH N N N N N N.
a N z qu. z mHmuoH N m N o m N m N H maaouu
anmH «unnumaHgo mmmcommom
wMHNNH ZOHHNQHH¢>ammONU
aowuuoavm m sapwoum mmmaHmzn m N
3wHumum .Nuumwuamvnoua .oaNONvmsnmum N smuwoua mucmwom w m
N.wnH\mH AmHun I mowaoaoom msonv unauHauwuwm o Emuwoun muum HmuonNH.m_ N
HmHun‘u wchuscv wdHummcwwam m vmvwomva: H

NNvaum :oN HHN3_um£3. NNH.M

 

.3358 533327330 can. coin—NHH; you .
NHNNHNNN auwsvmnHHHo HEN £2530; 09393“ .3353

NNNNNNNNN EN EN 2.26

ummfid

 

me H mm Nm MH. n meuOH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NLN NH, HH N N z

NN NN N N N

NN NH N N. H N

NNN. NN NH N-N NN NN N N N

.3. NN NH N NH H NH N H N
N Nx N 32, N- N433 N. N N N N N955

«Hana mumadmuwsu Noncommwm
NNNNNN. zdmaNNmNN? NmNNN.

NN HN NHSNN NNH NN NN NH. N. 238..

NN V NH. 2 NN NN NH N z

o NN V NH N NN. NN NH N N.
M HH _.A NH N NN HH N N N N
NN.N NN NH N..N NH A NH N NN NH NH., H‘ N
NN.V NN HH N NH V N N NH NH N N
NTNN N 32 N N N N-N 333 N N N. N N NN595

anmH NHNDUmanu , macauomuan . . ,Numaommmm
NNNNNN zNHNNNHNNN

HHN N

. ”HOE HO NHQS w

N.N-NNN NHSH 523 23H N

03°C 0

Avamvfimnoo wamHHoo HHm NNNHV :mHmmu50N pom mmHuHHNAHmmom ma vmuovwmdoo uw>w
:oN m>w£ mmmeHoo pans: anH ou concummmmv. Namwm :oN m>ms NwmeHoo mumsu mo mwdonuuwo unu.mo Name 3am NNH m

NNHNamm.coHuNNHHN>-NNouo Nam NNHNNNHHN> you

NHNNHma< Numsvmano can .Nosmsvmum NNaoaNNN..ouchaoHumoad

 

Nfigcflmmzu HSHSN NNH: 322.5.

IUHCEQW CCflUQCHHE>ImmOhU VG: CCfldfiUfiHm> HON
£dean 9156‘ [vhznivmnlflfiu US: qxnUCQZHUUHnH .UZCOQQUNH ..a._a.H.n.wCCO..NU.0NU-.Nv

Vlflu-.<7LZav-.-1......I-NNv .nlifii-nwiws u-wu~.- 1d7u- Fin-1..

 

 

 

:u

 

 

 

141

 

 

 

 

 

 

NN HIN NHNNNN NNH NN NN NN NH £33.
NN V HH 2 NN NH NH N. N. 2
NH V NH N NN N NH HH N N
NH V NH N NN N N. N N N
NH V N N NN N N N N N
NH V N N NH N N N N N
N-N N N-N NHSNN N N N N N 3455

Naowuooan . Nowaommum
NHHNNN 203.353,

HHN

muoe_uo «Hm:

ONmuo\m NHN; away NNNH

Nvouwmw> SON m>wn mmmeHoo NNNSN mo mmmamamo man we Name 30m omH m

03 0G

 

NNHNENN NNNNNNHHN>-NNNNU Nam NNHNNNNHN>_NNN

NHNNHNNN Numsumano Nam .NNNNNNNNN Nmaommma..NuHmcaoHNN

«SENENNNN H85. :3 6225

mid

\ONmax

 

142

 

 

N N N.N.N N N N.N N.N H =NHNNNN=
N N N N N N N N mw{, H. z
N N.N N N N.N N N H N
N N N N N N N N N H N
N N N.N N N N N N H N
N N.N N N N N.N.N H N
NH N N N N N N, N .wxm, H mNaoNN

 

NH vmxH H Naomumm mo umvuo.xcmm

 

 

me ma mm ca NH NN om NN mOH HNH NNH mHmuoH
mm mN mH. NN N NH oH N wN mm mm 2
Nm 0N NH 0N; N CH N N on wN Nm a.
NN HH. m mH N N N m mH HN MN 0
mN NH . OH . mH N n m m NH NN mN m
0H NH m NH N N m m NH NH mH <

mHmuoH o m m N o m N m N H manouo

mmmcommmm .

 

mNHmNH ZOHH<QHH<5

N>Humusmmmumwu mmeHoo

umgommu umnuocm

N.onH\0N uonmasoo mucmvflsw
nomoo NHNNHnum

umnommu anHwam

pmnommu Noamaom no Nuwumamnums

umnommu mmHusum HmHuom.

mmeHoo waflvamuum mmoamuafimuvum no Nuamaum
Hoonuw um mum: mvcmwuw

mucmuwm

\DNwO‘O
HNMNm

Nuw.u=opm ou.vowau SON m>m£ onmn vmumNH wdomumm m£u mo £0N53 .mmeHoo udonm wcwxcwsu GH HNH m

 

NuHNaNm NNNNNNHHN>-NNNN0 Nam NNNNNNNHN> you
NHNNHNNN Numsvmquo Nam .Noaosumum NNNNNN N .NuHmacoN

NNHNNNNHNNNNN HNNNNN NNH: 323

umde

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH. NH @ omw. N HH mHH m.x\mH mww. NH oH mHH mHmuoa
NM m on N‘ n‘ Nm H_ n mm N N mm 2
Nn. N H Hm m Hm H m wN N n aN a
NN N. N NH. N NN. H n‘ ON n N NH‘ 0
mN N m ON H N oN n NN H N NN m
NH H mH H mH N NH H{ H NHH <
3 .338. N N N N LNW H N N N NH N LH1 3.38
.4
1 NNN.N. Nag . «NNH. H8 338. N
mmuaomaum
mmHde ZOHHNGHHN5.
Nuow vam on HHHB owwHHoo.u50N sum 0 m N
.oao: aouw Nata m>HH vHIOSN sch umnumazh m N H o NNH
.ou ow anosm 50% oonHou £0H£3N o m N N
.oxmu vHaosm 50% No=um ma amusoo umnz _ m N H. o
vmmmaomHa Wmlmmfl a
NN.N-HNNN - NNH N NN.N..HNNH - NNH N. 352% Econ

Nmuamumn‘unoh nufiz vommsomwv u.cm>mn soN mono nowas Nam .usonm
mmuwmmav 50% mesa squa .uaonw owuwm muauumm Hack can no» Nana 50Hga gouge .30Han.gaouH «nu mo 50mm pom, NNH.m

 

Nonamm aoHumvHHN>ummcuo cam GNHumvNHm>.uom
NHNNHNHHN mumsvmnHHHu HEN .Nocmsvmum 2325mm 63233325

2438333 .5023 mug ozammm

 

luv‘li—Iwnftd‘ Kilt-NN~—RUC'4A-J jun: OKJIINN‘JI4-J)‘ ‘ ’11.!) 112-Iu‘

MCI-I’V’nZn.U-.-.whfll.-i .u-v-Jn-uiah unuuh.‘i wvz I,UIII§|.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH N mm NN NN NH N NHNNNN
NN H NH HH. N. N H z
NN N N NH N. H N N
NN N N N N H N
NNN. NN HH N-N NN H N N HH N N
.N.: oN oH Nam NH H. N N N H <
N mm- N uoz N NHNNNN N N N N N N NmmNuN
mHnaH mumnumuHsU Noncomwmd
M Na; zNHNNNHN<>-NNNNN
N.
HN , NN. NHNNNN NNH! N NN. mm NN NH NH NHNNNN
NH w NH. _ z NN H N NH N N N z
N» NN uv N N. NN N N. NH N N N N.
“w NH -. NH N NN N N N. N! N N
NN.H NN NH N-N NH. my N N NN N NH N H. N N
NN.HV NN NH N-N HH “v N N NH N N N <1
N» Nx N Noz N N-N N N-N NHNNNN N N N N N N NNNNNN
mHan NNNQUMIHso «doauuuan mwmcommom
NNNNNH zNHNNNHNNN
NNNH N
NNN N
N.N-NNN NNN N
NNN N
QfiOfl m

Nuou Nam HHN3,NN>HNNHNH uo NHHENN usom.um£u Noumxm NON ow NNHumosnm wmeHoo usoN mo mmmucmoumm was: uson< mmH m

 

NNHmaNm NNHNNNHHN>uNNouo.v:N NNHNNNHHN> new
NHNNHNN< Numsvmanu ccw .Noamsvmum NNNNNNNN .NNHNNNNHNNNNQ

MNHszoHHNNNN NoomNN NNH: NzHNNN

 

viva.<9,-z-N-.-.1nq.—..NN .-.-n--Huut

 

 

 

 

 

145

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH N N NH NN NN NN
NN H N N N NH NH 2
NN N N NH! N N
NN H H N N. N. N N
NN.N NN HH N-N NN N NH. NN H N
.Né NN NH N-N_ NH N N N N N
N NNH N 82 N NIHBNM N N N N N N N385
mHnaa oum5 wawnu Nuadommud
NN.HNNN. zNENNHHgNNNNN
NN NN NHSNN bmH m N NH. NH_ NLNF NN NHSNH.
S A NN z NN N H N N NN N 2
NH A NN N NN H N N. N NH N N
HH A NH N NN H H N N HH. H N
.HN.N NN HH N..N N A NH_ N NN H N N NH- N N
NH.V HN NH N-N N A NH. N NH N N N 4
NW NNH. Nuoz N N..N N N..N flmuom N N N N, N N LNIIN NNNN
NHMMH ouw5mw|H£o acqmmuman Nuucommom
NN.H; 2033349
NooH. N
NNN N
N.N-NNNH NNN. N
NNN N
UHHOG m

Nunon uaHuuuumn can Nwchumm umaE5m scum %ma ou uomnxw 50% 06 owuu5oouwm Nuns 95094 NNH.m

NoHasmm 50HNNNHHN>nNN0uu New 50HumvHHm> you
mHm%Hw:< oum5vmmH£o van .%0505vmuh mmaommum Nuquscowumm5d

EEOHHmm—Dd AOOHHDm SUHHH UZHMNHm

Vii-(VLZAv-:>-,ntn...--v .IAVAi-QHIIQA -1v-- NJz-zn-n

 

 

N5

 

146

 

 

 

 

 

 

NH NEH mum5umuNIHle.

ON.H NN wH NuN

.N.5 mm NH Huo.
m NN a uoz n

NNN.N mm NH N:N

.Nya mm m Huo
0 Nx < uoz <
NHANH muw5umnmmm.

.om.m NN oN NuN

oH.V Ho NH .Huo
N NW N 82 N.

Noo.m mN mH NuN

Ho.uv NN m Huo
N Nx N 32 <

 

 

 

 

 

 

 

 

NNH NH. NN NN HN NH NHNNNN.
NN N N N NH N, z
NN N N N NH! H N.
NN N N HH N N
NN H N N NH. N N
NH N N N N <
NNHNNNN. N N. “Nin N 3
Nommdmuum
NN.H; onNEumHgNNNNN
1mm NN NHNNNN NNH NH! NN NN NN N NHNNNN
NH A NN z NN N N N NN N z
NN V NH N NN N N HH, NH! H N
N. A NH N .NN. N N NH N N
N A NH N NN N H N NH. H N
NH. V N N NH N N N N H N
NuN a Huo, mHmuoa N m N H o Nm50uu
mmdwuoouwa, mandamem,

 

NN.H? NNN.NNNHHN. N.

.ou NNHHQQN mowwHHoo m0.qua5z mm .Hoo .nomH m

 

NuHmawm aoHumvHHm>1NNNuo can 50HNNNHHN>.NNN

NHN%Hm5< ouw5um&H£U Nam .%o:o5uuum oncommmm .ounadoHumo5d

MNH<zonNNNNo Hoomum NNH: NzHNmN

147

 

 

 

 

 

NNH NN NH. NN NN .m» NNH NHNNNN
NN NN N NH N! N NN z
NN. NH N NH_ N H NN N
NN NH N N N NN N
NN NH H N N NN N
NH N N Ne N N HH N
NHmuoa .m.d % x .m.5 .N. x ma50uu
.moHonU vaN . . mowono umH
.Nwmmmmmmmr

 

NNNNNN zNHNNNwNNN

mowosu vaN 05 no oHanHN>w N05 m5
vmuomhwu %
wouuHSNN x

.mouam ouuHuaouaau 0:» 5H.
.vuuownmp 500a 0>.50% nnga um Nuao may Nam vmuuHavm cmmn u>.50% noHsa ou N050 onu «0 mafia: «nu 5H ouHuz NemH m

 

NNHmamm.50HNNvHHN>uNN0uo can aoHuwvHHw> you
NHN%HN:< Num5vmuHsu New .%Nam5vmum mmaoamom .NNHNNNNHNNN50

NNHszNHNNNNN HNNNNN NNH: NzHNNm

148

 

mMH MMH M N HoH HH oN mHmuofi.
H

NM NM mN m @J z
NM MM H NN N a
NN MN H NH H o o
nN MN N NH m o m

H <

0H 0H MH

 

Immuoa .N.5 % x .m.5 N‘ x Nm50uw

N328 HEN NNHNNN NNH
NumaomNom

NN.HNNN. zNHNNmHHNN

 

 

mHsmumHonom w you %Hmmm N0: NHN m5
ummmo mHsmumHonom m vo>wmumu %
ANSNHNHonom w you NNHHQQN. x

.ummmo awnmumHonom m Nw>Huomu 0>.50%_
50H:3,80um Nwao man Nam A%5m NHV mAHnmumHosoN you NNNHQNN m>.50% soHsa Nu mmao man no mqsma «nu 5H oufluz. voMH.m

 

NmHmammtcoHNmNHHm>aNNouo van aoHumvHHm> now
NHN%Hma< mum5vm:Hso Nam .%Ncm5vmum oncommmd .NuHmccoHumm5d

la11fi553440d37 JDDNNJO 33.18. DrflJaHO

 

 

 

 

 

 

 

0N 0N HH N mN H NOH NM M No NH NH NH MN N m m No NM M M M no No mHmuoH
N m N N OH NN NH NH m M w m H H «NH MN N H H MH ON 2
N N N N. NH MN m NH m N N m H M N HN N mH NH O
N m H N NH NH N N NH H M N M H N H NH 0 H H H MH N O
o H N N NH N H MH N o M M H N H MH w H NH OH m
M n N H M H MH 0 N H N H N w H N N <
a N N N H N N N N N N N N N H N N N N N N N N N H Na
HmHoom msmoH m5mawo 50Humu5no¢ omu5oo
N.N-HNNN NNHH H208 NEW $3333 3&8 N N N H N HNH N
N.N.N.HNNN 2N3 NHNNHNNN N N N N N
m5¢fimo vcw 50HumuoH N M N H O ONH m
%.O.muH\MN Hoonom onu m0 50Huwu5mou Hmuuuuu m N m m %
uamz 50% %v5um N0 wmu5oo mgu waHuume N M H O aMH m

 

 

 

N
305% capo: Hw5mu. umuuum
u.=0O u5on<

Nm%w3 wdebHHow osu «0
50mm 5H .wcHuwvaaoo Nuo3 50% muonuo «nu ou mummaoo Hoonum mHau mmou 30: AamuH u>onm ecu ouv mm% NH HNHumMH m

 

 

 

M NNH N NH NHH NHSNN
1 NN N N NN z
NN N N NN N
NN N H HN N
NN N N HN N
NH. H N HH N
mHmuoH .m.: %‘ x mm50uu
mwmaommud

 

mNHMNH ZOHHdnHH<>

>~a

NNHosu umuHm mm mama N05 NH
wuHosu umuHm mm 08mm MN K

Nuawuum HHH3 50% mwuHHoo £0Hn3 30:3 %HmuoN5Hw0v 50% NO wMH m

 

NNHNamw NNHNNNHHNNTNNNNO HEN NNHNNNHHQ pom
NHN%HNHN< 333-20 HEN .Noaosvmum uncommon, .NNHNHENHNNNHHO

HMHNZZOHHmde Hoomom :UHm UZHme

 

wuHQEmm coauwwNHm>|mwouU flaw EOHumVNHm> how
m«m%452< UhGJUWINSU $25 .%Uflflzrvhh mmcommwfl .Bhflmizoqumvio

9. VII (ZZAJUJ‘..-.uuuunud~.v 1*. ifiVuQHbun‘ unwdd‘uq wiz‘~.U-tfla

 

 

 

 

 

 

 

 

 

 

 

150

.mN. Mm NH NuN

 

 

 

 

 

 

 

 

 

 

 

.mp: NM NH m
a Nx z uoz z NN NN mHmuoa NNH N oH N NN NN Namuoa
‘ NH A NN 2 NM M H N NH NH; 2
NO.MH ON NH N..N MN V N O Nm N H HH NH. O
Hoo.uv NN NN N nH ”v N o NN N N HH NA 0
. NH V N m Om H N, N NH. N N
a Nx z uoz z N. u‘ N < NH H m4 N 4
aoHumuzmwM¥ NuN O m mHmuoa %4 N N N m mnsouu
mHan mumadmuwno maowuomuHO mmmaommmm
NNNNNH zOHaquN<>-NNONu
N.N-H\NN NNNH HmNoom Nam NNNNNNNNUN msasmu N N N H N HNH N
N.o.N-H\NN magma UNNNHNNN N N N N N
msaamo Nam aONumooN N N N H o oNH N
N.O.NuH\MN Hoosum osu m0,aowumu:mmu Hmumamo N N N m N
udms 50% hcnum m0 mmu300 mnu wawumwmo N M N H O mMH m.

 

 

30am mmu03A Hmsdm‘ Houumm
u.:0O udon<

NmNma NGH30HHON o:u.
m0 sumo 6H .wdHumNchoo who: som‘mumzuo mnu ou mumg80o Hoosom many mmow so: AEmuN m>0nm mSu ouv mmh NH HNHammH m

NmHmamm cowumNNHm>umm0uO Nam aOHumNNHw>.u0w
NHNNHmn< mumsvmuwno Nam .Noamavmum mmaoammm..muHmac0Humm=d

afiszoHHNmpo .Hoomom NEH quNmN

VIVI-<7u7~A,-.-,1h..i_-‘d .-avAi-urmn quv-uc w~7H-,U.-.-m.¥.

 

 

I;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NMH HH HH NN NN mHmuOH
NM N N NH nH 2
NM H H N MN O
NN N N m N U
*HN.N ON mH mum On N m HH NH O
nOmV NN N N NH H H mH <
n Nx O uoz u NHMuoa o N N N mmdouo
mumcommam
wwHNMH cOHumNHHm>umm0uo
NMH m N Nm NN mHmuoH
an N H NH ON 2
NM H H OH NN O
NN N N NH N U
*mM.M NN NH mnN mN M NH OH O
NH.V NN N N NH N NH N
1
U a Nx Ouoz o mHmuom. o a N N mmHHouo
mwmaommum
wMHm<H ZOHHNOHH<>
xmms me whoa H0 muson ON m
O.NIN\N xmm3 Hem mason ON cmnu mmmH N

uaoc N

Nnon N:H%mn m an News Hum
xuos ou uommxo 50% ON .%ON MH .wEHu £058 30: udonm .Hoosom ou wcHow %HHmduom wu.=0% cmsz mEHu wnu NOHHDO MNH w

 

A.=0N you unnuuomaH mum: umsu mwaHsu on» HHm xum£0v
chmuum ou Hoonom NUHOB NcHNHumN mum3 30% cmss NaHE aH cm: 30% mcoHumnmNdeoo ucmuuomEH umoe mnu uuwz umnz NNH m

 

moHOme coaumvHHm>ummouo Nam :OHumNHHm> you
NHN%Hmd< ouwsumano Nam .Nuaosvoum wmaommwm .ouHmacoHumosd

NMH<ZZOHHmMDU Hoomom mUHm OZHmmm

 

 

 

a;

 

 

2,...-

Q .fVQL-‘c -

In

..-. .‘v

‘0‘.

 

152

 

 

 

 

 

 

 

 

 

 

 

 

mMH N H m oH NN Nm NN «kuoa .
mM H N HH NH N 2
NM‘ N H H HH NH n n
NN H N N 0H m o
HMo. NN MH ouM on H m N MH N n
ma Nm NH NuH NH n‘ o o N
a Nx m uoz m «Hmuoa \NW 0 n N. M. N H mmaouo
muHNNH ZOHHNQHA<>¢memo
mm mN meuos wMH m rm M NH NM mNI NN mHmuoa
NH An MN 2 . aM N H N N, HH NH 2
NH An NH 9 NM N H n w MH n n
m AH nH o NN H m CH m o
kma.m mM NH ouM NH UV m m nN N n m N N m
S .V 3 a N-H N A 2 4 3 N N N N m <
m‘ Nx m uoz m o-m a N-H mH38 N‘ o n N m. N H maaouu
moHan and: muaso , maOHuooan noncommwm

 

mMHMNH ZOHH<GHH<>

N.o-H\oN

n
o

A.Hoo:om aH awaHnu mo uwuamo any mucmmuunou mHouHu «nu mo uuuawo_assv N:ON mum mwaHnu
mo nouaoo can aouw ado 9mm Bow .HOOSOm um mama so ow udSu meuH>Huow unu voucummumou_mHouHu usu «mommam NNH m

 

megamm GOHuwwHHm>nmmono can aOHumvHHw> Mom
mHmNHu:<_mum:omuHso can .Nuamsvoum oncommmm .muHNGOHumuso

uzN<20HHmmad Hoomom moan 92Hmmm

... ~\-:.-.s

1... a .u..... v o N

 

1:.
...-~.- - -

 

 

 

ON.n NN MH m.ono

 

 

 

 

 

 

 

 

 

 

 

 

No.nv NN NH N-N NNH NN NN NH N N N N N N N NHNNNN
NN NH N oH H H N H N H z
N‘ NN N uoz N NN HH N H H N H N N N a
NN N N N H H N N H o
NN.N NN N N-o oN N HH N N H H N H N N
No.VVN. NN oH N NH oH N H N H N
IIINH‘ Nx xxwummmlle NHNuoN N N N N N N N N H o xxlmmmmmm
mmmmmmmmm
llnullnunuulinnunnluuiu
NNNNNN onNNNHHNNNNNoNo
NH.N NN HH N,N-o
NH.uv NN NH N-N NNH NN NN HN N N N N HH N N NHNNON
NN NH N N H N N N N N z
N x N NNHNIINN NN NH N N H H N N N N n
N NN N N N N H N H N o
NN.N NN N N-o NN N N N H H H H H N
:3, Ho.VuN NN HH N NH HH N H H N
1 II‘
a Nx N uoz N NHNNON N N N N N N N N H o NNNONN
N m m N N a w a n g
a D. 1. A . u .w m. m m
o a I I a I 3 x.
n I. a. E n... u 3 a D.
1.. .4 n m 0 w 3 F
T. rA J U. P. 9 A .l.
N s N N N u a
H u" n u m.
m. I u
aw aw u
3.
s
mumaommom

I‘llllull‘lnl

NN.HNNN onN<HHHq§
Novauuum HNONNN NNHm H .Hoo

 

moHNHEmm HSHumHuHHmeummouo Haw coaumuNHg Mom
NHNNHNGN uuuuumano Nam .Noumsuoum omaommom .ounacoHuNaso

muzzzofiamgd .Hoomom HHUHHH UZHMHmm

 

An-u - (7.....hnv 1.. \uvqunu‘.

 

 

 

 

 

NN.N HN NH N-N
NN.VN NN N H

 

m x U uoz o

N"NN.NN NHH N mum

 

 

 

 

 

 

 

 

No.vv ea c o
, , NNH NH N H N N NH NN oN_ NN NHNNON
9 ac . ur\
Nx N z N NN N H H N NH N N z
. u Nm. N H N N m 0H N NH 9
NNN H NN NN N H NN N H H H N, N N N
Nu m N o
mN N H m n N w m
NHNNOH o m m N o n N. m N N museum
$22.3 NNNNNINNNNN
2 NNNHN mNmeNmmsuN
O u. S J N1 o a T. . u.
m‘ o .4 .a o”% MN m U” N nv 0 E
“M Nx z u z z m m mwm. o .3 a o m“ I. s
.1 . Mum 1. n. .N. my m n. o M. 1. my
NN om NN ,1 m... M“ as a n: N m. a H ”m.
NNH NN 0H m a.” s um w u m. m. u m.
s 1» 0.4 .4 .L .i I. o n.
.N «as a a. n. m n“ n. q;
A NN 982 a m an m I a I a o
s .2. > N u N. s N
uNNN. NN NH NN.N .m uu w m w. a I
ma Nu m a NM. NV. .4 Sn 9 mu
0.. u .4 1..
» . «No a .
.W Nx N 32 N mm. m.
N5."
Nuanma mum: muano mmmmmmmmm
.ANHao «woo NNouuv Naauo\o chwvamuuw Boa ohm :0» mwmaaou any we mean «nu NN awn: N o

 

Noagaum coaumvNHm>1NNonu cam coNumvNHN> now
NNNSNNN 333-20 23 $3333 «2533“ 63353350

n54.14345344313y 14911414), 55414...)

 

 

L".

 

 

.c

‘3'

 

 

155

NHN.N NN NH N-N.o

 

 

 

 

 

 

 

oH.uv NN N H
ale. U uoz o
mHan muwsmwuNno NNH wH m N m w NH Nm HN Nm mHmuOH
Nm oH H m mH N o z
Nm N N w N o m n
muN NN m H m m m m o
*wm.mH ONH NH .m:o on m m N N N m m
Ho.uV H N 0 NH H N N H m H m <
NNx N “02 N NHNuoN o N N N N N N N N H NNNONN
mHan muwaumuwsu mmmaommmm
mmqm<H ZUHH<QHA<>ummOMU
%.muo\o Nwawwcmuum.3oa can 30% mmeHoo msu no mean msu mN wuss N n

 

 

NmHaamm GONquNHN>ummouo van aoNquNHm> How
NHNNHNNN NNNNNN-HNN NaN .NocNNNNNN NNaoNNNN .NNHNNNNHNNNNo

NNHszoHNNNNo NNNNNoo quNNN

 

 

on O ltlt.

156

 

 

NNH NNH N N H NHmuoH

mm on N H z

Nm Nm N a

NN NN H H o

mN mN m

A.moHo:u umuwm um nmuamoom mcomum>m .mdumuumm ozv 0H HH N N H 4
meuOH a w N o. mmaouw

mmmmmmmmm

 

mmqm<H ZOHH¢QHH<>

.Nmuuma u.ammov NH an: .aHmwm moHono mama mzu mxma anos H m
.5“ H mums3 Hmo kuumn an H N
.NUHoso umHHH NB mo mmeHoo msu ou maow m>m£ vHsoo H nNNB HHNum o

N.N-N\NN

A.moNosu uNHNH uNmsu mo mmeHou mag
cu vmuamoow uoa mmonu Na cmumsmc<v N=o% now use voxuos NwaHnu has msu uaoam Hmmm 30% ow Son .vHou HH<

NN o

 

NmHmamm cOHuwvHHw>ummouo Nam aoHumvHHw> Now

NHNNHmc< mumscmano can .Nuamsvmum mNaomNmm .NHHmaaOHumwso

no... I<I~9ln

 

 

 

 

 

 

157

EN

EN

\NN
\NN
\NN

mmo£u NHHmaHm so» Hoonum mSu wNv so: .muwonu HSON waNxms mama 30% mEHu mnu um do» ou vwammm uN m< OMINN o

A.aowumm=v.mN£u Hm3mcm uoc vHv mHmoma onuuNmHAMu ozv

mmHH HmNoom can NmHuH>Huua mamamo
Nammu oNumHnum

mamado can cowuwooH

Hoosom wnu mo aOHuwusmmu Hmpmamw

uama 30% hw=um mo mausoo man wawhmmmo

NOONOON
MNMNM
NonoN
HmHmnH
>50 >~O>s

 

306M mmuoz Hanan Hmuuom
u.aon udon<

vaummoow mum3 50% nowsa um mumSuo msu saws mummaoo

 

Nquamm aONquHHm>uNNouo can GOHuvaHN> now
NHNNHNNN 335-20 28 .5533; 333% .NNHNESENNNO

158

 

 

 

 

 

 

 

 

 

 

 

 

 

mmH m N o NHH NH m N NH Nm NH NN m o Nw m o N m oHH meuoa
an N H N Hm N m N H mN N m H N mN N N H H Hm 2
NM N on m H N NN m o 0 NH N H m wN n
NN H H NN N H m 0H m m N NH H N HN. u
mN N H H CN H H N mH N o N o m H H m 0N. m
0H N NH H H N NH H mH 0H <
mHmuOH N o n N m N. H o. x m N o m N m N H o x Nanchu
nonumm muonuwma avamwum Hanan:
mMAMdH ZUHHNGHH<>
%.mux\om mumumHm no mumSuoum .m m N H o x N
a.wuN\Nm3 umsumm usow .v m N o m N m.
%.mnx\Nmr mumnum0u uaow .u m N H o x %
N.N-NNNN NNamNNN uNmNoHo “new .N N N N N N N
h.mnx\Nm uwzuoa “sow .m m N H o x N
m mm mm NN.N mN
u AK 35 355 NM
3 mo. ”.3 0.3A Wd
x. .A I. a u. =.u.a INT.
u _ D. o 130 J u a
o o u n «an a .3 a
A 1 .. an ace 8
3 u. 1*u.u 3 a
3 3 3.0 “HP
“.1 3. 9‘
a a .1 .1
o n. m o.m u.m.
3 T. e o s .3 o
u:A D. n a u.n
a . a 1.3 3“:
13 p‘ u-
a a 3 3
J n10. u.
a a n 1.1.
A 3 u
a .3

wamHHou a mo NUHono HaaHH usom uaonm Hmmw.3onn vuuNHH,maomuma msu

vwv‘kbm onuNm U

 

aOHuNcHHw>nNNouu can cowuwuHHm> you

NHNNHNNN NumsvmuNso Nan .NNNNNNNNN NNaoNNNN .NNHNaaoHNNmso

IL! 1. llf‘ .‘Olll II;IIl-lI-I\I!¢i P!- a.IHIF..II:uII 1.1!“. 1:!!! § entaili fl O Uln‘uI-\

.cv- - l\vs~..... . . ..

 

 

 

 

 

 

159

*NN.m No m mum

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No.uv an NH N
me NH HN m NN mN nkuoa
m Nx <Vuoz < Nm m n H m HN. z
Nm_ N n N e mH a
OHHQH mummMWnHzo NN N m H NH 0
on N o H m HH m
NH H N NH <
no.N oN mH mum
no.uv No HH N deuoa m N o m N mqmouu
a Nx m 32 a 3:3;
«Hana mummumuHso mmmaommmm
mmHm<H ZOHH¢GHH<>ImeMo
«NH.N Nm H mum *mN.o mm 0H wnm
Ho.uv No mH N No.uv mN m N
W «x 4 uoz < a Nu m uoz m
avaoth _ ., avaoHuh...
mHnnH onusUmano mHan mumJMWuHso

 

.H.zoo mudm<H ZOHadfiHH<>

NwwwHHou a mo 00Hono HusHm Hack usonu Hmmm onmn cmumHH‘ucomumm m£u ch 30m omuNm u

 

mmHmaum cOHuwvHHw>ummouo can aoHumvHHw> now
mHmNan< mumsvmuHso can .Nocmswmum mmaoammm .muHmacoHummso

I I I. I II if I Ii-ndv.‘ 2’ [I IiillJ . u ' rl I I

...v. -f\9.u...-.v.3...A...-..V. ..n‘v.o s.......- ‘.3.

 

 

 

 

hvlv

 

160

 

 

 

 

 

 

 

NN Hm mHauoa me a HH NN Hm aHuuoa

NH An NN 2 mm m .d. NN 2

OH ..V NN 9 NM H m or NN 9

NH ' NH 0 NN m H. a NH 0

NN.N mm NH wnc O ..V mH m mN N n. mH m

ON. mN NH n m; UV MH < 0H N H NH N
a N u uoz 9 mac n m mHouoa m N o m unwouu

«Hana «unnamuHso mGOHuuwan momnomumm

 

mMHMdH onfidfiHA§>

.Hoosum umsuoau ou cm on
263 383 H 32 «was: 23 22 H HH 28 .335 as»? .2: 32. H xfifi face H "mama—2‘33 E H w

.muHoso uaoymHHHc a
wxma vHaoB H muam aw H .chwu um>o HHm uH ow ou_vmn H «H usm "mn<z_H moHomo may mHH3.HzmHzcu :4 H N

.umuumn
30:8 hum> ammo m>u£ uoa vH=o3 uH use .Hoosom umuumn a ammoao m>wn vHsoo thunoum H "QHHhmHH<m_z< H o

. . .mUHoso mama unu mxma HHHum deoa H .anwu.
um>o HHa uH ow cu on: H HH can .onOSu umuumn m onE m>m£ stoo H xaHnu u.aoc H "ammNMHm NMM> 24 H m

$.wnn\w.n NQOHuqumvw
mwuHHoo.u=o% now mum: 0800 cu mums :05 «UHOSU mnu usonm Homm 90% av 30: .uasouom‘ouaH waHsuhum>m wcmeH Nm

 

mmHmadm :OHuuwHHu>ummouu can GOHumeHm> How
mHthmc< mumsvmaHno vac .mucwsvmum mmaommum .muHma¢0Humm:o

mmH<zonHmmoo mwmqqco qummm

161

 

 

 

«mev we om ON ON Hm mN N mHmHOH

mm OH OH OH O N N 2

NM HN w H w m N H O

NN OH O m O O O N O

mN HH O m m m N O

OH O N N m N m <
:oHummsv mHnu umamcm uoa vHO mHmomm wo¥ mHmu0H m m N O m N m mmdouu

a:
mmmmmmmmm.
mMHm<H ZOHH<OHH<>

wmeHoo mnu scum GOHumEuomcH.umuumm w mmmeHoo whoa ou OmHHaOm o>an vHsozm H n
mOGMHuH NE scum mUH>Ow umuumm N Hoonom :an aw mumSommu.>B Eouw on>cw umuumm N
Hoosom msu um NwmmuHm muamwsum ou wmmeu m>m£ OHsosm H o muamuwm NE aouw moH>vm,umuumm m

*AvaHmHumm uoc
mum3 0:3 «menu Np OmumBma<v NmUHono Hmuumn m mxma ou Omvmm: m>mn 50% OHsos awn? .36: uH so xoan HOOH 50% m¢ mm 0

 

 

 

 

 

me. H HH mH mN co mHmuoa
Nm H N m NH mH z
Nm N, N, o ON O
NN H H m HH 0
mmO. O0 HH mum on m n NH OH H
.m.: mm HH m NH H N NH_ <
a NH 0 uoz o mHmuOH m m H o m unwouo
mHAmH mummMmuHso mmmaommmm
mMHm<H ZOHH<GHH<>ummomO
¢.w-m\mH HaOHuuuavm

monHoo Hack How mum: msoo cu mane 50% moHono ms» usonm Hmmm 50% ow so: .udsooou ouaH maHnuhum>m deme Nm 0

 

aoHOaam aoHumuHHw>ammoHO can :oHuucHHw> How
mHmmHoa< mumaamuHso can .Nucwavmum uncommon .mHHocaoHummao

11445453440337 55.31.1433 95.13..)

 

" it "" -'..‘H‘M_

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH H N NN NN NHNNOH
NN H N NH NH 2
NN H NH NH a
NN N N HH 0
NNH.N NN N N-N cm H NH NH N
oH.uv NN NH N NH H N NH N
a NN <uoz < mHmuoH. m m h @ union—U
mHnuH muwdumuHso mamaommmm
NNHNNH onHNnHH<>-NNoNo
,HN.H ON oH N-N
.m.a HN NH N
N Nx N uoz N mN‘ NN NHuuoH NNH N N NN NN NHuuoa
NN uv NH 2 NN N oN NH. 2
NH my NH 9 NN H H NH NH 9
NH ”v N o NN N NH N u
o. NHH.N NN N N-N NH An. NH N NN H N NH N
m NH.V NN HH N N A HH 4 NH H N HH <
N» Nx < uoz < N-N n N NHauoH N N N N mmwouw
«Hana mumaumano acOHuuman mmmdommmm

 

mmqm<H ZUHH<GHA<>

.cHo H umnu aOHumumamum umuumn a uow mucmc5um umozw uMOON w
.NHN H 533303 mag 2: 30pm How 3533 Haas "8453 Son.“ N
.muamvsum umoa saga cmumamum Hmuumn Ed H "HzmHHmONm o

NHoosom NNH;
use» aH uow 30% :oHuuumamHm Ho naHx way many 90% ow so: .mmeHou aH.oum£ mucmwaum Hague ou vmummaoo MN N

P

 

NmHQENm GOHHNNHHm>ummouo cam aoHumuHHm> pom
NHthma< mumsumano cam .mocmsumum mmaoammm .muHmc:0Humm=o

mmH<zonHmmDo momHHoo OZHMNm

 

343>_5~..< .HLCHHHNuml—Suv 7...: aNAUH....N..~.v.l..M-..~ “INC—.135: ILLHQCCCM N....s:.~v

N.I'II<C.-‘.-I¢Q In. .n‘..w. I .. I n 0.. . own it. '4‘

 

 

 

 

163

NN.N NN HH m
no.uv Om NH N

 

a NN o uoz 0 mac: Baum ha3¢¥

 

NHNNH NHNNNN-HNN

 

 

 

 

 

 

 

 

 

NNH NN NN N NNH w H Nn mm H ON NN 0 mm mN N7 mHouoa
Nm NH NN Nm HH mN wN N N MN NH H 2
NM NH HN Nm N mH NH NN N H NH NH O
NN o , NH H NN N N mH HH *NH H NH N O
NN w NH NN H H HH NH mH m N HH m H m
NH n OH H mH m o N H NH N NH m <
meHOH N m N . N m N1 o m N1 m N H \d n N ummmuu
\wN ‘ \NN .
mmmcommmm
mmqmdfi ZOHH<OHH<>
\wN .Hoonum :an :H cmnu “mun“: nose xuos on was
m>a£ H can ow ou uHaoHHHHv whoa none mmeHoo aH Muck msu canoH H N m N
\NN .umh mama mxma ou mHnm
soon He: m>ms H mum: Han NoaoHum mo uoH m and H Hoosom Hm s‘aH N m N
\NN .mwvasoax can NmmvH Ho mvHuoz 3m: mHons wmdmmo.
NuouoauumaH wsu mmamumn waHuHuxm Num> mp ou Ham» HNNHH wxu Nazca H o m N
\NN .maon Scam hwsu wnHon mH mmeHou aH cu tum: uww ou waHnu ummvumn «:9 m N H
\NN .umvums mum mmmusoo us» New usmumeHv mum NaomNASN

mnu uQSu ummuxm Hoonum sta mu mama msu none Nuumum NH mmmHHoo O x N
mwuwwNHO mmuw<

....u¢£u memeHc no mmuwm 50% on wNuNN o

NmHgaNm aoHuNNHHN>uNNoNO Nam aoHuNwHHN> How
NHNNHma< NHNSNm-Hno cam .Noamsvmum mNsoNNmm .muHmadoHuNmso

 

. *1_

 

l\p.
.rul ' I

164

 

NNH N HN NN NN NHNNOH
NN N N N HN z
NN N N N NH N
NN N H N NH 0
NN N N NH N
NH N N. 2 <

 

NHmuOH N N N Hg mmdouw

mmmaommmm

 

NMHM<H ZOHH<GHHd>

NIM\Om 02 N
mm» N

chHE 5H Hmom wEmN mnu m>m£ HHHum 90% on

N.N-HNoN 02 N
3N H

waoHHou :N How mummmum ou
twaaNB DON uMSu ummuao Ho aOHummsooo HNHSUHHHNN m m>m£ so» vwv .umnfimuamm ummH monHou dH vmuumum 30% am£3n on O

wNH Nw NN OH meuoa

 

 

 

NN ON NH N z
NN, NH NH H O
NN NH N H O
*NOO. mN NH N ON NN H N m
.m.a NN N N NH HH m H 4
a, x O uoz O NHuuoa N N H mnmouu
«Hana wum: muHso mmmaommmm

 

 

mmHm<H ZOHH<GHH<>umeMU

....uunu mmumeHv no mmuwm 30% on NN-NN n

 

Nonamm aowquHHm>nmmouo can coHuNNHHN>
NHNNHwa< mumsvmano ncm .Nuamsvmum mmaommwm .mNHNccoHuNmsa

NN<szoHNNN=o NuNNHoo wzHNNN

165

 

 

 

 

 

 

 

 

 

 

 

NNH N N N NH NN NN «Hauoa
NN N N H N NH N z
NN H N NH HH N
NN H N NH N o
NHN.N NN N NuN NN H N NH NH N
oH.uv NN N N NH H N1 N, N
A NN N uoz < mmmuoa N N N N N. N ammmwc
oHan unmamm:HAUt . . Noncommom,
NNHNNH onHNNHHNNNNNNNo
HN NN mHuuoH NNH N N N NH NN NNH mHuuoN
NN uv NH z NN N H4 N NH NH, z
NN uv NH N NN H N H N NH oH N
NH uv N o NN H N N NH N o
NNN.N NN. N N-N NH uv. oH N NN, H N oH oH N
NH.V NN N N N A. N N NH H N N <
N Nx < “oz ,N N-N N N NHNNON N N N N N N «macaw
mHnflH 0kg miwfiu mHHOHHOOHHQ mmmflommmm

 

mMAm<H onH<dHA<>

.uwmh axon use moat ou waHdauHm NHQUHQHHmv Em H

.ume wfiHsoHHON mag no Hawk uxma use Nona NHAunoum HHHs H

.mflawuaou HHNB H umnums3 munm uOd an H

.mouwov «nu ou oaaHucoo ou yaw? H umnu chuuou you an H as; no ow ou-amHa H
.HOOSUN mumsvmuw ca :0

ow ou wchamHm uoa an H uan .mmuwow usu.nom.:o_mnawunou en‘smHm NHmuHaHHmv H
.HOOAUN mumsvmuw ouVao cm can mmuwmv NE umw ou wcHdauHN ad H

WONG!)

Md’

N.NIN\NN NGOHuuusvo mmmHHoo use» wdHnaHuaou uaonm Hmow 50% ow 30: .3c: vauum mwaHnu m4 Nnvu

 

NmHNaNm aoHquHHN>-NNoNO was aoHuNNHHN> pom
NHNNHN=< NNNNNN-HNO Nam .Noamsumum mNaoNNuN .muHocaoNuNmad

HMH<ZZOHHmMDU momHHOU UZHmmm

NNH mN. HN N3 nHuuou.

NN NH n H z
NN. NH HN n
NN w mH‘ H U
NN OH NH m
0H H NH d,

 

 

NHuuoa N N N 11mmmmmm

 

mmmmmmmmm
NNHN<H ZOHHNGHH¢>
no» N1
oz N
N.N-N\NN Nva mHnmaoHHum mo mHnNuuHonuN haw waH>Nouon 90% 094. me o

 

 

NNH NH NOH N N NH unuoa

NN N NN H N N z
”m NN, H NH m u
mN, N NH H. N m
0H NH H .N d.

 

mHmuoa N N N N mnnlmmmmmm

Nouaommom

 

mmHm<H ZOHH<GHH<>

NowuHHoolanu mo.u:o menu cu auHm Hoz 0v H N

.xoun ow uu>oa NHouHaHmoO N

.Hoonua cu.xown om.aonu can «HHsk a you can a uou N
.oonHoo.nonuoau 0» cu n

:3

N.N-N\oN NuuuN an»: on soN HHHs uu NN.“

j

NonaNm NOHHNHNHHN>umNoNO ES 333:; you
NHNNHNHZ onus—$12". 28 $3333, oNaoNNmm 63.283330

mmH<ZZOHHmde NUMHHOU OZHmmm

 

Dd ..
‘4. a
. nu?"

 

 

167

NN.N Nm NH cuN
N n

 

 

 

 

 

 

 

 

 

 

ma cc NN - NN No mfimuoa and N Nm Na Na «N. ea udmuoa
3 A E 2 an a N S a m N .z
a «x 2qu z «N V 2 n «a H S S m m N a
S A 2 o «N H N o m a N o
:4 3 S o-m 2 V 2 m 3 o N m m N m
3 3 2 N-n 2 V m < 3 d N n N a <
INT ~x o 82 u o-» a N-n £38 o N. m ,N o m mmsloud.
NUHSNH. 9H3 midfiu NHHOHUOOHHQ NQNHHOQNGM
Nudge 538349 No2 a
.NmN. m
Non N
o.o-n\m .5 c
USOHH m.

Nuom Nam cu mo>HumHuu no NHfiawm u50N uuwnxo :ON ow mumoo mNmHH u umoN vacuum n30N mo «Nauauunma awn? uuon< NoH o

NNH NN NN N. NHMuoa

 

NN NH NN H
NN N mm
NN N NH

NN N NH H
NH. N. NH H

‘Nsncacazs

 

ages a m N masons

 

noncommom
was; 3.33 fig
mmw N
oz NV
N.NuN\~N NHNON amaflmmum Mach NaHuav uaHu Nam um N03M03.30N m>wm «N o

 

mngamm coHumvHHm>wmmouo can aOHuwvHHm>.u0m
mHmNHmaN onusvmano Nam .Nuaosvuum monoammm .ouHmacOHumoao

mmH<zonammpo momqgou qummm

168

 

 

 

 

 

 

 

 

 

 

 

 

 

0N NN NHmuoa NNH N NH. HN NH NN NN NkuOH
NH mm HN z NN. N N N N CH HH 2
NH An NN N NN H N N. N. NH oH N
NH uv. HH 0 NN H H N N. N N o
NN.H NN NH cuN N An NH N NN N N N. NH N N
NN.HV HN HH N-N N! .n HH N. NH N H N N. N N
mum Nu N uoz o N-N a Nuw. NHNNNN o N N. N N 1Nw+ NNmmmw
«Hana onusmmano NmoHuuuwHN Nommomwmfi
NNNNNH ZOHHNNHA<> NooH N
NNN N
NNN N
NNN N
«no: N
o.NnN\N NNNON oawuuuuma Nam Nwdwcumu.nuaaum Baum Nam ou uuomxm :ON ON uNmuauouun pus: uaoa< NoH o
NNH N. NN ON NNme NH NHmuoa
NN. H NH N. N N, N. z
NN H N NH N. H N, N
NN N N N N N o
NN. NN NN cuN. oN H H oH N oH N N
Na NN NH NIN NH - N N. N H N
N Nx zuoz z mHflNoHNWN N. N, N N“ N185,
0HNMH mum: NuHso Nomaommwm

 

 

mmam<H ZOHHNGHAN INNQMU

NuoN Nun cu N0>NuwHou no NHfiawm u30N uuoaxo.90N ow Numoo «NoHHou_ummN vacuum “:0N No mNmucoonmm wag? uaon<. NoH n

 

Nonawm doHumNNHm>wNNono Nam aOHumNHHw> pom
NHNNHNHNN 33$qu HEN genus—Noam NNHSNNNN .ouHNEHoHuNoHHc

NNHszoHNNmpo NNNNHNN NZHNNN

  

.‘(N - VN.~ N

 

169

 

 

 

 

 

 

 

 

 

NN NN NHNNNN NmN H NN NN NN NHNNoN
NH Au NN .«. z N . N NN N 2
NH .An NN NH NN. H N NN H N
NH NV NH N NN N NH N N
NN.N NN NH N+N N An NH N NN N. NH N N
NN.uv NN NH N N An NH N NH N NH N N
N NN N Noz N N+N N N NHNNNN N N N N NauouN
Naowuuuuwn mmmdommwm
NNNNNN zNHNNNHNN>
Gama NNON N
whoa oNoHHoo N
N.N-N\NH whoa HooNuN NNNN N

NuNoHHou.NaHNofiam.mum :ON amNu ouoa.uN Nahum.:0N NNN .Hoonum NNHN.H50N ou xuwn NflNMGNNH NHH :

It

 

 

 

 

NNH N NH , NH NN NN. NN Nkuom.

NN H N N N N NH z

NN N N N NH N N

NN. N N N N N o

NN . NN HH o..N 0N H N N N NH N N

2» HN HH NuN NH, H N N N. <
N Nu N Nam N NHNNNH N N N.{ N N n. NmsonN

oHNNH mum: muHso Nomaommmm

 

NMAN<H ZOHH<QHA<>1NNQMU

NNNON oaauuuuwm Nsm mmcwaumm umaEuN aoum Nam cu uomnxo.50N 0N,0quamoumm nun: u=ON<. NOH n

 

/
NoHaamm coaumvam>1Nmouo Nam coaumNNHm>.nom

NHNNHNNN NNNNNN-HNN Nam .NNNNNNNNN NNNNNNNN .NNHNNNNNNNNNO

NNHNZZNHNNNNN NNNHHNN NzHNNN

[\NN...

 

 

 

' Kiri" “.5 I'-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN NN NHmuos NNH N ON NN Nm NHmuoH
HH An NN. z NN. . N N NN 2
NH Au ON O NN H N HH ON O
NH A NH N NN H N N NH N
OH .An NH N NN N. N NH N
N A HH N NH N N HH <
NuN O N NHmuoa.aOn, N N N ammouO
.NdOHuoouHO g
NNANNH ZOHHNOHAN>
umNuwH N
uuHHwaN N
ONNN NNNN N
O.N-N\HH
NuNoosu NON NH503.NOH33 .oco.uoNumH m
on no .Hoonum uoHHmaN m o» no .ono NHNN No NNHN msu Hoosom nosuoam ou NNHON mo ooHonu HNON van :oN NH HHH O
nu
“u
NNH NN ON NN. NHmuoa
NN N NN N z
NN N NH N O
NN NV NH N O
«Ho. NN N N+N ONH N HH N N
Na NN NH N NH N. NH H N
m NN O uoz O NHmuoa O N N N mmmmuO
«H58 «NNNNN-H.N. ‘ fig

 

 

NN.HNNN. 239.3 54>. NNNNN

NmNuHHoo.NaHN0ncu.mum noN :NNN once uN Nahum NON NNN .Hoonom NNH: MOON ou sump NawxaHsH OHH O

 

NNHNamm aoHuNNHHN>NNNouO Nam :NHNNNHHN>.uoN
NHNNHNNN.NNN:NN-HNO Nam .Noausvaum mNcoNNNN .NNNNNNNHNNNNN

NNHNZZOHHNNNN mumqqoo Nszmm

 

mmanHN—Svm FNONUNNHufiHQNwImNmOkU Hvfiuu CONUNNHUNHGNI “0...!
NWNNINJNNC< 59.31:va -_nv Hus: .xnuuflunvNJHVrULunu QICORH-afiva .fl-qflduunuuunviunmflv:~u

qua-[\QNV-Aul nu‘ o.v\. n . I.-.|..u.l .I‘,.l'l‘l\\

 

 

%

 

may.
. A4

 

 

 

171

NNH N NN HOH mHmuOH

 

 

.Nm, «Ny NH NN z
NN N N NN N
NN N NH N
NN N NN. N
NH N NH N

 

mmduoa N. N H mmaouO

Noncommmm

 

NNNNNN zNHNNNNmN>
02 N
New H
N.N-H\NN
NOMONMOHN HON uNON .omHH
amumH NH O: BOHHON Ou oxNH NHnos NON Nam OuONu mum .HOONON um mums Nowvaum NON muomnnam «Nu HHm NO NHH ON

 

 

 

NNH N NHH ON NHmuOH
wa N HN N z
NN. N NN N O
NN H HN. N O
NN NH N N
NH H NH H N
mHmuOH \mwwmam Oa mmeo.mM uON mNOHO.:Iw wmmmmm

 

NMHN<H ZOHHNGHH<>

.NQOOHN NuH>Huuw uO NNOHO Nam Ou NNOHON no: ON H x
NNNH
ANONOO an NNOHNNO Nam NNOHOV .somo ON NHON NON OOHNNO
may umHH .Nam NH HOOHNNO am mum OON NH .uonama m NHOOONONN mum NON sows: OH ONOHHOO NH NNOHO Nam uNHH NHH O

 

NOHNENN OONNNNHHN>ummouO Nam OOHNNNNHN>.NON
NHNNHNN< oumfianHzO Nam .NOOOONOHN umdommmm .muHmdNOHummsc

NNHNZZOHHNNOO NONNHOO OZHNNN

172

 

 

NNH 0N 0N NN NH q mHmuoN

mm NN o N u z

Nm NH N N m. H a

«N NH m N N. H o

NN «H o m N‘ m

NH oH N N N <
mHmuoN NW N o n N «O N 11H. mmuouu

mmmdommom

 

mmAm<H ZOHHNGHA¢>
NNN.N muNos whoa no N

haw m muNON N uNONN
NNN.N muNOn NH uNonm
NNN.N uNOn H uNONN
Nam w HNON w uNonw
NNN.N HNON w NNNu mmoH
ONON umOaHm HO .oNON

Hund'mxon

N.N-H\NH
NmmmHo NH vNuNm
NON Nae» «nu Ou NOHuHNvm NH .mOmHNOO HNON NO xuOBNwNHON vaNm NON ON .meuo>m mnu NO QEHu NONE sow NHH o

 

mOHgamm NONumNHHm>1mmouo Nam NOHumNHHm> HON
mHmmHmN< onqum:Hno Nam .NONONvmum umNONmmm .OuHmNNOHumoNo

mmH<ZZOHHmmpo mOmNNOO Oszmm

173

. NNN.NN NN NH NNNHN Noxamu uoz
Hoo.uv. N. NH uNuNN Noxamm
. . NNNHHNN NsonHHum

 

9 Nu N uoz N

 

NuHmmwonmsmN-Hso

 

 

 

 

 

 

 

 

 

 

 

NNH NN‘ NN. HN NH NN NN NH NN NH NN NN. NN N NN NN. NN mHmuOH
NN HH NH N . N NN N N N. N N NH NH N NH HH 0H 2
NN NH N HH N N N N NH N N N N N 0H NH N a
NN N N. N N NH N N H N N N NH N N N o
NN N N N N NH N N H N N NH N N N N N
NH N N N N NH H N H N N HH H N N. N N
mHmuOH N N N‘ H N N N H N N. N H N N N H mmNouu
:oOquNuoom: :NOHNHHOH: :NNHNHmmH: :muNonmN: . .
NNHNNH ZOHHNQHH<>
N.NnH\HN mumNuO Na voxHH NN- Nmummuom NNHon
N.NuH\ON monHon mNOHNHHau NE Ou NN NNH>HH
N.NnH\NH HOONOm NH «HNHNNON ma NoNa mm NNHNummH
N.NuH\NH. muNmumN NE NNHNNNHN

A.mnHmNNOHum0NN HNNHNHNO «Nu NH mNOHuONuumNH uHOHHNNw whoa nu N Ou H memv NamoNu
NO Noam mH uquuOQBH BON umNe .mNmN mNmHHOO MNON NNHNNN HON o>Huum NON mNNHNu oNu NNQE< .uNONN
uNNNONu‘NHNNNONN,w>NN NON NNNHNN oaOm mum wumm .mNNHNu uNmummmHN NON o>Huum oHNOON uNomommHn HNHuNHH o

 

mmHNamN NOHumNHHm>1mmouo NNN NOHNNNHHN> NON
NHNNHNNN «NNNNNIHNU NNN .NONvaoum omNOwaN .mnHmNNOHummNo

mdezzOHammNo NmeNOO OszNm

174

 

 

 

 

 

 

 

 

 

 

 

HN NN NHNNNN NmH H NH Nm NN HN ymN NHNNNN
HN W NH 2 NN N NH N N N 2
NH V NH N NN. H N NH N N H N
N. V HH N N, H NH N N N. N
No. NN N N-N NH V HH N NN N NH N N N. N
N: N NH N..N N A NH N [S H N N N N N
N Nx N Noz N N-N N N-N NHNNNN N N N N N N NmNNNN
OHN—wH 9H9.— mldfiu NdOHu00HHQ NQNHHONNGM
‘ NNNNNN zNHHNNHNN>
3003 N Nuaau umuNu no can NNNu whoa .mmN N
3003 m NOHNN NO moNO .NNN N
Nxoos QNNNu HO O39 Num>o.ouNO uNONm .NNN N
NuNOa w ouNO NmNu NNNH .NNN N
N.N-N\NN oz N
NNNNN NoN oN NNH N
NNH N Nm NN HN NN NHNNNN
NN H NH NH N N z
NN.N HN NN NNNHN NNNNNN Noz NN N N N N NH N
HN.uv NH NH NNNHN NNNNNN NN N NH N H N N
NNNHHNN NNoHNHHoN NN H N NH N N N
NH NH N H N N
a NN 9 ”~02 Q NHmuOH N N m N .— Nan—ORG
NHNNN mum: N-HNN =NoHNHHoN=

 

 

NmAm<B ZOHHNQHA<>1NNQMU

NONoNu mo Noam NH uNmquNaH BON umNn .NNNN mNmHHOO NNON NNHNNN NON w>HuuN NON NNNHNu mNu NNOa< h
.uNONN NNNNONu mHnmnouN o>mN NON mNNHNu NEON mum mums .NNNHNN uNmumNmHN NON m>Huum mHNONN uNmuwmmHn HNHnNHH o

 

NNHNENN NOHumNHHm>1mNONU NNN NOHumNHHm>.NOm
NHNNHNN< NNNNNNuHNU NNN .NONNNNNNN mmNONmmN .mNHmNNOHummNo

NNH<zzOHHmmNo momqqeo ozHNmN

2.17;! —::< Cc. ..:r\4

.d! 5 .rd.\

‘-V. o INVNV.u‘ -

V-.~.-,d ..\..-N..~

 

 

 

175

NN NN NHNuOH NNH N NN NNV NN NHNuOH

 

 

 

NHA_ NH m N! NH N NH 2
N _An HN N NN. N N HN N
. N A NH N NN N N NH N
NNN.N NN N N N u N N NN N. N N N
NH.V NN N H N V N N NH N N H N N
_N. Nx N_moz N N N H NHNNNN N N N 1H» mmsoNN
NmOHuuouHN NamNomwwm

 

NNNNNH ZCHHNNHANN
.uH oxwa u.NON NON NH NNHNuNu0>o

uNOaHm umOH NON NOHN3 NH uNN .mNmuN «Nu uxma NON NH .oaOONH NOON NHNamuuxm Na mNNN NOHN3 NOH N N

N.N.H\NN

NNHNOH NO «ONNNO ONuON N 0>NN NON NOHNB.uNN «BOONH NOON m mNNN NOHN3 NON m N
NNHNoox NO muNN mum NON NOHNB.NNN maOONH ouwumNoa m mNmN NOHN3 NOH m H

quHN NON NHNOa NOHNN .muHONO NNON NNN NON NH .NNON uthmNN N mthu mum mumN NNH o

NmH N HNH NN NHNNNN

 

 

NN NN NH 2
NN NN OH 9
NN N. NH N o
NN H NH N N
NH H NH <

 

NNNNNN N N N NNNNNN

 

NNNNNN zNHNNNHNNN .
oz N
N.N-N\NN. New N

NANONV HHHN NNO NuHB NNNoum ON NON.ON NNH U.

 

NmHNamN NOHumNHHm>1mNONU NNm NOHumNHHN> NON
NHNNHNNN muvaNuHNU NNm .NONoNquN NNNONNMN .muHmNNOHumwNo

mmH<ZZOHHmmNo NONNNOO OZHNNN

 

 

 

30H»:
:NINA... 623%

:ta :1 N 1.:1 . _ < Z
. . O hmv
NN.H :57 T... d £3 HNH—5 NAU- 3

..-. -t\-~.-.. .

vsum :N:3::;:

9,“. h n N

.N... - NN.-NN.

«a «n NHNNOH NNH _ NH en. mm Nn NHNuOH

 

 

 

 

 

 

 

 

 

 

 

 

NN M NH .llz. NN. N HH HH NH 2
NH A NH N NN. H N N NH N
NN V N N NN N N N N
HN.N NN NH N..N NH V NH N NN N H N, NH. N
NN.V NN NH . N NH V N N NH N N N N N
NP NN NJoz N N..N N N NHNNNN N N N m N 3:8.-.IN
Gaan Guamldfiu wdowunumMNQ muwdomwfld
NN.HNNN zNENNHNN>
HN>NN N
NuNOa.N NONO NNNu NNNH o
suNQB N NuaNu m.Ou H n
x.Nu¢\mN 3003 Nuo>o N
NNUNNNU NNNNNN NON ON NouNO Bow. omH o
n
1
NNH N. HN NN NN NHNNNN
NN H NH N NH 2
NN. H N N NH N
NN N N N N
NNN.. NN N N NN H N N. NH N
2. HN N H NH. N N N N
N NN N 82 N NHNNNN. N N. -N.. .H.- .NNIINNNN
oHnNH NHNN muHso NNNNONNNM

 

mmam<fi ZOHedaHH4>1memo

NxOHm NON NHNOa sows? .NONONO HNON Nu: NON NH .NNOH uNmumNNNN ummgu mum whom NNH u

 

NNHgaNm NONNNNNHNNNNNONU Nam NONNNNNHN> NON
NNNNHNN< muvamnNnu Nam .NONNNNNNN NNNONNNN .NNHNNNONNNNNO

NNHNZZOHHNNNN NONNNOO OZHNNN

 

 

 

.i f

 

—€.v —.lr-<

N rk.... .. .

.

 

 

 

 

 

 

 

 

 

NNH ON NN N mm mN H NOH NN H OHH NN N mN NHH H NNH NH H Hm nN H NHNuOH
mN HH NN H NN HH H NN. m H NN OH H OH NN H 0N N H mN w H z
NN NN HH H NN m NN. o NN N H N HN NN N ON NH 9
NN OH NH NH OH NH N ON N H H NN ON N NH HH 0
NN OH NH NH o «H. N HN N n ON ON n NH N m
NH 0 OH NH NH N NH N 0 OH NH HH n d
NHNuOH N N m o n O N N x N H N o n O N N N N H N mmNOHO

 

mmHn<H ZOHH<QHH<>

.wNHENNuvaNNN NHNNNB ONHN NNuNO H m N o
.NNHNNONHNN NH: uNNNNwN ow Ou mNs

NNENuNaON N: .Nuun NNNOHN NBOHO wNHNNNH NNu NO uuNm on Ou NNNNS NOHHNN N NH o n O
.HOONON uN HHN3 Om wNNON NON EN H N N x \NN
.HOONON NHNN NN mwNHNN HONuNOO O53 SNN N NuN NNNNH N H N NNH

.NHNONN £uw3 Np Ou umNfi mwNNHNNONw HNHOON NONNN H o n O
.NHNNNE EOHN NNNuNNNNN NNONEON Np NHNos H .NNNuu NHNOO H NH N N x \NN
.NNOuw N NN uNO NEON NONuNN N NO NNNNHHNNN HNNM NNH N H N HNH

NNNMNNHQ NNMNN

. . .uNNu NNHwNNHN NO NNHNN NON On NNHaHNH O

177

 

 

 

 

 

NNH N HH NM on No mHNHOH
NM H m m m 0H 2
NM H N N ON Q
NN H m N O n U
oo.m No HH Nun om . H H o w mH m
No.UV. NN ON N NH N o m 0 ¢
Nfi NM 0 qu n NHNNOH N‘ N o m N NmNonu
NNNNONNNM

 

 

NMHNNH ZOHHNQHH¢>ummOMO
NnuuNnO NNNuuN NON ON NNuNO Bow ONH O

mNHmaNm NOHuNONHN>1NNOuO NNN NOHuNNHHN> HON
NHNNHNNN NHNNUNIHSO NNN .NofioNvmum NNcONNNm .NHHNNNOHNNNNU

NNHNZZOHHNNNO NNN.NNOO OZHNNN

 

 

' "In

178

NNH No NN H mHNuOH
NN NH OH H z

 

 

 

 

Nm mH‘ NH 9
NN oH NH o
NNoo. aq nH N on «H NH . m
as Nm N N NH N oH 4
:4 Nu a 32 a 333. N N N 385
moHnNH NHNN muwnu mMmaomwNM

 

:mNNaNNnvu NNN.. Rn

 

 

 

NNH NHH NH N meuoN

Nm. NN N H 2

«NH mu. m H n

NN ONO N o

NNN.H mm NN N on. NN N N

m: HH N H NH mH N 4
.mm Nx 2 “oz A z mHmuoa N ; H. N None

 

momcommmm

moHnNH NHNN mnHNo
:moNuNHNNv HNmn: \NN
mMANNH ZOHHNQHH<>1mmOMO

 

. . .uNsu NNHwNva no omuwN NON on NNH-HNH O

 

 

 

 

 

NN.N NN NN w *HO.N. No mN N
HO.UV. on HH N OH.uuv NN w H
m Nx a uoz a m Nx z “oz 2
:mNNaNmuquNv: \mN :mmNuNHNNu HNNu: \NN
. menNH muNNumuaso
mmANNH ZOHHNQHA<>
A.uNooV. . .uNau mmuwNme no quwN NON on NNHsHNH U

 

«Haeam NOHNNNHHN>1NNOHO cam NONNNNHHN> HON
mNmNHNNN ouNNvmuNno vNN .NONNNNNHN mmcommmm .NHNNNNONumNNO

MNHNZZOHHmNNo momNNOO Oszmm

Eud+n>fl ch< GIG-.nhac‘ flul- 11!: I

,‘Oo.-<$-...-.-...u-Su .-.-:-V-.-n~ v 1d5L-v-u-1h

 

 

 

 

an O NN NN NHNuOH

 

 

 

179

 

 

NN. H NH HN z

NN‘ H HN OH 9

NN N. NH N. O

NH. NN. NH N ON H NH HH N

m: NN HH H NH N N N N
m N m uoz m NHNuON. N N H. mange

 

NHNNH muNN muuno Nwmdommwm

 

NMAN<H ZOHHNnHA<>1meMU
NNH NH NN HN NHNuOH

 

 

 

 

 

Nm» N NHI NH 2
NN. H NNA NH N
NNH NA HH NH N
NNN.N NN NH N NN H NH N N
3V NN N H NH N N N N
LN x N 32 N 3331M N LN 3:03
acmflommflm
NNNNNN zoNNNNNN<>
02 N
N.N-H\NN no» H

Nwzbuu mNHNNmH map NO uuNN ouN NON uNNu NNm NON NHNoz NNH a

t

moHeaNm NONNNNNHN> macho NNN NOHNNNHHN>.HON
mNmNHNNN NHNNumuwno Nam .NONQNUNHN ONNONNNM .NHHNNNONNNNNO

NNHNZZOHHNNNO mumqqoo ozHNmm

Nun—d....~>—-na-< EH‘ICIQ‘U ‘Iiri I

..-vo.’\?vu7~.v¢ ..~ ~.-‘« n...n\- .N- u .us.‘1‘ts.slu

 

 

 

 

180

 

 

 

 

 

 

 

 

NN NN NHNuOH NNH Nw HN OH NN N NN NN NN NHNuOH
HH .AH NN. 2 mm NN. N. N N N NH m NH 2
N An NH a NN ON N N N N NH N. NH 9
N An NH O NN NH N. N H H N N HH 0
N An NH m NN NH N H N N N HH m
N An OH 4. IINH a N H N H N NW N N
N+N n N+N NHNuOH O a N N x. N N N mmNouu
NNH o +.NNH o HON NaONuuonNn NNH u NNH_O
shag u.NOa
mmANNH.ZOHH<aHA<> _uNHooaun
53333
O.NuN\NN NaouNO uaoafiuuNoum NHNON.NON sauna NON ON NuuNm snag? .:uNuNNomoNNN: Noxoonu NON.
uNNNNmmoNNH
uNuOOaun
NN.N-iNN amuNHNsmoN

Nuuumwwmu NON NHNOB_BON .NONuuNHo uxuu «Nu NH houNNwNu.Ou NN: NON

 

 

 

 

NN H .NN NN NH NHNuOB

NH a NW N. 2

ON N an N a

HH N N N o

NH H N N N m

N N. N N
NHNuOH .m. N N m» N.NO 0

mg
mg onHNNHNg

muNo u.don
x.NuN\NN Oz

NON

NH NNH o
N
N
N

NH NNH U

N
N.
N

NNNOHO.NNHNNNH NNu NO uuNN up Ou m& H NON NHNoz "Oz NH NNH O.

 

NmHmaNm NONquNHN>1mNONO NNN.NONNNONHN>_MON
NHNNHNN< ouNNumuNso NNN .NONNNUNHN omdoammm .NHNNNNONNNNNO

NNNNZZDAHNNDU mwmaqoo GZHmmw

iflmshdpyfiudx‘ UHQUUMNINZU 3:: ¢>LCC=IC$L~ -.-..
r—v— -.<7.?:o - ~ c.....~.-V‘ J: 1.1-... :6 svsfl Wv~.-1u

 

 

.1

 

 

181

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNH N L WNW NN NN N NH OHOuON.
NN H HH N, NH H N z
NN H O N NH H N N
NN H N H N N N O
*NN. NO OH NH ON N OH. NH N N
ON NN N N NH H H N N N H H 4
m Nx 4 uoz < l OHONOH N1 N N N. N N H ONNONO
“OHNNN. oumN NEED OOONONOJN
mafia. 55%? $95
HN NO. HOuON. NNH H N HN NN HN N ON 3.33.
N A NN 2 ON H N N NH N z
N A NN a NN N N OH N N N
N A NN. O NN N O OH N O
*O0.0 NO N NIH N A NH N NN H H N N N H N N
HO.V NN O N O V N < NH H O N H H N
a «x < 32 N n a .3 333 H O m N . H
“OHAON. 9.3 @230 madmuomuwn mmmaommum
NN.HNNH ZOHHNN HNH;
63mm? ONO .9258 N
HOONOOOHO N
unmfi N
ONONE HmHNnHOO HONNO N
N.N:H\NN OmONHmo N
HHON Nam NOON H

Numoa NONNO NON ON 0.32:. NO NN.—UH p.25 NNH u

 

moHOaum NONumNHHO>ummouo HONO NONNNNNHO> MON
mflmNHOON OuwNOmnNsO Nam .NOaONOONN OmOOOmON .ONNONEONOOONO

HMQZZOHHNMDU m0 M1500 OZHmnHm

 

 

 

 

 

 

INN NO mHuuoe NNH H H H NH mm, an N mHmuoa

NH An ON 2 mm H N NH NH N 2

ON An NH O NO N NH HH n O

NH » NH 0 ON H N O NH o

2 NH V NH O N H NH OH N. O

m N A N < OH H N m N H N
N-N O O-n mHmuoH x o N N N o m mOsouO

ONONuOOuNN mOmcomOON

 

NNNNNH ZOHHNNHAN>
NOON O OONBO OONu Ouoa

NOON O OOHBu uNONO

NOON O Ooao uNONO

meOB OOuNu NO can NNO>O.ONONO

x.OuN\NN mmOH no NuNOa.O OOOO uNOgO
HO>ON quaHO no .HO>ON

“\ONQOSO

NOON>OB ONN Ou ON NON ON NOuNO 30m NNH O

 

OOHOEON NONOONHHO>1OOOHU ONO NONOONNHO>_uON
mNmNHOON OOONOmnNNO NON .NOOONOOON OmOOOmOm .ONNOOOONONONO

ONHOZZOHHOOOO OOOHOOO OszOm

 

u—‘UA'Fu‘< (I'llll‘c ‘ F: I
v-quu.~.u..az,v “4244—..13

OH: 1- (ZZAJH‘.~. T. v.3...-

 

 

 

 

183

 

 

 

 

 

 

 

 

 

 

 

 

NN ON OHOuOH NNH N ON NN HN ON OHOuOH
OH V. NH 2 NN N N N N NH 2
OH uv NH N NN N N N NH O
OH An HH 0 NN N N N HH 0
.OHOEOO NH *An NH N ON N N N NH N
OONuOONHON aOHN OOONO>OH N AM OH N NH H H N N OH N
OON ANV NOHuOOHN0.00NNO
OOanOqu NON OHONOOnNno NuN N H mmmuos m N N N H OmNOHO
ONONOOONNN OOONONOON
NNNNNH ZOHHNNHNN>1NNONO
HN NN OHOuOH Nmm, M} NH NH OH me mHOuOH
NH An NN z ON H N. N N. NN 2
NH An HN N NN H N. N N HN N
S A NH O NN N m N NH O
OH.N NN NH NuN NH *uvv HH N NN N N N HH N
OH.uv HN HH H N An NH N NH H H H NH N
n Nx m 32 m N-N IO H 333 N N m N H «9155
OHNOH ONON NuNsO ONONOOONNN OOOOONOON

 

 

NNNNNH ZOHHNNHNNN
mOaON OaON ONO HHO

mOaON O30: ONO NO NHON NOsu Ouoa
N.NuH\NN mOaON ann ONO NO NHOn uNONO
mOaON ans ONu NO NHON NONO mmOH

HNMN‘I'

NHOON ONNu ONOqu NON ONO mOaON HHONqumOn ann Ofiu NO NONuHONOHO OOON NHONNom NNH O

 

OOHOEON NOHuOONHO>1mmOHO ONO NONuOONHO>.uON
ONONHOON.OHONONuHNO ONO .NusoNOOuN OmOOOmOm .ONHOONONumONO

NOHNZZOHHOOOO OOOOOOO ozHNOm

avahf-vu:< c“II1 UI fluid" I

 

 

 

 

 

 

184

 

 

NNN.N NN N unuNN uoxauu uoa NNOON.auz

 

mN NN O OOHNN OOxNOHNOOOOH.3Oz
Ov NN N 32 N

NN.N NO oN. umuNO Nassau uoa OONN NaNoom

no.uv oN mN unuNO Oaxauu «ONH NuNuom

 

.ml. x O uoz O
NONOON mummmm Nnu

 

 

 

 

 

 

 

NNH N ON. NN NN NH .NF NN ON, NN NH N NN NH NN NN N NN HN ON HN OHOOON.
ON N HH NH O N N. OH NH O N , N OH N . N NH N N N NH NH 2
NN N N HH HH N H NH NH N N H O N. N NH H N N OH NH O
NN H N N N N H N N N N H N N N O H N N N N O
NN H N OH N H OH OH N N N. N NH N N N N N
NH H N N N N. N N N O N H N N N O N
NNauoN N. N N N .N N N N N N . Naw N N N.. N . N. N N. N N Ommmmm
:OONONHN: \HH :ONNH HONOOO: \OH ..NN.N>NH O NN.NxOB: \O ..OOOO._.. RON: \N

 

 

NN.HNNH ZOHHNQHAN>

NN.N-HBO OuOOuOuNN Na OHONO ON? OONONHN 3ON
N.N..H\OH ONNH HONOOO ONO OONONEEOO ONOEOO
N.NuH\O . NNN>NH O NNNO—Oa NON NONNOOONOON. IIII
N.NuH\N OOOON 3ON NO. NONuOHNaNuO ONO

, TONNONNONOOONO HONNNNMO ONO NH NO>NN
ONONuOONNO ONONHNNO, Ouoa MN Ou H xNONv...OOONO OHNOHN ONO NN MOON—5N ONO ONO": NOON ON ON HHN3
NNNDOHHON ONO NO NOOO. MNNNN NON OO uNOquNaH so: .ONOHHOO. NN NOON ONSN ONO.» Ou OHOOMON NNNNOOA NNHnONH_ O

 

OOHNNON NONuOONHO>uOOONO ONO NONOOONHO> NON
ONONHONN ONONONINNO ONO .NONOOOOHN OONOOOON .OONONNONOOOOO

HNHNZZOHHNNDU momfloo UZHNnHm

‘qlill..v‘l‘ I l\ 1 ll

iii/\ZZ:1....2333 3.33.1.3: 32,—1...?"

 

 

 

185

 

 

 

 

NONNO N NO ON ON NN ON OO O ONOOON
ON O NO O NH O O ON N z
NN O. NO N NN N O ON O
NN N NO N ON N N. ON N O
ON NN N O O O ON N N
ON N N O O N O N N
NNNOON O N O N N. N N N ONO

 

OOONNO zOmNNONONN

O.N-N\NO

  
 

OOONONOON

Oaau OOON O HON uNO
HOOOOOHHOONN

OONON>NOOO HOONOO NN O>NuOO
RON: Ou uON Ou ONO:

NquO

NNOOOON
ONO: OHNNN ON ONO ONONu

HNMd’fiOh

NHOONOO uO ONO; OHNNN ONO NO uOOa.uNN OOHON OaOuN ONO NO NONNS

.OHOEOO NONOOONHO>
BOON OOONO>OH OOB NONuOOuNO
OONNO OOanOqu uON OHOOOOuNNo

 

NNOONN ZONNNONNN>1NOOOO

 

 

ONH N ON NN Om ON ONOOON

NN N O O NH O z

NN N NN O N. N O

NN N O ON N N O

ON NH O O N O

NH N N N Mn N N
mmwuoa N. N N

 

NNH U

NOON Ou On HHOS NNNOOHHON ONO NO NOOO NNNNO NON OO uNOuhONaN 3o: .ONOHHOO NN NOON uxON MOON Ou OHOOHON NNNxOON ONH O

 

OOHNNON NONOOOSO>OOONO ONO NONOOONHO> NON
39322 333.30 ONO .NONOOOONN OONONOOm .ONNONNONOOONO

NMHNZZOHBNNDO NUMAAOU UZHMNN

mdwzdwc< Uhfidcmldzu fits .>U:::t:$k :e1: I

n.:—h.<ZZNN73:22:} w....v.......:.... wOZ‘HJZLJNu

 

 

 

 

186

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we N NM a w ON MW. NN NN NN oN M NN NN ON oN m NN nN MN ONOuOH
NN N N N N N N N N M N N M M N M N N N N 2
MN oN n N N N M N N m N N m o M N m m w 9
MN 0 N N N N m M N N N M N N M n N 0
MN N N N N M N n N N M. o N N N N N m
o m N N N N N M N N N M N
fl Inflow N N N N N H N N Nfi N N N NF! N LN. N N N: N N NmmouO
.OuO .NNOxOn .OuO .NONM .OuO .Nudaw .NNH NO .3 ONuNu
NN wN NN MN m N mN mN NN NN N o mN eN oN NN wN oN MN ONOOOH
NN _ NN m n N o N N oN M m ON 0 a N. o m m z
NN M M M N N M N N N N N N N M n M n
NN N n N N m M M N M m N N N N o
NN w M N N N m N N N N M N M N N m
oN N N N N N M M N N N M n o M N N
ONOOOH N‘ N M N N N N M N N N N M N N N N M N N Omupuu
.OuO .xONM .OOO ..H.H.z .OuO .NONOONNMN .NNH NO .3 ONON
OBOE zohNflNNNSN
NONnOz OuOum NNONNNNH HO OwONNOU NHONOOOH ANOMOQ N.NnN\wN NOuNuOo no .NNNHONO .NONM
NOONNNO .NNNNOOO .xoaON _. N.N-N\NN .58 NodmaNNOO :NnNNN
MOOOO> .NONOONNOS .NuNam N.NuN\oN ONON .NHO>HOm..NOOOONNHm
NNSNNNN No 5383:: N.N-N\NN $833 No 5333::
AmNuva AmNONV
A.OHNONNONNNONN NONNNNNO NO NNHNw HON NOON OOOHONOm .ONNONNONumONN

NONNNNHO NO ONONuONHuNNN ONONNNxO Ono: n .ONOV .OONONO‘OOHNN HNON ON N "N Ou N NOHN xNOm
.EONO wNOaO OOOONO Ou NO: NON NN .OONONONOHN NNON NO mauOu NN mOwONNOO NNHBONNON ONu NNOM

omNummN o

 

OONNaOm NONuOvNNO>1NNOHU NNO NONNONNNO>.HON
ONONNON< ONONchNnu vNO .NoNONNOHm OONONNOM .OHNONNONONONO

EHZNKANHFN/N fii‘a....'r\ . u..
v..._<7__~&.v..NN..N.~_..: ...:..........-0.0 «:é—vr—u...

 

 

 

Nay-N—pq N __—

187

 

 

 

ONN N NO NN NN NO NN O NN ONONON
NN N NN N N ON N N ON z
NN N ON O N NN N N ON N
NN NN N O ON N NN O
ON N ON N N NN N N ON N
ON N NN N N O N O O
ONONON O N O O N N N N NNOONO
uwmfiommmm
O.N-NNNN ONNNOH zONNONNN<>
OONON>NOOO ONONNOO NN ONO NNON OO wNNNNNk N
NOONONNO NNNONOOONNOONN N NNoaNOONOO NON? NOOO ooO M
NONOOOON Nab NNONO NONO ONOE NNNNOOOO NONB NONNOoaou o OONNOO OOO N
MNNNOOOO NON? NONO ONOB NONOOOON N30 NNONO NON? NONNOONOO m NNNNONNN N

NNOONOO OO ONON ONONOOOO ONO NO Omoa ONN NONON OEOON ONO NO NONNS mmN U

 

 

 

wMN N mN ON mm 0N MN mm M ONOOOH
NM N mN NN NN m N mN 2
NM N NN m NN M m MN N a
NN N oN N NN N m NN o
mN wN w m N M wN N m
0N N m o o m m N <
ONOOOH m N o m N M N N ONNONo
OOONomOOM
mmNNNH onH<nNN<>
NONN OO OOw OO NNON N
wNNuN\wm OENO woow O NON ONO N NNONN M
NOONOONNOOON O NNONONNN N
OONON>NOOO NOONOO NN O>NOOO m ONON ONON oN ONO ONONO N

NHOOfiUm um OHQS m%0£ 030 MO “mos “HM 3OH0£ mENUH 05“ HO 30H£3 QMH U

 

OONNEOm NONOONNNO>IOOONU vNO NONOONNNO> NON
mNmNNON< ONONvmuNNu NNO .NoNONuONN OONONOOM .ONNONNONOOONO

 

OGNQEdm COfiumUHHG>ImmONU US“ SOHUQUHHQ> NOH
mfithflfl< @HQSUWINSU VCB .kUEQJUONk UQCOQWUM .UNHQSCONUQOJO

N.-v.<O7Nva_...T.N.:.~u u.-NNN.-.-.:~..u NVZH-nhun

 

«vmw

..'- . O -.-1 .. ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mMN N o N NN mm‘ MN ONOOOH
NM N M MN ON 2
NM N N N n NN N a
NN N N N NN N o
mm.N mm NN NIN OM N N MN m m
oN.uv MN 0N N NN N ON <
a NO z umm z mNauoN O N O N N N N N mmaouo
OOOOONOOM
mMNm<H zoNH<nHN<>ummOMo
MNN NN ONOOOH wMN N N NN Nm NN ONOOOH
NN uv ON 2 mM N N N ON ON 2
4, NM uv M 9 NM N NN M n
NN ”V N o NN N N N mN N o
M Nw.M Ow NN NuN 0N V m m mN N N N 3 m m
1. mo.uv NN 0N N NN Uv N O ON N MN N <
N‘ NO 2 umm z N-N a N ONOOON O N O O N N N N mmsouu
OOONOOONNO OOOOONOOM
mMNMOH ZOHHNONN<>
OOOOO NONNONOO No NONONOOONN O ON N
NONOOBunNOOOO N NNOONENOO NONONO>NOO O ON M
w.NuN\oo O80: OOO>NNO O ON 0 OO>NOONON NOOOO OON3 N
NONNONO No NONuONO NONONO>NOO O ON m OONONOOw No OOOONOO he OONS OEOO OO N

No>NN soN oO mums: OON o

 

OONmEOm OONOOONNO>IOOONU OOO OONOOONNO> NON
ONONNOON ONOOvmuNnu OOO .NoOOOuONm OOOOOOOM .ONNOOOONOOOOO

OONszoNNOmOo NOONNOO ozNOOO

 

E*E:Fqc< (KI‘IIlfU c I! I

. \ i . ., . an!
.3: N,<ZZC.~..~:...N..~...~. N....«N.-...=~.O OZ u 2.7.

 

 

.43?

 

rk..

 

 

.IONN NN OO O O N O NN N O O N ONBON
ON N ON N N N N O N N N N z
NN O NN N N O N N N O
NN N NN N N O N N N o
ON N O N N N N N N O
ON N N N N N O <

 

«NauoNNxomONOONNNNamsoNO
OOOOONOQM

 

muNNOH zoNHOQNN<>
h.x.mno\oo
Au>0NO OO OOOO OEOOV
NO>OO 50% OO OOOONNN OoOO NO NOOE 30m .ONO OOOONNN OOwOuOmONNOO OOON NOON 0&3 NO OOOONE O NON NONOH OONINON o

 

o ONN NN OO N N N N NN O N N N 338.

w. ON N ON N N N N S N N N z
N O NN N N NN N N O
NN N NN N O N N N N O
ON S N N O N N N O
ON N O N N N <

 

ONSONOOHO NO N O ONN N N «~85
OOOOONOOM

OONBO. 2323.3,
.3". TON NO

 

NON NO>O u x .ONnm ..OOO .MIN .NIN .NIOV OOOOOOOO
NOON omONO>O OO OO OOOOONE O>NN OOOON OO NON OONs NNOO OOO OOOO Och ow OOOOOOOO NOOE 30m NONIMON o

 

OONOEOm OONOOONNO>nOOONo OOO OONOOONNO> NON
ONONNOOO ONOOvmuNOo OOO .NoOOOvONm OOOOOOOM .ONNOOOONOOOOU

NNNszoNNONOo NOONNOO OZNOOO

 

33.7. .232... ti. .57. NO 1.7...

a. .1<ZZ~N TN...

 

 

 

an

 

‘-

4..A - N. .

 

 

 

 

 

 

 

 

 

 

ONN O N NN NN ON NN ON N N N O ONOOON
NN ON N O N O N N z
NN N O N O N O N N N O
NN N O N N N O N O
NOO.N ON O N-O ON N N N N N N ON N N O
.m.: ON NN O-O NN N N N N N N N N N <
O NO: N uoz < NquoN N‘ O O O N O O N N N N OOOoNO
UHn—mm. OHflDWWIHg womflommg
OONONN zONNNONN<>-OOOOO
ON OO ONOuoN ONN O N NN N ON NN ON NN N O N ONOOON
W. NN V NN z IN..N N ._N NN N N N N N N N z
1. ON UV NN O NN N NN N O N N N N a
NN u NN O NN O N N N N N N N O
NNN.N NN O N.O ON “v O O ON N N O N N N N N N O
ON.uv ON ON O-O O AnO ON < ON N N N N N N N N N <
O NO < ummx‘N N-O O O-O mNuuoN N O N O N O O N N N N OmsoNO
ONNOH ONO: muNOo OOONOOONNQ mOmOomOOm

 

 

mMNm<H onH<GHN¢>

N.mno\NO
aO>ONO mO OOOO OaOmV NOwONNOO OO ONO: OOOOOOOO ONO NOwOOOOONOOV OONONOOONO OOOB .OOOONNN OOOOO No NNNINON o

 

OONOOOO OONOOONNO>uOOONo OOO OONOOONNO> NON
ONONNOO< ONOOumnNOu OOO .NoOOOVONN OOOOOOOO .ONNOOOONOOOOO

OON<zzOHNOm=o NONNNOO OzNONO

 

ON... .<ZZC .9332? 3.53.7.3“. .JZ _ 2.3..

 

 

.:~.—-—.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ ONN N ON ON NN ONOOON
NO N NN NN O 2
MN N NN ON NO
NN O ON N O
OO. OO ON O ON ON NN M m
.O.O NN NN NuO NN NN N N O

O Na m Ooz m ONOOON O O N O O OOOONO

QHAQH. whgwmlfifiu ammgmug
ONNOOH zOHNOONN<>1OOOOO

NO NO ONOOON ONN NO NN O N ONOOON
NN uv ON 2 ON NN ON m z
2 NN V ON O NO .NN O N O
m NN V ON O NN NN O N N O
ON.N NN ON O ON .An ON O ON ON NN N N m
OO.V NN ON NuO NN V N O ON NN N <

a Nx O Ooz O O O NuO ONOOON O O N O O OOOONO

ONNOH ONOOdmINOU OOONOOONNQ OOOOomOOM
ONNOON ZONNOONN<>
mawlm\m.m mfiomarvm wdeOflHm GHQQOHH ONN m mfiomfiom mud—Hm find H mwaflumfiom 0
wfiomEOm HN.—"m dmu H NAHHNNNQNN h NOHQS mer m.“ UN. .mm% m

NOON OON3.8ONNONO OOO OOOOONO OO OOOOEOO OOw OO ONO: ON ON .OOOEONNOOO OOONO O NO>O NNOO OO OOOB 50% NH MNN U

 

OONOEOO OONOOONNO>uOOONO OOO OONOOONNO> NON
ONOONOOO ONOOOmuNOO OOO .NOOOOOONN OOOOOOOO .ONNOOOOOOOOOO

NONNzZONNOOOO OOONNOO OzNOOO

OONOEOO :oNuOONNO>Imm0No OOO OONOOONNu: stu
mHm%N@C< QNGSUWINSU UCQ NMUCQSUUNK UQCOQmUZ «GNHQECONOQQJO

v.1: H<7NZAJ .74. n .. mi 3.1. ”.3 «U u- J . =J u v .. :4 O‘ V: H.- a...

 

 

 

 

 

193

 

 

 

 

 

 

 

ON NO ONOuoN ONN ON ON NN ON ON ONOOON
NN An NN z ON N NN NN N N z
N .n ON O NO O N NN N O O
. O .n NN O NN N O O O N O
. OOO N ON O N ON uv NN O ON O O N N N O
OO uv ON O N-N ON ”v. O 4 ON N O N N N N
O NON <32 < N O N..N ONSON O N N N N 33.5
ONNOH ONOOOmuNOU OOONOOONNQ OOOOONOOM

 

mNAMOH ZCHH<QHN<>

ONNONNO>O OOO I O OOO .NON OOO n N .NON OO KOO n M .NOO OO NON u N .NON.BOOOOA u N "ONNONOONU .O.N NN .NOO

 

 

OMN M NO Om NN NM ONOOOH
OM NN ON M O 2
NM NN NN N N a
NN MN O N O 0
ON M NN N N N m
ON O O N O <
ONOOOH O m N O O N OOOONU

OOOOomOOm

 

mmNm<H ZCHH<QHN<>
:OOOOn: No ENOO OOO OO ONOO OOONNN O O

O.OIN\NN OOOOO NO OOoO w NOOOONONO NO NOOOONOOON OOO O
OOONO EONN OOON N OOOOBOO N OOOE BOON N8 N

NEONNONO O OOOO OOONO OONOVNNOOKONNOOOO OOO ON 053 NNN u

 

OONOOOO OONOOONNO>umOONO OOO OONOOONNO> NON
ONOONOOO ONOOOOuNOO OOO .NOOOOOONN OOOOOOOO .ONNOOOONOOOOO

OONszONNOOOO OOONNOO OzNOOO

”a: ,_.<ZZC NN.N..nOOZNNv. 33.....7: N., N .QZ NINE

 

 

 

 

- -

-_.—{

 

194

«NO.N NN ON N
.O.O OO O MIN

 

O} Nx N uoz N

 

mNOON. ONOOOO-NOO

 

OONOOHXZONONONN<>-OOOOO

mNOONNOS OOO u O OOO .NON OOO u N .NON OO NOO u N .NOO ou NON u N .NON OOOOOO u N

 

 

OMN N NO NN NN ON ONOOOH
NM O MN ON N O 2
NM N NN N O N O
NN M O N N O O
OM NN O O N m
NN N ON N N N <

ONOOOH O N M N N Omwouo

mmmaommmm

"ONNONOOO .O.N NN .Noo

 

OONOaOm OONOOONNO>IOOONO OOO OONOOONNO> NON
ONONNOO< mumsvmuNnO OOO .NoOOOOONm OOOOOOON .ONNOOOONOOOOO

9*4“F‘D‘) ‘- "'f

v— “UNA. —. :3 u uyvz .~ I.— m...

E: H<ZZD fifln O: .3

 

 

 

 

Jl‘.
. ,.v‘

195

 

 

 

 

 

 

OO NO ONONOO ONN NN ON ON NN ON O NN O NN ONOuoa

ON uv NN 2 ON O O O N O N N O 2

ON uv NN O NO O N N N NN N N O O

NNN.O OO N ONO NN ”v O O NN N N N N O N O u

N N ON .uv O O ON O N N N N N N N N O

OO.uv ON NN N+O N An NN N ON N O N N N N N
O} NO N uoz N O+ONN-N O. N+O ONONOO O O N O O N N N N OOOOOO

ONOOH mumswmuNso OOONOOOHNG OOOOONOOM

 

mmNm<H ZOHH<GHN¢>

A.O.N HO OONummdouo.O.Hm£umm do .m.: ..O.NV OOOONOGN OOO n O

m M .N N .000 m.um£umm

M N

DU
39::

 

N N N N
ONNOOOOO .O.N

OONOOOOOOO O.Nmnuwm OOO .o.N O.uOOOOum ON .NOO

 

OONOaOm OONOOONNO>1OOONO can OONOOONNO> Now
ONONNOON ONOOOOINOO OOO .Noaosvmum OOOOOOON .OHNOOOONOOOOO

532205520 NONNNOO OZHNNNO

 

 

 

7‘1 $1, ‘7,

'.

 

‘ '11.,
,...,.. .‘- _

w—v_‘_1<ZZNv Niki—2N»

H— ” \.‘-

-\O\

Oil!

 

 

196

 

 

 

 

OMN O OM NO ONOOOH

OM M NN ON z

NM ON NN a

NN N N ON 0

ONO.N OM N N om ON N O ON O

.O.a OO NN N 03 ON N NN <
O Nx < qu N ONOuOH O N N Omaouo

ONOOH OuudmwuNnu OOOOONOOO

 

OMNM<H ZOHH<DHN<>

O n ONOONNO>O OOO OOO .N n AU Ov HONNOO OONO .N n Au 3v NONNOO muan NOONOOOOOOO m_umaumm ON .NOU

 

 

 

 

ONN ON NN ON ,NN ON O NN O ON ONOOOO

NN NN N N N N N N z

NN N O N N N N O N N O

NNN.N OO O O+O _ NN N N N N N N N N N O

N-N ON N N O O N N N N N O

.O.O ON ON N+O NN N N O N N N N N
O NO Nwmoz N ONOOOO O O N O O N N N N OOOOOO

ONOOH mhmauOuNnu mumdommmm

 

mmNm<H ZOHH<QHN<>umeMU

OONOOOOOOO O.nmnumw OOO .O.H m.uOOOOuO ON .NOO

 

OONOEOO OONOOONNO>uOOONO OOO.OONOOONNO> you
ONONNOON OOOOOOuNOO OOO .NoOOOOOum OOOOOOOO .OONOOOONOOOOO

OONszoNOOOOO OOONOOO OzNOOO

ON:_<ZZC-N..N.N...N.:.3 N.N..x....‘._...3 NOZNZAON

 

 

 

9;: a

 

 

197

 

 

 

 

 

 

 

 

 

OMN NON O ON ONOOOH
NM OM N O 2
NM NN M N a
NN ON N O O
*NMO. NO NN M OM MN N O O
.O.O NN N N NN MN N 4
O NN a qu O ONOOOH .0.0 M N N OOOONO
ONOOH mumndmuNsu OOOOONOOO
OMNONH ZOHH<GHN<>uOOOOU
OMN N NON O ON ONOOOH
OM ON M O 2
NM NN N NN a
NN N ON N O O
ON.N OO NN M ON ON N N O
ON.uv ON NN N ON ON N
m‘ Nx a uoz O ONOOOH .m.c M N N Omwouw
ONOOH OHOOOOINOO . meaommmm

 

OMNONH ZOHHNGHN<>

OOHON u M OOO .ANOOOOO nONm ONOO ONNOONOOV O NOOOOO I N .NNOEO u N NONNO NOOOOO anm NN .NOO

 

OONOEOO OONOOONNO>IOOONO OOO OONOOONNO> HON
ONONNOO< ONOOOOINOO OOO “Noamsvmum OOOOOOOO .ONNOOOONOOOOO

OONszoNOOOOO OOONNOO OzNOOO

3.: 3x223 NN.:ONNN? 3.53.1: N... 3.2.1.2..

 

 

 

 

'13.? “

 

198

OMN NO NN ONOuOH

 

 

 

 

 

 

 

 

 

NM NN ON 2
NM ON NN O
NN O ON 0
NNN.N ON NN ONONO ON NN ON O
.O.a OO O ONOO NN ON N N
O» NM 2 qu z ONOuOH ONNNO ONOO OOOOHO
ONOOH ONOOOOINOO Ommaommmm
OONONH ZOHHNOHN<>uOOOOO
OMN OO NN ONOOOH
OM NN ON 2
NM MN NN O
NN MN NN O
ON.O OO NN ONONO ON NN NN O
NOOv NN ON OOOO ON O ON N
O NN z qu z ONOOOH ONHNO» whoa OOOOHO
ONOOH mumsdmuNsu mmmdommmm

 

OMdeH ZOHH<QHN<>

ONHNO n OuO OOO ONOO u NuO xmm m.uOOOOum N .Noo

 

OONOEOO OONOOONNO>1OOONO OOO OONOOONNO> NON
ONONNOON ONOOOOINOO OOO .Nocmswmum OOOOOOOO NONNOOOONNOOOO

OONszoNNOOOO OOONNOO OzNOOO

 

199

APPENDIX B

An Illustration of Hemingway's
Multiple Agreement Analysis

 

1Peter W. Hemingway, "Multiple Agreement Analysis," (unpublished Ph.D.
dissertation, Department of Psychology, Michigan State University, 1961),
30-35.

 

 

 

 

200

Heminggay's Nine Major Steps pi Multiple Agreement Analysis

"A hypothetical example will be used to illustrate each of these
steps. The responses of 9 subjects to 6 binary items are given in Table

1 (Y denotes a yes response and N a no response).

Step 1, The agreement score (number of identical responses)
between each person and every other person in the response matrix is
computed. This requires ngn-lz computations. These scores are listed

for the example in Table 2.

Spgp g. That pair of persons with the maximum agreement score
is selected as the starting point for the initial obtained class. In
the event of;a tie among two or more pairs for maximum agreement score,
an empirical test has revealed that the same structure is obtained
regardless of which pair is used. Hence the computer program arbitrar-
ily selects the last pair to attain the maximum as the starting point.
The mathum scores are circled on the agreement matrix of the example.
Pair'HI, being the last computed, was selected as the initial starting

point.

§ppp‘§. Those responses upon which this pair agree are
selected to form the initial scoring key. This scoring_key is then
used to compute the agreement scores of all remaining persons with the
initial pair. In the example, this would be all 6 of the I and H

resPonses.

Step 3. That person agreeing most highly with the initial pair

is tentatively chosen as the next member of the class. Again ties are

3P5

 

 

201

TABLE 1

THE RESPONSE MATRIX

 

 

 

 

 

 

 

 

 

 

Subjects
Items A If c D N E F *G 11 T
1 ‘ Y YT Y A N N N.
II
2 N N N Y Y Y
I
3 Y Y Y N N
a E Y Y' N N N N
IV . _
5 N N Y Y Y
: III
6 Y N N Y Y Y,

 

 

 

TABLE'Z

The Original.Agreement Score'Matrix

Subject A B C D E F G H I
A - 5 5 4 4 1 1 1 1
B - (9 3 3 o o 1 1
c - 3 3 o o 1 1
D - @ 3 3 2 .2
E - 3 .3 2 2
F - © 5 5
c - 5 5

202

settled by the rule that the last person to tie for the highest agree-
ment score be the one selected by the computer. Thus person G, with an

agreement of 5, is the first tentative choice.

St p,§. The products of the two sets are now compared. If the
inclusion of the new person does not reduce the information accounted
for by the class, he is accepted as a class member. If the product is

less when he is included, his classification remains tentative.

gppp g. .A new scoring key is now prepared, based on the
responses common to the augmented set of persons, regardless of the
outcome of step 5. Steps 3, 4, 5, and 6 are repeated until all persons
who agree with the current scoring key on at least two responses are
tentatively included in the set. In the example, persons“G and F are
included and the procedure terminated, as no remaining subject agrees

with the four-item scoring key on more than one item.

§£gp.z. That point in the formation of the set with the maxi-
mum product is now'chosen as the best estimate of a hypothetical-class.
.Again ties are settled by taking the last maximum. The computer there-
fore prints out the person and the response pattern which form this
obtained class. In the example, this corresponds to the class FGNI,

with response pattern NYNN--. The product of this class is 16 (4 sub-

jects thmes 4 responses).

Step'g. The submatrix corresponding to the forst class is now
eliminated from the original response matrix. This is in accordance

with the requirement that no response be used more than once to

 

203

classify the same subject. In the example, this corresponds to the

elimination of that labeled I in Table l.

Spgp.2. Steps 1 through 8 are repeated on the reduced response
matrix. This cycle of Operation is repeated, with each cycle defining
a new class, until there remains in the response matrix no agreement
score equal to or greater than some predetermined criterion. For the
example, these classes are presented in Table 3 in their detail of
formation, but without showing the recomputation of the agreement
matrix of Table 2 for each cycle.

The-complete analysis of the example, with the criterion that
no agreement score less than two will be used in forming classes,
results in the class structure listed in Table 3. Several character-
istics of the method are illustrated. First, the first two classes
classify all the subjects, giving a complete classification correspond-
ing to major mutually inclusive classes. Second, class 3 is a cross-
-classification, containing members of both major classes. Third,
classes 4 and 5 are sub-classes of the major classes. The responses
defining the major classes are those listed. Those-defining the sub-
‘classes include those listed plus those-defining their respective major
classes. The cross-clas is defined by those responses listed, plus a
No response to Item-4, which would not be realized as it had been
previously used to classify the class 1 subjects. Thus, patterns of
sub-classes consist of their common responses plus the common responses
of theierajor classes, while cross-class patterns can be completely

determined only by inspection of the original response matrix."

 

 

204

TABLE 3

‘MULTIPLE-AGREEMENT ANALYSIS: CLASS STRUCTURE OBTAINED

First Classification: Output

HI - 6 Product - 12

HIG - 5 Product - 15

HIGF - 4 Product - 16 Max. FGHI Key: N Y N N * *

Second Classification:

DE - 6 Product 12

DEA --4 Product 12
DEAC - 3 Product - 12

DEACB Product - 15 Max. ABCDE Key: Y N Y * * *

Third Classification:

DE - 3 Product - 6
DEC - 2 Product -~6
DEGF - 2 Product - 8 Max. DEGF Key: *'* *'* Y Y
Fourth Classification: ABC Key: * * * Y N *

Fifth Classification: HI Key: *‘* *‘* Y N

 

 

» ""313: 77:2 '5‘;—

" '1‘ v

 

llHllWlU Illlllll

3 1293