Title: Status inconsistency and mental health
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098867/00001
 Material Information
Title: Status inconsistency and mental health
Physical Description: xvii, 248 leaves : ; 28 cm.
Language: English
Creator: Fuson, James B., 1949-
Copyright Date: 1976
 Subjects
Subject: Role conflict   ( lcsh )
Social status   ( lcsh )
Adjustment (Psychology)   ( lcsh )
Sociology thesis Ph. D   ( lcsh )
Dissertations, Academic -- Sociology -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: by James B. Fuson, Jr.
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 227-247.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098867
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000179007
oclc - 03145466
notis - AAU5521

Downloads

This item has the following downloads:

PDF ( 11 MBs ) ( PDF )


Full Text










STATUS INCONSISTENCY AND MENTAL HEALTH


BY

JAMES B. FUSON, JR.























A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE
OF DOCTOR OF PHILOSOPHY





UNIVERSITY OF FLORIDA


1976






























































Copyright

1976

James B. Fuson, Jr.


























Dedicated to

my

teachers














ACKNOWLEDGMENTS


The race of mankind would perish did they
cease to aid each other.--We cannot exist with-
out mutual help. All therefore that need aid
have a right to ask it from their fellow-men;
and no one who has the power of granting can
refuse it without guilt.

Walter Scott



The above quotation is, of course, another way of

expressing John Donne's well known concept that "no man is

an island." In the process of obtaining an education and

completing this dissertation I have certainly had to call

upon many people for help and have not been refused.

Acknowledging all who have provided assistance would be an

impossible task. Therefore, I shall restrict myself to the

formidable task of acknowledging those who have been most

influential in my educational process or have contributed to

the preparation of this dissertation.

My parents inspired my original motivation to learn, to

attempt to expand my knowledge beyond what appeared to be

the obvious. They always supported my educational interests

regardless of their direction as those interests shifted

from mechanical engineering, to chemistry, to philosophy, to

history, to political science, to sociology. Throughout this

entire period of shifting interest the support of my












parents, both emotional and economic, remained steadfast.

My first interest in sociology as a discipline was

stimulated by Marvin Henricks at Indiana Central University

under whom I had much of my undergraduate sociology. He has

left a lasting influence on my view of the social world,

particularly concerning the influence of culture on human

behavior.

My education was further molded and refined at Ball

State University where I was particularly influenced in the

area of social stratification by Bernard Blakely. At the

same time, Whitney Gordon provided a new perspective on

social philosophy and humanistic sociology. I am deeply

indebted to Joseph Tamney who was chairman of my Master's

committee and instilled a great respect in me for the

relevance and complexity of social theory.

At the University of Florida the mystery of statistics

was unravelled for me by Alan Agresti. This process was

further assisted by Charles Holzer who in addition to

increasing my knowledge of statistics taught me all I know

concerning computer programming and made a significant

contribution to my understanding of methodology. I am also

heavily indebted to Benjamin Gorman for his teaching in

methodology, theory and social stratification. Anthony

LaGreca has also contributed to my knowledge of methodology

and urban sociology. Gordon Streib has given generously of

his knowledge of social gerontology. Felix Berardo has











provided insight in the areas of the sociology of the family

and social theory. He served as my Ph.D. committee chairman

through the completion of the written and oral qualifying

examinations. In his own unique way, and I believe unknown

to himself, he provided the stimulation which pulled me

through the first year of the Ph.D. program.

Gerald Leslie developed my first real interest in the sociology

of the family and as chairman of the sociology department

during the interim of my Ph.D. work served as a constant

source of advice and encouragement both personal and

academic. I am indeed heavily indebted to George Warheit

who has served as my Ph.D. committee chairman for the writing

of this dissertation. He has heavily influenced my thinking

in several areas, including social organization, social

theory and especially in medical sociology. I should also

like to thank him for making available to me the data on

which this dissertation was written. In this connection I

should also like to thank John Schwab who was the original

principal investigator on NIMH GRANT 15900 and Roger Bell

who was the original principal investigator on NIMH Contract

HSM 42739 OC. Without the work of these two men in

gathering the data this dissertation could not have been

written.

Last, but certainly not least, I extend my loving

gratitude to my wife, Becky, whose patience, understanding

and helpful comments have greatly facilitated the completion










of this work.

In short, I have called upon many for help of one kind

or another, including many who are not listed here. I have

been fortunate to be among those who have received help when

they needed it. These people who have given so generously

of themselves in my educational process I consider to be my

teachers, including my parents and wife. Thus, I wish to

sincerely thank my teachers for making my education possible

and the rewarding experience it has been and I hope will

continue to be.















TABLE OF CONTENTS




Page

ACKNOWLEDGMENTS ................ . . iii

LIST OF TABLES . . . . . . . . . . . . IV

ABSTRACT . . . . . . . . . .. ... .. xiii

CHAPTER I INTRODUCTION . ............... 1.

CHAPTER 2 LITERATURE REVIEW . .. ..... ... .. 8

Political Liberalism ......... . . . . .. 10
Democratic Vote ...... . . . .. .. .. 11
Political Change . . . .. ...... . . 12
Presidential Assassins . . . . . 12
Right-Wing Attitudes .... . . . 13
Power Distribution . ........... . . 14
Prejudice .... . . . ... .. .. .. . .. 14
Social Participation . . . ... .. ...... . 15
Individual Unrest . ................. . 15
Job Satisfaction ...... . . . . .. . 15
Self-Esteem ........ . . . . .. .... 17
Voluntary Organizations . .. . ......... . 17
Decision Making . ... ............... 18
Peptic Ulcers . . .. .. ... .. ....... .. 18
Coronary Heart Disease ... .. .......... 19
Rheumatoid Arthritis . .... ..... .. . .. 20
Morbidity .................. . . ... 21
Social Stress . . ... ................. . 21
Mental Health.... .. .. . ............ 27
Sight of Flying Saucers .... ............ . 31
Summary . . . . . . .. . . . . .. 32

CHAPTER 3 THEORY . . . . .... ... .... . 33

Symbolic Interactionism and Role Theory ... . . .. 34
Structural Functionalism . . . . . . . ... 37
Distributive Justice ..... . . . . . . 39
Expectancy Congruence . .. ............ 43
Relative Deprivation . . . . . . . . .
The Predominant Process in the Relationship Between Status
Inconsistency and Poor Mental Health . . . . 46
Social Stress . . ..... . .. ... . . 52
Theoretical Propositions ... . ........ . 57











Table of Contents
(Continued)






CHAPTER 4 METHODOLOGY .. . ....


Methodological Criticisms of Status
The Data . . .
Measurement of Status Inconsistency
Measurement of Mental Health . .
Control Variables .. . ....
Statistical Procedures ..
Summary . . . . . . .

CHAPTER 5 FINDINGS AND DISCUSSION .


Inconsistency


Status Inconsistency and SES . . . . . . 110
Thirteen Status Inconsistency Types and Mental Health 113
Three Status Inconsistency Types and Mental Health . . 115
Four Status Inconsistency Types and Mental Health . .131
Six Status Inconsistency Types and Mental Health . . 146
Analysis of Propositions. . . . . . . . .169

CHAPTER 6 CONCLUSION ...... . . . . . . .176

Summary . . . . . . . . . . . . . 176
Conclusions . . .. . . . . . . . . 179
Future Research . . . . . . . . . . . 181
Implications . . . . . ... . . . .. .183

APPENDIX A Income, Education and Occupation Scores . . .189

APPENDIX B Six Measures of Mental Health .. . . . . 207

APPENDIX C Thirteen Status Inconsistency Types by Six
Measures of Mental Health . . . . . ... .221

BIBLIOGRAPHY .. . . ..... . . ... . . 227

BIOGRAPHICAL SKETCH ... . ....... . . .


Research.
















LIST OF TABLES


Table

1. Thirteen Status Inconsistency Types . . . . . .

2. Three Inconsistency Types . . . . . . . .

3. Four Status Inconsistency Types . . . . . . .

4. Six Status Inconsistency Types . . . . . . .

5. Mental Health Measures, Their Ranges and Corresponding
High Score Levels . . . . . . . . . .

6. Correlation Coefficients for Education, Occupation
and Income . . . . . . . . . . .

7. R Values for the Regression of Education, Occupation and
Income on Six Measures of Mental Health . . . . .


8. Three Status Inconsistency Types by SES

9. Three Status Inconsistency Types by the
Survey Score . . . .

10. Three Status Inconsistency Types by the

11. Three Status Inconsistency Types by the

12. Three Status Inconsistency Types by the
Score . . . . . . . .

13. Three Status Inconsistency Types by the
Breakdown Score . . . . . ..

14. Three Status Inconsistency Types by the
Score . . . . . .

15. Prediction of Health Opinion Survey Scor
Status Inconsistency Types, Control Vari
Interaction Terms .. . . .. ..


Scores ..... ll

Health Opinion
. . . . 116

Depression Score 117

Mood Score . . 117

Anxiety Symptom
. . . . 117

Worry, Nervous
. . . 118

Psychopathology
. . . . 118


'es from Three
ables and


Page

79

80

82

85


93


94



95











Table Page

16. Prediction of Depression Scores from Three Status
Inconsistency Types, Control Variables and Inter-
action Terms .................. . . 24

17. Prediction of Mood Scores from Three Status Incon-
sistency Types, Control Variables and Interaction
Terms . . . . . . . . . . . . . 125

18. Prediction of Anxiety Symptom Scores from Three
Status Inconsistency Types, Control Variables and
Interaction Terms . . . . . . . . . . 126

19. Prediction of Worry, Nervous Breakdown Scores from
Three Status Inconsistency Types, Control Variables
and Interaction Terms. ... . . . . . . .127

20. Prediction of Psychopathology Scores from Three
Status Inconsistency Types, Control Variables
and Interaction Terms .... . . . . . .128

21. Four Status Inconsistency Types by the Health Opinion
Survey Score . . . . . . . .... . . . 132

22. Four Status Inconsistency Types by the Depression Score . 132

23. Four Status Inconsistency Types by the Mood Score .... .133

24. Four Status Inconsistency Types by the Anxiety Symptom
Score .. ....... . . . . . . . .133

25. Four Status Inconsistency Types by the Worry, Nervous
Breakdown Score . . . . . . . . . 134

26. Four Status Inconsistency Types by the Psychopathology
Score ........ . . . . . . ..... 136

27. Prediction of Health Opinion Survey Scores from Four
Status Inconsistency Types, Control Variables and
interaction Terms .. . . . . . . .... 139

28. Prediction of Depression Scores from Four Status
Inconsistency Types, Control Variables and Inter-
action Terms ... .... . . . ... . . 140

29. Prediction of Mood Scores from Four Status Inconsistency
Types, Control Variables and Interaction Terms . . . 141

30. Prediction of Anxiety Symptom Scores from Four Status
Inconsistency Types, Control Variables and Inter-
action Terms . . .. . . . . . 142










Table Page

31. Prediction of Worry, Nervous Breakdown Scores from
Four Status Inconsistency Types, Control Variables
and Interaction Terms ... . . . . . . 143

32. Preri. Urin of Psychopathology Scores from Four
Status Inconsistendy Types, Control Variables
and Interaction Terms ..... . . . . . . 144

33. Six Status Inconsistency Types by the Health Opinion
Survey Score .. . . . . . . . . . .148

34. Six Status Inconsistency Types by the Depression Score . 149

35. Six Status Inconsistency Types by the Mood Score ... .150

36. Six Status Inconsistency Types by the Anxiety Symptom
Score .. ..... . . . . . . ..... .. .151

37. Six Status Inconsistency Types by the Worry, Nervous
Breakdown Score . . . .... .. ... . . 152

38. Six Status Inconsistency Types by the Psychopathology
Score . . . . . . . ... . . . 153

39. Prediction of Health Opinion Survey Scores from Six
Status Inconsistency Types, Control Variables and
Interaction Terms . . . . . . . . . . 156

40. Prediction of Depression Scores from Six Status
Inconsistency Types, Control Variables and Inter-
action Terms ...... . . . . . . . .157

41. Prediction of Mood Scores from Six Status Incon-
sistency Types, Control Variables and Interaction
Terms . . . . . . . . ... . . . . 158

42. Prediction of Anxiety Symptom Scores from Six
Status Inconsistency Types, Control Variables and
Interaction Terms . . . . . . . . . . 159

43. Prediction of Worry, Nervous Breakdown Scores from
Six Status Inconsistency Types, Control Variables
and Interaction Terms . . . . . . . . 160

44. Prediction of Psychopathology Scores from Six Status
Inconsistency Types, Control Variables and Inter-
action Terms ... . . . . . . . .161











Table Page

45. Beta Coefficients on Six Measures of Mental Health
for the Variables of Underrewarded, Consistent and
Overrewarded Who Perceived Themselves as Under-
rewarded After Considering the Interaction with Age
as Calculated from Tables 39 Through 44 . . . ... .166

46. Thirteen Status Inconsistency Types by the Health
Opinion Survey Score . . . . . . . . ... 223

47. Thirteen Status Inconsistency Types by the Depression
Score . . . . . . . . . . . . . 222

48. Thirteen Status Inconsistency Types by the Mood Score . 223

49. Thirteen Status Inconsistency Types by the Anxiety
Symptom Score . . . . . . . .. .... . 224

50. Thirteen Status Inconsistency Types by the Worry, Nervous
Breakdown Score .. . .. . . . . . . 225

51. Thirteen Status Inconsistency Types by the Psycho-
pathology Score . . . . ... . . . 226


xiii









Abstract of Dissertation presented to the Graduate Council
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



STATUS INCONSISTENCY AND MENTAL HEALTH

BY

James B. Fuson, Jr.

December, 1976

Chairman: George J. Warheit
Major Department: Sociology

The purpose of this research is to examine the rela-

tionship between status inconsistency and mental health.

Since Max Weber first set forth his ideas concerning the

relationship between class, status and power, sociologists

have been enamored with the prospect of a multidimensional

approach to social stratification. Within this context, the

notion of status inconsistency has been a controversial

research topic in the recent sociological literature.

In this research, literature was reviewed which re-

ported the relationship between status inconsistency and

twenty other variables. Additionally, methodological

criticisms of status inconsistency research were reviewed,

and the procedures employed in this analysis for dealing

with those issues were explained. The three variables of

education, occupation and income were measured according to

a procedure developed by the U.S. Bureau of the Census and a

status inconsistency measure developed by the same agency










was utilized. Mental health was assessed on six different

measures. The analysis controlled for the effects of race,

sex, SES, age and marital status on mental health, in order

to determine to what extent the findings could be attributed

to status inconsistency as opposed to other status vari-

ables. Utilizing multiple regression procedures, these

control variables were allowed to explain all the variance

in mental health scores attributable to them before the

status inconsistency variables were introduced to determine

what portion of the remaining variance was attributable to

status inconsistency.

The data for this analysis were obtained by combining

two epidemiological field surveys collected between 1970 and

1973 in central and north central Florida. In each case,

the procedure for selecting the sample was a systematic

probability method with interviewers utilizing the Kish

method for selecting respondents within household.

The general theoretical approach of this analysis was

the structural functional perspective which maintains that

it is functional for the social order if positions requiring

greater training (investments) are more highly rewarded.

Within this general frame of reference the perspective of

distributive justice as explicated by George Homans provided

a more specific approach emphasizing the proportionality of

investments and rewards. According to distributive justice,











discrepancies between investments and rewards are experi-

enced as an uncomfortable or stressful condition. The

stress experienced from being status inconsistent would then

be expected to have a negative influence on mental health.

In this analysis, education and occupation were considered

as investment variables and income was considered as the

reward variable. The findings supported the theoretical

expectations, status inconsistency of an underrewarded type,

rewards below investments, was found to be associated with

poor mental health.

A serendipitous finding was found when the interaction

between SES and overreward was considered, the overrewarded

type of status inconsistency actually had scores indicating

better mental health than the status consistent. As SES

increased the mental health scores indicated better mental

health for the overrewarded than for the status inconsis-

tents.

The predominant process in the relationship between

status inconsistency and mental health has been a debated

issue. Some maintain that status inconsistency is a

stressful condition which has a negative effect on mental

health; others argue that poor mental health is likely to

result in status inconsistency. Two independent tests were

developed and executed to aid in determining the predominant

process in the relationship between status inconsistency and











poor mental health. The results of both tests clearly

favored the first position that status inconsistency has a

negative effect on mental health.

A serendipitous finding as a by product of the second

of these tests was that the perception of underreward was a

better predictor of poor mental health than more objective

measures of underreward after the interaction with age was

considered.


xvii














CHAPTER 1
INTRODUCTION



From the beginning of man's existence his social order

has included a hierarchical status ranking of individuals.

Certainly man's biological predecessors have a known pecking

order, or dominance-submission pattern. A question which

has intrigued social scientists and others is, What are the

effects for man when the pecking order is uncertain, when

the dominace-submissive patterns are unclear? This work

deals with this issue. More specifically, it is designed to

examine the relationship between status inconsistency and

mental health. It asks, What are the implications for a

person's mental health when the statuses or positions which

he holds in society are decided by unequal or inconsistent

criteria? Does ranking high on some statues, e.g., educa-

tion, occupation, or income, and ranking low on other sta-

tuses, e.g. education, occupation, or income, result in a

stressful condition which has a negative effect on mental

health? In investigating the relationship between status

inconsistency and mental health, this work will deal with

the following issues. Is there any association between status

inconsistency and mental health? If so, how strong is the

relationship and are they positively or negatively related?











Additionally, if such an association does exist, are there

any means whereby the direction of the predominant process

in the relationship may be assessed?

Some of man's earliest recorded history has shown that

the issue of social stratification, the ranking of indi-

viduals or groups in a vertical manner according to some

criteria, is as old as man. In fact, in some sense,

Original Sin as recorded in Genesis was related to the

system of social stratification. The serpent told Eve that

if she ate of the tree of the knowledge of good and evil she

would become as God. Therefore, instead of being ranked

below God on the criteria of knowledge of good and evil,

she would be ranked equal with God. Thus, Original Sin was

an attempt to circumvent the existing stratification system.

Aristotle, in his work Politics also recognized the

existence of a social stratification system. He wrote:

Now in all states there are three elements:
one class is very rich, another very poor,
and a third is a mean. It is admitted that
moderation and the mean are best, and there-
fore, it will clearly be best to possess the
gifts of fortune in moderation; for in that
condition of-life men are most ready to fol-
low rational principle. But he who greatly
excels in beauty, strength, birth, or wealth,
or on the other hand who is very poor, or
very weak, or very much disgraced, finds it
difficult to follow rational principle.
(Aristotle, 1943:190)

In the latter part of the quote, Aristotle appears to be

giving credence to a multidimensional approach to social

stratification when he lists various areas in which a person










may excel rather than a unidimensional approach in which one

area is emphasized.

Probably the person generally considered to be the

leading exponent of the unidimensional approach to stra-

tification is Karl Marx. It was Marx's conviction that a

person's position in the stratification system was deter-

mined by his relationship to the means of production. For

the most part this meant whether one was an owner of the

means of production or a worker in the means of production.

Marx held this principle to be true regardless of what the

means of production might entail.

Without becoming overly involved in Marxian theory, it

should be pointed out that Marx believed classes started

from a position of class an sich or "class in itself" and

moved to a position of class fur sich or "class of itself"

(Barber, 1957:219). In the former state the class was not

conscious of itself; in the later state class consciousness

had developed. Under class consciousness, the class took on

an almost community-like quality in which the class recog-

nized its interests and acted to achieve them.

Max Weber reacted to the writings of Marx. Weber

maintained that classes do not achieve the type of class

consciousness attributed to them by Marx; class members may

unite on transient economic issues, but they do not achieve

the level of consciousness Marx indicated. More important










for the issue under consideration here, Weber introduced the

first significant statement of a multidimensional approach

to social stratification. Weber recognized three dimensions

of stratification: class, status, and party (Gerth and

Mills, 1946:18-95). Class was basically an economic dimen-

sion; it represented the economic order. Status was

basically an hcror or prestige dimension which represented

the social order. Party was basically a power or political

dimension which represented the legal or political order.

Since the time of Weber, scholars have been giving

increasing credence to the multidimensional view of stra-

tification. Warner et al. (1949) constructed an index of

social class based on a multidimensional approach.

Stonequist (1937) in The Marginal Man, expressed the notion

that the marginal man was between statuses, having some of

the characteristics of a particular status but not all of

them. Sorokin (1947) also viewed social stratification from

a multidimensional perspective. He pointed out that the

individual's rank within a group was dependent upon his

relationship to the primary bond which held the group to-

gether. Thus, in a religious group, the individual's

religious status was important in determining his position.

In an economic group, the individual's economic status was

the important factor. Sorokin also noted, however, that

many groups were of a multibonded nature, and thus, more

than one status of the individual became important in










determining his position. Similarly, when discussing caste

Sorokin noted that:

The upper and lower castes are superior or
inferior not on a single basis but on a
multiple basis. The Brahmins are superior
to the Sudras in race, religion, occupation,
kinship, language, education, and so on.
Likewise, the social estate of the nobles
in comparison with that of the free popula-
tion and slaves is superior not for one
reason, but for three: occupation, econo-
mic, and state. (Sorokin, 1947:289)

One of the first persons to recognize the multidi-

mensional nature of stratification and to point out that

discrepant positions in these various dimensions could have

implications in and of themselves for the individual was

Emile Benoit-Smullyan (1944). He maintained that the term

social status was employed too indiscriminately and de-

veloped the neologisms of situs and locus to aid in dealing

with the situation. He intended the term situs to be used

in relation to horizontal distinctions and groupings rather

than vertical or ranking distributions and groupings. A

social situs is an aggregate of persons who are distin-

guished by society on the basis of some common charact-

eristic either real or imaginary. Locus refers to the

"socially standardized function which an individual performs

in an organized group" (Benoit-Smullyan, 19114:154).

Benoit-Smullyan agreed with Weber that there are a rela-

tively few status hierarchies which are of primary impor-

tance. He believed these important status hierarchies to be










economic status, prestige status, and political status,

which are synonymous with Weber's class, status, and party.

He realized that a "high degree of concomitant variation

exists" (Benoit-Smullyan, 1944:156) among these three status

hierarchies. But he argued that the three hierarchies could

and should be differentiated and that many persons did not

rank evenly on these hierarchies. In fact, Benoit-Smullyan

pointed out the process of "status conversion," by which an

individual might employ his resources in one status hier-

archy in order to improve his position in another. Thus,

any individual might use his political power in order to

gain wealth or prestige; any one status position might be

utilized to improve the other two. He suggested that, "as

a result of status conversion processes which are normally

at work in every society, there exists a real tendency for

the different types of status to reach a common level"

(Benoit-Smullyan, 1944:160). This tendency for

fusing status was labeled as the "status equilibration"

process. He believed "when legal, customary, or other

barriers seriously hamper the equilibration tendency, social

tensions of revolutionary magnitude may be generated"

(Benoit-Smullyan, 1944:160).

Following these earlier leads, Stuart Adams (1953)

studied U.S. Air Force bomber crews in order to determine

the implications of status congruency within small groups.

It was found that increased status congruency at first had a










positive effect on technical performance, but changed to a

deterioration in technical performance with the passage of

time. Increased status congruency, however, did lead to

increased group cohesion, harmony, and social performance.

Zaleznik et al. (1958:59) later pointed out that Adams'

findings suggest that from conditions of high status

congruency the social leader of the group is likely to

emerge, while from conditions of low status congruency, the

task leader of the group is likely to emerge.

This introduction has attempted to accomplish two

tasks. First, to define the purpose of this work, which is

to examine the relationship between status inconsistency and

mental health. Specifically, the following types of

questions will be examined: Is there any association

between status inconsistency and mental health? If so, what

is the direction of the relationship and how strong is it?

Is there any evidence which would aid in assessing the

direction of the predominant process if a relationship were

found?

Second, the introduction provided a brief historical

review of the social thought concerning social stratifi-

cation which has been instrumental in the development of

theory and research related to status inconsistency. The

major components of this social thought up to the time

Gerhard Lenski expounded his ideas concerning status

crystallization are briefly reviewed.














CHAPTER 2
LITERATURE REVIEW



The literature concerning status inconsistency is vast

and varied. For this reason the literature will be reviewed

according to the dependent variables which various re-

searchers have attempted to relate to status inconsistency.

It should be noted that most of this literature has appeared

since 1954. In that year, Lenski published the article,

"Status Crystallization: A Non-Vertical Dimension of Social

Status," which provided the impetus for most of the sub-

sequent research which has been done relating to status

inconsistency. In fact, one of the basic criticisms one

often hears concerning status inconsistency is that we are

still too heavily dependent on Lenski's ideas and we have

not made significant progress beyond them. Therefore,

consideration will be given first to Lenski's original

articles, and second, the literature will be considered

according to the variables related to status inconsistency.

Lenski (1954), in his first test of status inconsis-

tency, examined inconsistency in the four status hierarchies

of income, occupation, education, and ethnicity. The status

inconsistency measure was obtained by finding the midpoint










of the cumulative frequency distribution for each status

hierarchy. Then, the square root of the sum of the squared

deviations from the mean of the four hierarchy scores of the

individual were subtracted from one hundred. Lenski found

that those characterized by low status crystallization were

more likely to vote Democratic. The low status crystal-

lization group was also more likely to express liberal

attitudes concerning government health insurance, price

controls, and extension of government powers. It was also

found that regardless of the particular form or type of low

crystallization, those having any type of low status

crystallization tended to have more liberal attitudes. He

did find, however, that certain types of status inconsis-

tency were more closely related to political liberalism than

other types. He concluded from his findings that "one might

Predict that the more frequently acute status inconsis-

tencies occur within a population the greater would be the

proportion of that population willing to support programs of

social change" (Lenski, 1954:412). It was Lenski's belief

that the status inconsistent individual represented a

particular type of "marginal man" who was subject to

pressures not felt by the status consistent individuals. He

went on to suggest three ways in which the individual might

react to his status inconsistency: he may blame others; he

may blame himself; or he may withdraw socially.










depending on which issues are under consideration. They

point out that political liberalism is not a unidimensional

concept. They also warn against the interpretation of votes

for the Democratic party as indicators of political

liberalism. This would seem to be especially relevant in

light of the traditional voting habits of many ethnic

groups. They argue that ethnic background and social class

membership are "far more important determinants of political

attitudes than the degree to which persons are status

consistent or inconsistent" (1966:27).

Similar results were obtained by Laumann and Segal

(1971) in comparing the effects of status inconsistency

versus ethnic group identification. They found greater

support for ethnic identity in explaining political and

social behavior than for status inconsistency.

In a study of data collected before and after the 1964

presidential election, Eitzen (1973) found that socio

economic status was a better predictor of consistent and

persistent political attitudes than was status inconsis-

tency. It should also be noted that neither dispropor-

tionately liberal nor disproportionately conservative

attitudes were found among the status inconsistent people.



Democratic Vote

Ascribed and achieved status discrepancy was found by

Segal and Knoke (1968) to be more highly associated with










Political Liberalism

Kenkel (1956) attempted to verify Lenski's findings by

retesting his hypothesis. Kenkel's research failed to

produce supporting results. Lenski (1956a) criticized

Kenkel's retest, maintaining that Kenkel had used status

dimensions such as prestige of neighborhood and area of

residence which did not occupy a "central place in American

society."

Lenski (1967) further tested his own theory by a

secondary analysis of data on twenty-five national surveys

of voting behavior in Canada, Australia, Britain, and the

United States. He found in twenty-one of the twenty-five

surveys that discrepancies between occupational rank and

socioreligious rank increased liberal tendencies. Britain

was the exception to this general pattern. Britain's

departure from this pattern was felt to be at least par-

tially attributable to the relatively homogeneous religious

structure of that country.

Schmitt (1965) used the statuses of education, eth-

nicity, husband's occupation and husband's income to measure

status inconsistency in a sample of 153 white married

females. He found the status incongruent white females to

be more politically liberal, and those having an educational

level below their husband's occupation or income to be the

most politically liberal.

Kelly and Chambliss (1966) found that social class

groups vary in their relationship to political liberalism










Democratic party vote than was discrepancy between achieved

statuses. Later, Segal (1969) also found that discrepancies

between socioeconomic status and ethnic status were associ-

ated with a Democratic party voting preference.

Using a sample collected by the University of Michigan

Survey Research Center for studying presidential elections

from 1952 to 1964, Smith (1969) obtained similar results.

He found that status inconsistency among achieved statuses

was more highly associated with Democratic party vote for

those under age forty-five than for those over that age.



Political Change

Laumann and Segal (1971) found no significant dif-

ference in preference for political change among those

consistent and inconsistent with respect to educational and

ethnoreligious status. Status inconsistency was also found

to be unrelated to attitudes concerning political change by

Olsen and Tully (1972).



Presidential Assassins

Wilkinson (1970) conceptualized status inconsistency as

the discrepancy between desired or expected status and

actual, achieved status. He found this type of status

inconsistency to characterize all the people who have

attempted or completed an assassination of an American

president.










Right-Wing Attitudes

Utilizing the status varibles of income, occupation,

and education, Rush (1967) in a study of attitudes on

social, political, and economic issues found greater support

for extreme right-wing attitudes among the status incon-

sistents than among the status consistent. But he did not

test for the distribution of scores on left-wing attitudes.

Eitzen (1970a) found Wallace supporters (who were used

as they indicated an adherence to the radical right) were

below the community median in education and occupation, but

above the median in income. This was interpreted as

evidence that those who experience structural imbalance

(status inconsistency) are more likely to support extreme

political movements.

In a secondary analysis of data collected during the

1964 presidential election, Hunt and Cushing (1970) found

that particular types of status inconsistency were related

to favorable or unfavorable attitudes towards the John Birch

Society. Specifically, they found that unfavorable atti-

tudes were highest among the education high-income low, and

the racial ethnic status low-achieved status high, types of

inconsistency. The high income-low occupation and low

income-high occupation inconsistency types displayed the

most favorable attitudes towards the John Birch Society.

Ethnicity and social class were found to be related to

political nationalism by Loh (1975), but status inconsis-

tency as measured by the interaction effects of ethnicity










and social class were not found to be related to political

nationalism.



Power Distribution

In a study of desire for change in power distribution,

Goffman (1957) was the first to suggest that status incon-

sistency be viewed from the perspective of discrepancies

between ascribed and achieved statuses. Others who made use

of this type of approach included Jackson (1962), Lenski

(1965), Segal and Knoke (1968), Olsen and Tully (1972), and

Jackson and Curtis (1972). The typical procedure employed

is to consider education, income, and occupation as achieved

statuses and age, sex, and race as ascribed statuses.

Goffman found the high achieved status and low ascribed

status inconsistent to have the greatest desire for a

rearrangement of power.



Prejudice

Income and education were the status dimensions

employed by Treiman (1966); he found that status incon-

sistency showed no significant association with the dis-

tribution of prejudical attitudes. Treiman was criticized

by Geschwender (1970), however, for utilizing only two

status dimensions in his analysis of status inconsistency.










Social Participation

Employing the status variables of education and

ethnoreligious status, Laumann and Segal (1971) tailed to

find any significant difference with respect to social

participation between status consistent and status incon-

sistent individuals.



Individual Unrest

Geschwender (1968a) found support for the hypothesis

that status inconsistent were more orone to exhibit

symptoms of individual unrest. Specifically, underrewarded

inconsistent (high ethnicity-low occupation, high educa-

tion-low occupation, high ethnicity-low income, high

education-low income) were more prone to individual unrest.

Overrewarded inconsistent did not show this pattern.



Job Satisfaction

It has been suggested by Kimberly and Crosbie (1967)

that the dissatisfaction associated with status inconsis-

tency is the result of discrepancies in cost and reward

rather than of status inconsistency per se. They con-

ceptualized status inconsistency in terms of position and

ability discrepancies: difficult position-low ability, or

easy position-high ability. They employed a laboratory

experiment in which subjects were led to believe they had a

certain level of ability according to their performance on










certain tests. Subjects were then free to choose tasks at

defined levels of difficulty and tested for their satis-

faction in how they performed these tasks. Kimberly and

Crosbie found support for their general thesis that status

inconsistency itself was not as highly associated with

satisfaction in performance as was the cost-reward ratio.

In a study of professional social workers, Kolack

(1968) found that compared to status consistent, status

inconsistent were less likely to obtain satisfaction from

their jobs, were less likely to view social work as their

terminal occupation, and had generally more unhappy ex-

periences in social work as a profession.

Research was undertaken by Kasl and Cobb (1971) to

examine several different types of status inconsistency.

Job satisfaction was found to be lowest for status incon-

gruent males of the types education lower than occupation

and education lower than perceived social class.

Erickson et al. (1972) found job satisfaction to be

significantly related to status inconsistency, which was

defined as job advancement being incommensurate with years

of experience, age, and marital status. Those who were in

step with their peers indicated greater job satisfaction,

while those who were out of step with their peers indicated

less job satisfaction.










Self-Esteem

Kasl and Cobb (1969a) found males who were charac-

terized by status inconsistency of education higher than

occupation or education higher than perceived social class

were particularly prone to low self-esteem. But, status

consistent males generally had the highest levels of self-

esteem. In addition, they found both males and females

reported lower self-esteem if they were from families

characterized by status inconsistency between the mother and

father. In addition, they found both families characterized

by status inconsistency between the mother and father.

This finding is similar in nature to Geschwender's (1968a)

finding of a relationship between underrewarded incon-

sistents and individual unrest.



Voluntary Organizations

Lenski (1956b) found that status inconsistent persons

participated less in voluntary relationships, have more

inactive, long-standing, voluntary ties, and are more likely

to establish and maintain voluntary ties for reasons other

than sociability.

Similarly, in his study of professional social workers,

Kolack (1958) found that status inconsistent social workers

participated less in voluntary associations and that when

they did participate they were far more likely to parti-

cipate for "nonsociable" reasons.










Decision Making

In their study of decision-making groups, Exline and

Ziller (1959) found that status inconsistency interfered

with the free flow of communication and the development of

satisfying interaction. They found that the group composed

of status inconsistent individuals encountered considerably

more difficulty in reaching consensus than the group

composed of status consistent individuals. This finding

supports Adams' (1953) earlier finding that status con-

gruency within a group increased group cohesion, harmony,

and social performance. It would also tend to make more

tenable the suggestion of Zaleznik et al. (1958) that social

leaders of the group are more likely to arise from conditions

of high status congruency.

In an experimental study of group behavior, Brandon

(1965) found status inconsistency to be a more accurate

predictor of group tension and feeling of unfairness when

consistency between the statuses was expected rather than

when it was not. Brandon pointed out that under certain

specified conditions, status inconsistency has a stronger

effect than an examination of the effects of status in-

consistency without specifying any conditions would

indicate.



Peptic Ulcers

No association was found between peptic ulcers and










status inconsistency by Kasl and Cobb (1971). An associa-

tion was found, however, between parental status inconsis-

tency and ulcers (Kasl and Cobb, 1967) and between husband

and wife status inconsistency and ulcers (Kasl and Cobb,

1969a).



Coronary Heart Disease

Shekelle et al. (1969) undertook a longitudinal study

of coronary heart disease at the Hawthorne Works of the

Western Electric Company in Chicago. It was found that

incidence of coronary heart disease increased as the number

of status inconsistencies increased. Subjects with four or

five different types of status inconsistency were found to

have six times the risk of coronary heart disease when

compared to subjects with no status inconsistency.

Horan and Gray (1974) reanalyzed the Western Electric

study data reported by Shekelle et al. (1969). They found

that the utilization of multivariate statistical techniques

allowing for multiple controls yielded very little asso-

cation between status inconsistency and coronary heart

disease.

In a reply to Horan and Gray, Shekelle (1976) admitted

that status inconsistency accounts for a relatively small

proportion of the variance in coronary heart disease. But,

Shekelle (1976:86) maintained that "this is a common

characteristic in research on the epidemiology of chronic










disease." He went on to say:

The importance of a variable is determined
not only by the magnitude of its coefficient
in a regression equation, but also by its role in
a theory of pathogenesis. (1976:86)

According to Shekelle, the notion that status inconsistency

is associated with coronary heart disease is strengthened by

the absence of a statistically significant relationship

between social status itself and coronary heart disease.

Shekelle (1976:87) maintained that:

At best all we can now say is that the weight
of evidence supports the hypothesis that men
with certain behavior and social characteris-
tics e.g., status inconsistency, the Type A
behavior pattern, social mobility, have a
higher risk of CHD than men who do not have
these characteristics.



Rheumatoid Arthritis

In the study of status inconsistency and rheumatoid

arthritis, Kasl and Cobb (1969a:276) list the following

among their findings directly related to rheumatoid

arthritis:

1. Women with rheumatoid arthritis are more
likely to come from families where parents'
marriage is status discrepant and the father
is congruent on education versus occupation.

2. Men with rheumatoid arthritis are somewhat
less likely to come from such status incon-
sistent families.

3. Several forms of status incongruence of the
respondents themselves bore no relationship
to rheumatoid arthritis in women and to
rheumatoid arthritis in men.










4. Marriages where both spouses were healthy
were less likely to be status discrepant
than those marriages where one or the other
of the spouses had rheumatoid arthritis.

5. Healthy husbands who were themselves status
incongruent were more likely to have wives
with rheumatoid arthritis than healthy
husbands who were status congruent.

Kasl and Cobb (1971) obtained the same results for rheuma-

toid arthritis as they obtained for peptic ulcers. There

was no association between the individual having rheumatoid

arthritis and being status inconsistent (see three above).

There was, however, an association between parental status

inconsistency and rheumatoid arthritis (Kasl and Cobb, 1967)

and between marital status inconsistency and rheumatoid

arthritis (Kasi and Cobb, 1969b).



Morbidity

In a sample consisting of 10,621 white employed males,

ages twenty-five through sixty-four, Wan (1973) found that

as the level of status inconsistency increased, the level of

morbidity also increased among the non-poor. This positive

relationship, however, did not hold for the poor or for the

sample as a whole.



Social Stress

Utilizing the three status variables of racial-ethnic

status, educational status, and occupational status, Jackson

(1962) attempted to analyze the effects of status










inconsistency on symptoms of stress as measured by a series

of items drawn from the Health Opinion Survey. The sample,

drawn from a national sample survey (Gurin et al. 1960),

consisted of 2,460 adult males and females. The males were

assigned their own status on each of the status variables

while the females were assigned their own status on race and

education, but they were given their husbands' occupational

status. Jackson (1962:473) found the consistent to have

the lowest percentage scoring high on symptoms of stress (16

percent). The moderately inconsistent had a slightly

higher percentage (18 percent) scoring high. The sharply

inconsistent with no like ranks had the next highest

percentage (24 percent) scoring high on symptoms, and the

sharply inconsistent with two rank deviates had the highest

percentage (35 percent) scoring high on symptoms of stress.

In addition, Jackson (1962:479) found the strongest

direct relationship between symptoms of stress and the

following status patterns:

1. Racial-ethnic rank superior to occupational
or educational rank.

2. For males, occupational rank superior to
educational rank.

3. For females, educational rank superior to
husband's occupational rank.

Jackson also compared his findings to those of Lenski (1954)

He found that the status inconsistency patterns associated

with high stress scores, the high ascribed-low achieved










patterns (R/O and R/E), were not associated with political

liberalism. The inconsistency patterns associated with

political liberalism, the low ascribed-high achieved

patterns (O/R and E/R), were not associated with stress.

From this Jackson concluded that although status incon-

sistency is likely to have effects upon the individual,

these effects may be manifest in different ways.

In'another study, Jackson and Burke (1965) drew on

Lenski's (1964) suggestion that statistical interaction

effects of the status variables could be thought of as

status inconsistency effects. Employing this method, they

undertook to examine the effects of status inconsistency on

stress through regression analysis. They made use of the

same status variables: education, occupation, and racial-

ethnic status. Several commentators (House and Harkins,

1975:399; Knoke, 1972:27) have pointed out the weakness of this

type of methodology. Jackson and Burke themselves recognized

that the use of interaction effects as a measure of status

inconsistency was questionable. They stated that, "If

significant deviations from the predicted rates occurred

among inconsistent groups, the presence of status incon-

sistency effects could be inferred" (1965:577). Thus, they

do not claim that interaction effects are status incon-

sistency effects. Rather, their claim is that the presence

of interaction effects provides a basis for inferring the










presence of status inconsistency effects. Jackson and Burke

admitted that "the test for interaction is not fully

appropriate for our needs because it collects all deviations

from the predicted scores into one interaction term"

(1965:558). Evidently they were aware that while the

interaction term may represent status inconsistency effects,

it may also represent other types of effects.

Jackson and Burke developed two models. In model one,

the three status variables provide a direct additive effect

and terms representing status inconsistency are simply added

to the model. This model leads to the conclusion that

education and occupation are negatively related to symptom

level, while racial-ethnic status is positively related to

symptom level. All forms of marked status inconsistency

were found to produce elevated symptom levels. In model

two, additive terms for occupational and educational status

were included along with interaction terms for occupation

and education. But for racial-ethnic status, only a term

representing high racial-ethnic status and low occupational

or educational status along with a term representing low

racial-ethnic status and high occupational or educational

status was included. This model also led to the conclusion

that educational and occupational statuses were negatively

related to symptom level and that marked education-occu-

pation discrepancies produce elevated symptom levels.

Unlike the first model, however, the second model indicates










that high racial-ethnic status and low occupational or

educational status have a much greater impact in increasing

symptom levels than does low racial-ethnic status and high

educational or occupational status.

Model one accounts for more of the variation in symptom

level than does model two but at the cost of including two

more variables. Jackson and Burke (1965:654) prefer model

two since they

find the assumption that inconsistency
between high ascribed and low achieved status
is especially likely to lead to a symptomi-
zation response, theoretically more palatable
.. than the assumption that high racial-
ethnic status is itself positively related
to symptom level.

Employing the twenty-two item Langner index as their

measure of stress, Meile and Haese (1969) examined the

relationship between stress and status inconsistency as

measured by educational and occupational discrepancies.

They found an inverse relationship between occupational and

educational levels and stress. They found no significant

relationship between stress and status inconsistency

regardless of whether the inconsistency was moderate or

marked. They did find a significant relationship between

the type of status inconsistency and stress. Specifically,

they found that having an educational status higher than

one's occupational status was associated with higher levels

of stress. This finding is consistent with those of

Geschwender (1968a) in connection with self-esteem. Thus an

underrewarded type of status inconsistency has been found to










be related to individual unrest, self-esteem, and stress.

Jackson and Curtis (1972) in a catch-all study of the

effects of status inconsistency examined the following

dependent variables which happenened to include stress:

formal social participation, informal social participation,

political liberalism, satisfaction and symptoms of stress,

intolerance, anomia, aspirations for son, leisure acti-

vities, self-perceptions, salience of rank, perceptions of

responsibility, and legitimacy. They employed a sample

consisting of male heads of household in six American

cities. The methodology they employed was to compare the

amount of variance in the dependent variables explained by a

simple additive model of status rank with the amount of

variance explained by a more complicated model, including

status inconsistency as measured by the interaction effects

of the status variables. Thus, the measurement of status

inconsistency in this study is subject to the same weakness

discussed earlier in relation to the measurement of status

inconsistency in the study by Jackson and Burke (1965).

When comparing the two models, Jackson and Curtis reasoned

that if the more complicated model including status in-

consistency does not explain a significantly greater amount

of variance, then one should choose the simpler additive

status ranks model on the grounds of parsimony. Their

conclusion was:










Most of the realtionships appeared to be
additive. The interactions which did appear
were not clustered with respect to any par-
ticular independent or dependent variables,
usually varied in form from city to city, and
did not resemble patterns expected on the basis
of mobility or inconsistency theory. The
findings suggest that multidimensional addi-
tive models adequately represent the effects
of social stratification on the individual
(Jackson and Curtis, 1972:701)



Mental Health

Tuckman and Kleiner (1962), using a sample of first

admissions to a state hospital, constructed a "Discrepancy

Index" as a predictor of schizophrenia. Utilizing education

as an indicator of aspiration and occupation as an indicator

of achievement, they found that the greater the discrepancy

between aspiration and achievement, the greater the risk of

schizophrenia. The findings of Tuckman and Kleiner, how-

ever, merit cautious interpretation. Since Dunham et al.

(1966:223) conclude "that while schizophrenics in their

developmental years are able to show fair educational

achievement they are at a distinct disadvantage when they

enter the job market."

Using a sample composed of 7,109 adults drawn from

western Jerusalem between 1962 and 1964, Abramson (1966)

attempted to examine the relationship between emotional

disorder and status inconsistency.

Status inconsistency was measured as a discrepancy

between occupational and educational ranking. Emotional










disorder was measured as a score of thirty points or higher

on the Cornell Medical Index. Abramson found that the

status inconsistent were considerably more likely to score

high, indicating greater emotional disorder.

Kasl and Cobb (1971) utilizing both a sample drawn from

a study of rheumatoid arthritis and a national survey

sample, examined the relationship between status incon-

sistency and mental health. They found that status incon-

gruent men reported more anger-irritation, more symptoms of

acute physical illness, more psychological anxiety, more

occasions of being fatigued, and more depression. The

occupation higher than education type of status incongruent

males consistently reported better mental health than those

with education higher than occupation. The differences

between these two types of status incongruent groups were

statistically significant for anger-irritation, immobili-

zation, depression, and selfconfidence. In terms of status

incongruency between perceived social class and education,

it was found that the incongruent group reported more

frequent ang more depression,and anger-irritation,ad a greater

re to change. This finding was particularly strong for

the type of inconsistency in which education was higher than

perceived social class in which case, mental health was

"invariably" reported as being poorer than for the education

lower than perceived social class. Kasl and Cobb (1971)

believed that status inconsistency of the type of education










higher than occupation was somewhat more likely to occur

among younger males, thus age would be an important variable

to control. Therefore, they conducted separate analyses for

males under forty-five and for males over forty-five. They

found that there was no significant difference between the

older and younger age groups; the level of association was

basically unaltered. Thus, Kasl and Cobb (1971:78)

concluded:

It can be seen again that the congruent group
has better mental health than either the in-
congruent groups, and that within the latter
two, the E>0 group has particularly poor mental
health... Incongruent effects of education vs.
perceived social class show a strong direction-
ality: E than men who are classified as E>uSC.

These findings are in line with those reported earlier

by Kasl and Cobb (1969a) concerning self-esteem; by Meile

and Haese (1969) concerning stress; and by Geschwender

(1968a) concerning individual unrest. An underrewarded type

of status inconsistency was found to be related to all these

factors.

In a study of mental patients in a state institution,

Eitzen and Bair (1972) found that if status inconsistent

patients were merely compared with status consistent

patients, both categories had almost the same probability of

being diagnosed as schizophrenics. The education higher

than occupation type of inconsistency had a disproportiona-

tely high probability of schizophrenia; furthermore, the

greater the magnitude of differential between higher










education and low occupation, the greater the probability of

a diagnosis of schizophrenia. But as Eitzen and Bair

(1972:71) point out the direction of causality is unclear.

In addition, these findings should be interpreted cautiously

since as Turner et al. (1969) have indicated, samples drawn

exclusively from state mental institutions tend to yield

biased distributions on many variables. According to Turner

et al. (1969:295), these biases "are likely to result in

findings or hypotheses that may be seriously misleading in

regard to the occurrence and etiology of schizophrenia."

Erickson et al. (1973) reported that status inconsis-

tency, defined as being out of line with one's Deers in

terms of job advancement and marital status, was signi-

ficantly related to poor mental health. They concluded

that, "The degree to which a man is in step with his peers

reflects his adjustment" (Erickson et al. 1973:400).

Baldwin et al. (1975) analyzed data on 11,325 psy-

chiatric out-patients to determine the relationship between

severity of psychiatric diagnosis and status inconsistency

employing the statuses of occupation, marital status, and

welfare status. They found support for the hypothesis that

patients having high status inconsistency would receive more

severe psychiatric diagnosis than patients having low status

inconsistency. However, other status variables, such as

race, sex, age, and occupation were found to be more

powerful predictors.










In a study of 333 employed white males, House and

Harkins (1975) found that for the older age group (age

greater than forty-five), the status inconsistency pattern

of high occupation, low education showed a significant

relationship to psychological strain.



Sighting of Flying Saucers

Using data from a 1966 Gallup Poll, Warren (1970)

analyzed the sighting of flying saucers in relation to

status inconsistency. He found that the status inconsis-

tents were the most likely to have reported seeing flying

saucers. Specifically, he found that the underrewarded

type, with either education or occupation high but income

low were by far the most likely to have reported sighting a

flying saucer. In answering the question: Just what do you

think these flying saucers are? It was found that "status

inconsistent persons who reported seeing UFO's are far more

likely than other groups to define them as extraterrestrial

vehicles," (Warren, 1970:603) rather than giving some more

"acceptable" explanation such as meteors, weather balloons,

or airplanes. Of the moderate inconsistent, 17.4 percent

reported them as extraterrestrial vehicles, while 28.6

percent of the sharply inconsistent gave that type of

response. "In fact, all such answers came from the status

inconsistent respondents, and not from the status consistent

saucer sighters" (Warren, 1970:603). Thus, Warren found










that, "it is not low income alone, but low income with

moderate to high education or occupational status that

produces a higher level of saucer reporting" (1970:602).

From this finding he concluded:

it is not, therefore, the uneducated, credu-
lous or the uninformed individual who reports
saucers. Rather, it is the individual whose
reward structure is out of line with his in-
vestment. (Warren, 1970:603)




Summary

In this chapter an attempt was made to review the

research which has been undertaken in relation to status

inconsistency. Literature was reviewed which reported

research on the relationship between status inconsistency

and twenty different variables. These variables are very

heterogeneous and represent diverse areas of human behavior.

The commonality they seem to share is that most of them

represent some type of deviant activity. The general

contention of most status inconsistency research is that

being status inconsistent is a stressful condition which

affects the individual in a manner resulting in deviant

behavior patterns.

Drawing on this line of reasoning, this research will

investigate the contention that the stress generally

associated with being status inconsistent has a negative

effect on mental health.














CHAPTER 3
THEORY



The theoretical support for this work is derived

principally from the perspective of distributive justice as

delineated by Homans (1961). In order to make this general

theoretical perspective more applicable to this particular

research endeavor, portions of other theories will be

included to enhance clarity and precision.

The preceding review of the literature enumerated

research on at least twenty different variables in relation

to status inconsistency. Almost all of these variables

represent some type of "deviant" or nonnormative behavior.

Thus, the general contention of status inconsistency

research is that there is something about being status

inconsistent which causes the person to engage in some type

of deviant behavior. As Zaleznik et al. have said, "from

persons of low status congruence we assume, expect, and

often find a certain kind of 'trouble'" (1958:57).

Thus, the task of the theory section of this research

is to explain why status inconsistency is distressing. The

particular task, then, becomes one of explaining why status

inconsistency would be expected to have a negative effect on

mental health.










Symbolic Interactionism and Role Theory

Probably the most common theoretical approach utilized

in status inconsistency research involves combining segments

from the perspectives of symbolic interactionism and role

theory. The supporters of symbolic interactionism maintain

that the individual develops a self-concept based largely on

the basis of how other people react to him. Cooley (1964)

in his development of the concept of "the looking glass

self" pointed out three basic steps in the formation of the

self-concept.

1. The imagination of our appearance to
the other person.

2. The imagination of his judgment of
that appearance.

3. Some sort of self-feelings, such as
pride or mortification, based on our
imagined appearance and his imagined
judgment.

In a similar.manner, Mead maintained that the individual's

development of his self-concept is dependent on the way in

which his "significant others" and later the "generalized

other" react to him (1972).

The supporters of role theory maintain that the

reactions of others is to some extent dependent upon the

statuses the individual occupies and the roles he plays in

connection with these statuses. Linton (1936) and Hughes

(1944) maintain that accompanying any given social status is

a set of normative expectations. This set of expectations

includes: expected rights and duties of persons occupying










the status as well as expected rights and duties of others

towards that status.

Therefore, combining these two perspectives in relation

to status inconsistency, it is concluded that if the

statuses are inconsistent, then people will tend to be

uncertain how to react to them. These uncertain or in-

consistent reactions will also tend to have a negative

effect on the individual's ability to maintain a stable

self-concept since his "significant others" or his "looking

glass self" reflects an inconsistent image or inconsistent

reactions. The general contention, then, is that this lack

of consistency results in a stressful condition with which

the individual attempts to deal by engaging in some type of

"deviant" or nonnormative behavior. For the purpose of this

research, this stressful condition will be seen as having a

negative effect on the person's mental health.

There are some problems with this type of theoretical

approach to status inconsistency research. First, it is

unclear which, if any, particular status the individual may

adopt as his rank from which he will relate to others. It

has been argued (Galtung, 1966) that the status inconsistent

individual will tend to view himself from his highest status

position and will expect others to view him from that same

perspective. While others (Himmelfarb and Senn, 1969; Segal

et al., 1970; and Laumann and Sega], 1971) argue that the

individual evaluates himself according to an average of his

statuses.










Second, and more important, this general theoretical

approach does not adequately distinguish between two

separate ways of viewing status inconsistency. The first

view of status inconsistency is the one taken by most

researchers, although it is usually not explicitly stated.

This perspective views status inconsistency as a character-

istic of the individual. Thus, according to the above

theory, it is a characteristic of the individual, his self-

concept, which is threatened, and a stressful condition

results. The second view considers status inconsistency as

a characteristic of a relationship. Mitchell (1964)

maintains that status inconsistency research should be

concerned with examining inconsistency within relationships,

rather than inconsistency within the individual. According

to Mitchell (1964:317) research designs for status incon-

sistency research should include status measures for both

the subject and his role partners. Malewski (1966) like

Mitchell, believed that status inconsistency should be

examined within the context of a relationship and is

characteristic of a relationship between two or more

individuals rather than a characteristic of an individual.

According to this view it is the inconsistent reactions of

others which result in an individual stressful condition.

Although it is probably the most widely used theory in

status inconsistency research, these weaknesses, along with

a more adequate theoretical formulation from other sources,










which are more explicitly concerned with the statuses

considered here, provide the reason for not utilizing this

perspective as the primary theoretical support of this work.



Structural Functionalism

Even though it is usually not explicitly stated, the

structural functional perspective on social stratification

provides the basis for believing that status inconsistency

might be stressful. In one of the classic works from the

structural functional perspective, Davis and Moore (1945)

argued that in order for society to insure that the most

qualified people filled the most important positions

(statuses) these positions must be highly rewarded. In

other words, it is functional for society if positions are

rewarded relative to the amount of effort (investment)

required to properly fill those positions. Thus, Davis and

Moore contend that unless positions requiring a great deal

of work and training (investments) were highly rewarded,

people would not put forth the effort to properly fill these

positions, and society would suffer.

It is this basic theoretical reasoning from structural

functionalism which leads to the expectation of status

consistency as the usual or "normal" condition, and status

inconsistency as the unusual or stressful condition. This

tends to be especially relevant for the statuses employed in

this analysis which utilizes occupational, educational, and










income status. There tends to be a high positive corre-

lation between these statuses. That one obtains a high

educational status which enables him to obtain a high

occupational status which combines to yield a high income

status is common folklore in American society. As Hartman

(1974) has pointed out, there are many dimensions in which

status inconsistency may be assessed. The implications of

status inconsistency are dependent upon the relative im-

portance of the status dimensions being compared. The

status variables employed here have been historically

correlated in American society. In addition, Americans

generally continue to expect and believe that education,

occupation, and income should be commensurate. This is

probably less true for many other variables which have been

employed in status inconsistency research. For example, the

expectation that education, occupation, and income should be

commensurate with race or sex is declining in this society.

But, most Americans would continue to support the view of

Hodge (1962) that, "...education may be seen as the in-

vestment whereby 'occupational stock' is acquired and upon

which income is the 'dividends.'" Thus, the theoretical

questions of concern are as follows: What are the impli-

cations for the individual when the functional theory of

stratification is not realized in his particular case?

What is the effect for the individual when he is not

rewarded relative to his investments? What is the effect










for the individual when American folklore is found to be

invalid? Does this inconsistency of investments and rewards

have a negative effect on his mental health?

In order to more adequately investigate this issue, the

suggestion of Lenski (1956b), Blalock (1966), Geschwender

(1968a), and Eitzen (1970b) to concentrate research on

different types of status inconsistency rather than on the

magnitude of status inconsistency has been adopted.



Distributive Justice

A theoretical perspective which fits the general

approach outlined under structural functionalism is dis-

tributive justice as proposed by George Homans. According

to Homans, "The rule of justice says that a man's rewards

in exchange with others should be proportional to his

investments" (1961:235). This general rule of distributive

justice and its application to status inconsistency research

is made clear by Zaleznik et al. in their discussion of

Homan's work.

In listening to the complaints of workers,
one hears frequently such statements as
"service counts too much or too little
around here," "education counts too much
or too little," "ability is not rewarded,"
"woman's place is in the home," "the
youngsters are still wet behind the ears,"
"those foreigners don't know how good they
got it here," "married employees are given
preference over single employees,"
"brains don't count." In asking employees
to elaborate upon these complaints they
frequently come down to some explicit or
implicit assumption of this sort: "My pay










is 'not in line' with my age, skill, senio-
rity, education, ethnicity, sex, responsi-
bilities, marital status, etc., and this is
not just..." According to Homans when the
investments of an individual member or sub-
group are higher than those of another,
distributive justice requires that their
rewards should be higher too. This is the
"emaning" of "status congruence." It is a
condition of equilibrium because it is a
condition of "felt justice." Thus, com-
plaints do not arise when status congruence
exists in a person, group, or job becuase
this condition is felt as "just" by all the
members of the group. (1958:50-53)

Homans is careful to point out the relative nature of

this balancing process between investments and rewards. The

process by which men gauge this discrepancy between invest-

ments and rewards is a subjective one.

We must not argue for a minute that people
can meausre the rewards and costs of what
they do in cardinal figures, but they
surely do assess them in ordinal ones;
they put the activities of different
persons and subgroups in rank-order of re-
wards and costs. In these terms distribu-
tive justice among men is achieved when
the profits of each are equal, in the
sense that their rank-order on rewards is
also their rank-order on costs. (Homans,
1961:202)

As previously noted, this particular research endeavor

examines status inconsistency as a discrepancy between the

levels of education, occupation, and income. A person's

effort to obtain an education, and the work he does in his

occupation are considered as his investments. The income he

receives as a result of having made these two investments is

considered as his reward. Thus, a person who ranks

similarly in terms of investments and rewards would be










considered to be status consistent or in Homans' terms, as

having experienced distributive justice. But, a person who

,iiinks dissimilarly in terms of investments and rewards would

be considered to be status inconsistent as having experienced

distributive injustice.

This study will examine three main types of status

inconsistency as it relates to mental health. These are:

1. The Underrewarded Type: which consists of
those who are underrewarded relative to
their investment.

2. The Overrewarded Type: Which consists of
those who are overrewarded relative to
their investment.

3. The Consistent Type: which consists of
those who are consistently rewarded relative
to their investment.

According to Homans' terminology, the consistent type

has experienced distributive justice, while the under-

rewarded type and the overrewarded type have experienced

distributive injustice. Although both the underrewarded and

overrewarded types of inconsistency may lead to feelings of

distributive injustice, they would not be expected to have

the same results. As Homans pointed out:

'The more to a man's disadvantage the rule
of distributive justice fails of realiza-
tion, the more likely he is to display the
emotional behavior we call anger.' And
finally we argue that men are rewarded
by the attainment of justice, especially
when just conditions are rewarding in other
ways. For instance, I am more likely to
demand justice when justice would bring
me more money than when it would bring me
les. (Homans, 1961:232)










Thus, the theoretical perspective of distributive

justice would predict that the underrewarded type of status

inconsistency would be felt as the greater injustice. This

would be experienced as the most stressful of the three and

have the greatest negative effect on mental health.

The overrewarded type of status inconsistency would also

be felt to be unjust. But, the level of stress experienced

is likely to be less, since, as Homans pointed out, the

injustice favors the individual instead of penalizing

him. Therefore, while still having a negative effect on

mental health, the effect should be less than for the

underrewarded type. The consistently rewarded type would

be predicted to have experienced distributive justice

which would produce little or no stress. They would be

expected to have little or no negative effect on their

mental health resulting from status inconsistency.

As the theoretical perspective is developed here,

status inconsistency is considered to be a characteristic of

an individual rather than a characteristic of a relations-

hip. The frustration or stress experienced by being

underrewarded is not peculiar to any specific relationship.

The stress-inducing agent is not considered to be the

individual's interaction with others who react inconsis-

tently to his inconsistent statuses and thereby threaten his

self-concept which is the contention of role theory and

symbolic interactionism. Rather, the stress inducing agent










to be the frustration and injustice (stress)

individual within himself, from the experience

2rrewarded relative to his investments.

justice deals more explicitly with this type of

nd reward discrepancies than either role theory

interactionism. It also avoids the weakness

rlier in relation to role theory



ism in that they did not distinguish whether

nsistency was considered to be a characteristic

idual or a characteristic of a relationship. As

964) has noted, a concern with either incon-

thin relationships or inconsistency within the

may be a legitimate research concern. The

should be careful, however, to distinguish

two and to indicate with which he will be





Expectancy Congruence

son (1963) in his article concerning expectancy

e maintains that it is not status inconsistency in

which is stressful. Malewski agreed with Sampson.

In order to establish the incongruence of
status it is not sufficient to point out,
as is commonly done, that some factors of
an individual rank much higher than others.
One should also show reasons for maintaining
that those differences are inconsistent with
the normative expectations of the environ-
ment in which the given individual moves.
(1966:304)










In an explicit test of expectancy congruence as a

advanced by Sampson, Brandon (1965) found inconsistency

between statuses of expected congruence to be a better

predictor of feelings of unfairness and group tensions than

is inconsistency between statuses that were not necessarily

expected to be congruent.

The statuses employed in this research to measure

status inconsistency meet the requirements of expectancy

congruence as developed by Sampson. The common folklore

assumption in this society is that increments in education

are believed to make one more marketable for higher status

occupations, which in turn combine to yield elevated levels

of income, has been previously discussed. In addition, many

commentators have noted the high correlation among educa-

tion, occupation, and income (Broom and Jones, 1970:999;

Hartman, 1975:708). The general expectation in this

society, both objectively and subjectively, is that these

statuses will be congruent. Therefore, conforming to the

expectancy congruence discussed by Sampson provides further

evidence for suspecting that status inconsistency, as viewed

from the distributive justice perspective, results in a

stressful condition which would have a negative effect on

mental health.



Relative Deprivation

The relationship between status inconsistency (con-

ceived of as discrepancies in investments and reward) and










distributive justice (as delineated by George Homans) has

already been noted. Homans went on to point out, however,

that "it is always relative deprivation that arises the

question of distributive justice" (Homans, 1961:2L3).

Thus, he is saying that it is not only the actual extent or

degree of deprivation or inconsistency which results in

feelings of injustice, but that these feelings of injustice

are also relative to other factors. What are these other

factors? There are probably several, but only two which are

felt to be very important will be discussed.

In the first as has been previously noted, Sampson

(1963) maintained the degree to which deprivation in the

form of inconsistency was experienced would be relative to

the extent to which the statuses involved were expected to

be consistent in the society at large. Second, the amount

of deprivation in the form of inconsistency which was

experienced would be relative to whether or not the in-

dividual himself perceived his investments and rewards to be

discrepant regardless of the "objective" discrepancy between

them. This point will be developed in more detail.

The reasons distributive justice would predict under-

rewarded inconsistent to be under more stress than over-

rewarded inconsistent have been previously noted, in a

similar manner, from the perspective of relative deprivation

one would expect the underrewarded to feel relatively

deprived and the overrewarded to feel relatively undeprived.










As Runciman and Bagley have commented: "We should expect

those of high education and low income to feel in general,

a sense of relative deprivation, but those of low education

and high income to feel relatively gratified" (1969:366).



The Predominant Process in the Relationship
Between Status Inconsistency and Poor Mental Health

The preceding theoretical discussion has approached the

relationship between status inconsistency and mental health

from the perspective that status inconsistency is stressful

for the individual and, therefore, has a negative effect on

his mental health. Some (Dunham et al., 1966) have argued

that the predominant process in the relationship mav go the

other way. It may be that having poor mental health results

in the person being status inconsistent. This is based on

the reasoning that regardless of what a person's educational

attainment might be, if he has poor mental health he is more

likely to have difficulty in obtaining and holding a job,

especially a higher status job. Therefore, his occupational

status is more likely to be suppressed and to fall below his

educational status. His income level would be expected to

accompany his occupational status. In an attempt to bring

some clarity to this issue, two modified approaches to the

relationship between status inconsistency and mental health

have been utilized.










The First Test for the Predominant Process in the Relation-
ship Between Status Inconsistency and Poor Mental Health

For the first modified approach to test this issue let

us momentarily assume that those who argue that poor mental

health results in status inconsistency are correct. If this

is true, the next question is what particular types of

status inconsistency are most likely to result from poor

mental health? Dunham et al. (1966) maintain that indi-

viduals with poor mental health may function fairly ade-

quately in the somewhat protective environment of the

educational system. In fact, unless the mental incapa-

citation is severe they will be required by law to attain a

number of years of education. Additionally, some may desire

to remain in this protective environment and thereby attain

education beyond the high school level. But when called

upon to participate in the "real world" of the job market,

they have greater difficulty in sustaining an acceptable

performance. Therefore, even though they may have a

moderate to high educational level, their occupational level

tends to be somewhat suppressed. Their income level would

probably tend to accompany their occupational level.

A measure of this type of status inconsistency was

developed and labeled as the "impaired" type. The central

characteristic of the impaired type is that occupational

status is always lower than educational status. Following

the reasoning stated above, if the primary relationship

between status inconsistency and mental health was such










that poor mental health resulted in status inconsistency,

then one would expect to find a strong relationship between

the "impaired" type of inconsistency and poor mental health.

Comparing the relationship between the impaired type of

status inconsistency and mental health to the relationship

between the underrewarded type of status inconsistency and

mental health could give some indication as to which is the

predominant process in the relationship between status

inconsistency and poor mental health.

In this comparison the test would be to determine which

type of status inconsistency was more highly associated with

indicators of poor mental health. If the impaired type

showed the stronger association, it would favor the argument

hat poor mental health results in status inconsistency. If

the underrewarded type showed the stronger association, it

would favor the argument that status inconsistency is a

stressful condition which may have a negative effect on

mental health.



The Second Test for the Predominant Process in the Relation-
ship Between Status Inconsistency and Poor Mental Health

The second modified approach involved the introduction

of an intervening variable which would be expected to alter

the relationship between status inconsistency and mental

health. House and Harkins (1975) have pointed out most

studies of status inconsistency have attempted to assess a

direct relationship between status inconsistency and some










dependent variable. There has been a tendency to ignore

intervening variables which might alter the relationship.

They contend that the explanatory value of status incon-

sistency might be significantly improved if the specific

conditions under which "particular discrepancies involving

particular status dimensions" were specified (1975:407).

The intervening variable employed in this analysis was

perception, or more specifically, perception of being

underrewarded. Regardless of what the objective measures of

the individual's investments and rewards might indicate,

Does the individual perceive himself as underrewarded? Or,

from the perspective of distributive justice, Does the

individual think he is fairly or justly rewarded?

The theoretical importance of perception in relation to

structural variables is well known. Probably one of the

best-known statements in the sociological literature is

W.I. Thomas' statement that, "If men define situations as

real they are real in their consequences" (Leslie et al.

1973:188). In other words if a person defines a situation

as being real receivess it as being real), then that

situation will have consequences or effects for him. W.I.

Thomas explained it as follows:

We must put ourselves in the position of
the subject who tries to find his way in
this world, and we must remember, first
of all, that the environment by which he
is influenced and to which he adapts him-
self, is his world, not the objective
world of science--is nature and society
as he sees them, not as the scientist sees
them. The individual subject reacts only










to his experience, and his experience is
not everything that an absolutely objective
observer might find in the portion of the
world within the individual's reach, but
only what the individual himself finds.
(Janowitz, 1966:23)

In attempting to relate a stressful situation such as

status inconsistency to mental health, it becomes important

to consider whether or not the individual perceives the

situation. In discussing the relationship between stresses

and disease, Wolff noted the importance of the individual's

perception of the stressful situation.

The stress accruing from a situation is
based in large part on the way the
affected subject perceives it....it is
not the particular nature of the forces,
pressures and preferences that engender
a threat to the individual in any parti-
cular society, but how they are perceived
and the amount of conflict directly or
indirectly engendered. It is not the
specific behavior toward parents, power,
possession, sexuality, the hours of work,
or even the type of work or the amount of
individual freedom of action, but it is
the threat engendered by the culture which
becomes pertinent to the development of
stress with its ensuing protective reac-
tion patterns and disease. (Wolff, 1953:10,
14-15)

Thus, Thomas has noted that in considering the effects

of a situation on the individual one must take into account

how the individual perceives the situation. In a similar

manner Wolff noted that one must take into account how the

individual perceives the stress in considering the effects

of stress on his health.

Introducing the perception of underreward also










provides assistance in determining the predominant process

in the relationship between status inconsistency and poor

mental health. If one claims that it is an underrewarded

type of inconsistency which has the greatest negative effect

on mental health, as is done here from the perspective of

distributive justice then one would expect perceived

underreward to have a greater negative effect on mental

health than nonperceived underreward. Since, according to

Wolff, the perception of stress is an important factor in

determining its effect. If one argues that poor mental

health results in status inconsistency then the perception

of underreward would be expected to have little impact on

the relationship between status inconsistency and mental

health. If a variable does not have an effect, then the

perception of that variable should also not have an effect.

As both Thomas and Wolff have indicated, it is how one

perceives the supposed causal factor which is important.

Thus, the test would be to introduce perception of

underreward into the relationship between status incon-

sistency and mental health. The first argument would be

favored if perception of underreward resulted in an

increased negative effect on mental health. The second

argument would be favored if the perception of underre-

ward had little effect on the relationship between status

inconsistency and mental health.










Social Stress

Central to the sociological approach to illness is the

notion that social factors may act as etiological agents in

the disease process. Chief among these social factors which

have received attention in this respect has been social

stress. In relating the social model of illness to the

medical model, Freeman (1960) pointed out what he refers to

as the "man-environment relationship" which he said occurred

at three levels and affects health and illness. The three

levels discussed are:

1. The Physical: in which man contends
with the "elements," light, dark,
hot, cold, humidity, etc.

2. The Biological: the major con-
sideration here being microorganisms.
This has been the area which has re-
ceived the major portion of medical
attention.

3. The Social: as man interacts with the
elements, with plants, animals, and
microorganisms, so man also interacts
with his fellow man. The effects of
this area of interaction on health are
just beginning to be appreciated
(Freeman, 1969:8485).

Freeman argued that in all three of these "man-en-

vironment relationships" there is the presence of stress and

strain which can have effects on the health or illness of

the individual. Thus according to Freeman, there may be

physical stressors, biological stressors, and social

stressors.

In examining the relationship between social stress and










disease, Wolff discussed stress in terms of mechanics.

...'stress' is the internal or resisting
force brought into action in part by ex-
ternal forces or loads. The change in
size or shape of the member as a result
of the application (2) (Maurer, 1917).
The load in biology becomes the stimulus
or the external environment agent. Loads
may be considered as a) sustained and of
low or moderate intensity;*
b) repeated and of low or moderated
intensity; and c) brief but of high
intensity.

The stress becomes the interaction
between external environment and orga-
nism, with the past experience of the
organism as a major factor. The strain
is the alteration or deformation in the
organism that then ensues. The magni-
tude of the latter and the capacity of
the organism to withstand the strain
determine whether or not there will be
re-establishment of homeostasis or a
'break,' with disruption and death.

...the impact of man on man may
be as seriously traumatic as the
assaults of microorganisms, climate,
chemical and physical forces. Disrup-
tions, hindrances and threats, stemming
from the interaction of man and man,
both singly and in groups, evoke
adaptive responses indistinguishable
from those set off by other environ-
ment forces. (Wolff, 1953:5-6)

Thus, Wolff appears to be in basic agreement with



*It should be noted that the stress associated with the
under-rewarded type of status inconsistency is probably the
type which Wolff said "...may be considered as a) sustained
and of low or moderate intensity..." (Wolff, 1953). This
is based on the reasoning that inconsistency between the
status of education, occupation, and income are likely to
persist over a number of years and thus, is of a sustained
nature. Also, for most people, stress of this type is not
likely to be of a high intensity. For most, it will pro-
bably be experienced as a more or less constant nagging
feeling of frustration, unfairness or injustice.










Freeman. Both would maintain that just as physical and

biological factors may act as agents of stress, so may

social and cultural factors. They also agree that stress

is a force acting on the organism and that strain is the

result of that force, the change or tissue damage.

Wolff illustrated his basic argument with the following

example. For both arms, the tone of the small blood vessels

of a subject's skin was tested for their capacity to hold

the contents of the blood. The left arm was then struck.

This resulted in the appearance of a red area which is asso-

ciated with a fall in capillary tone. The right arm was not

struck but a similar capillary change occurred. The experi-

ment was repeated only this time a mock blow was delivered

which stopped short of actually touching the subject's arm.

The anticipated threat, however resulted in capillary changes

which were similar to those which occurred when the subject

was actually struck. The experiment was repeated only this

time the subject was informed that the blow to be delivered

was a mock blow which would stop short of actually touching

him. This resulted in no observable change in capillary

tone (Wolff, 1953:5). In discussing Wolff's theoretical

perspective as illustrated in this example, Mechanic has

noted that:

Wolff argues, as his basic theoretical
assumption that the bodily protective
responses pattern to a physical blow
has been generalized to a symbolic blow.
It is his contention that when such re-
sponse patterns occur with great frequency
in response to threats to status and other










forms of stress, they result in symptom
formation and tissue damage. (1968:315)

Both Mechanic and Wolff have noted that these symbolic blows

or threats may take many different forms.

Mechanic has conceptualized these symbolic blows or

threats in terms of demands made on the individual.

It appears that it is useful to conceive
of stress as characterizing a discrepancy
between the demands impinging on a person
--whether these demands be external or in-
ternal, whether challenges or goals--and
the individual's potential responses to
these demands. (Mechanic, 1968:301)

This type of approach draws heavily on the work of W.I. Thomas,

who emphasized that the crisis lies not in the situation it-

self but in the interaction between the situation, the indi-

vidual's perception of the situation, and the individual's

resources for dealing with the situation (Mechanic, 1968:302).

Thus, in discussing social stress, consideration should be

given to normative expectations, excessive demands, percep-

tion of demands, inadequate resources, legitimacy of demands,

and the absence of alternatives.

This multiplicity of interacting factors partially ex-

plains why it has so often been noted that what is stressful

for one individual may not be stressful for another. As

Simmons and Wolff have observed:

The stress accruing from a situation is based
in large part on the way the affected subject
perceives it: perception depends upon a mul-
tiplicity of factors including the genetic
equipment, basic individual needs and longings,
earlier conditioning influences, and a host of
life experiences and cultural pressures. (1954:25)










This raises an important question. If the conditions which

result in stress are not constant, but may vary from

individual to individual how can researchers go about study-

ing social and cultural factors which may act as stress-in-

ducing agents in the disease process? In responding to this

issue Basowitz et al. (1955) have noted that certain stimuli

may be considered as stress inducing agents regardless of the

responses they may evoke in any particular individual. They

are considered as stressful "because of their assumed or po-

tential effects, although we well know that in any given case

the organism's adaptive capacity, threshold, or previous lear-

ning may preclude any disturbance of behavior" (Basowitz et al.

1955:7) In a similar manner, Mechanic has maintained that,

...the designation of certain circum-
stances as stress situations if based
on an assumption: the investigator
intuitively selects various aspects
of the physical, social, and cultural
environments that he assumes are likely
to lead to experiences of discomfort
for most people living within some de-
signated group, the discomfort being
reflected by both social and psycholo-
gical responses. (1968:297)










injustice which may be a stress inducing agent which would

have a negative effect on mental health.



Theoretical Propositions

The research reported here employed structural func-

tionalism as the broad theoretical perspective. Structural

functionalism places emphasis on the structure of social

organization and how the structure functions for the

effective operation of the social system. Davis and Moore

(1945) have amply presented the structural functional per-

spective on social stratification. From this view it is

functional for the effective operation of a social system

for the most difficult positions (statuses) requiring the

greatest effort and training (investments) to be highly

rewarded in order to insure that the most qualified people

occupy those positions. This theoretical perspective

provides the basic theoretical framework for status in-

consistency research. The notion that education, occupation

and income should be consistent is a derivative of struc-

tural functional theory, even though in some cases it may

not be explicitly recognized.

Within the broad structural functional perspective the

particular approach utilized in this research is distributive

justice as delineated by Homans (1961). The reason for

adopting this particular approach was because of its primary

emphasis on proportionality. The central core of the

distributive justice perspective is that rewards should be










proportional to investments, and if they are not it is

likely to be stressful for the individual. Different types

of investment-reward discrepancies, however, are likely to

have different effects. Those overrewarded relative to

their investments will perhaps feel some guilt, but they are

also likely to feel as though they have beaten the "system,"

they have succeeded. Those underrewarded relative to

their investments are more likely to feel as though the

"system" has beaten them. They are more likely to experi-

ence stress in the form of frustration, anger and resentment.

The distributive justice perspective was supplemented

by the perspectives of expectancy congruence and relative

deprivation. The perspective of expectancy congruence

points out it is not meaningful to talk about status

inconsistency in relation to any collection of statuses.

Rather, status inconsistency becomes meaningful only in

relation to a set of statuses which according to the general

beliefs of the society are supposed to be consistent. It

was noted that the three statuses employed in this analysis

(education, occupation and income), were both objectively

and subjectively expected to be consistent in this society.

The perspective of relative deprivation was introduced to

indicate that there are factors other than the absolute

level of deprivation or inconsistency which affect the

relationship between status inconsistency and mental health.










There were two such factors which were given consideration.

First, the effects of status inconsistency probably vary

relative to the extent to which the statuses involved were

expected to be consistent; this was dealt with by selecting

statuses which were generally expected to be consistent.

Second, teh effects of status inconsistency probably vary

relative to the degree to which it is perceived by the

individual.

Drawing on the broad theoretical perspective of

structural functionalism as developed through distributive

justice and supplemented by expectancy congruence and relative

dprivation the following working propositions were advanced.

1. The underrewarded type of status inconsistency
is associated with poor mental health.

2. The overrewarded type of status inconsistency
is not associated with poor mental health.

3. The relationship between status inconsistency
and mental health identified in the first two
propositions, will continue to exist after con-
trolling for the effects of race, sex, SES, age,
and marital status on mental health.

4. The predominant process in the relationship be-
tween status inconsistency and poor mental
health is that the underrewarded type of status
inconsistency has a negative influence on mental
health, rather than poor mental health influ-
encing the development of status inconsistency.

These working propositions provide the context for the

organization and presentation of the data. The methodology

employed in the analysis of these propositions is presented

in the next chapter.














CHAPTER 4
METHODOLOGY

The preceding chapter presented the theoretical

reasoning for expecting status inconsistency to be a

stressful condition which could have a negative effect on

mental health. The purpose of this chapter is to explain

the methodological procedures by which the propositions

developed in the previous chapter were tested. Before

this specific task is undertaken, however, attention

must be given to some of the general methodological

criticisms of research in status inconsistency and how

this particular study has dealth with those criticisms.



Methodological Criticisms of
Status Inconsistency Research

In his original research on status inconsistency

and political liberalism, Lenski (1954) utilized the

following procedure to measure status inconsistency.

First, for each of the four social statuses (education,

occupation, income and ethnicity) considered in the

research, a cumulative frequency distribution was calcu-

lated. Then Lenski proceeded by "taking the square root

of the sum of the squared deviations from the mean of

the four hierarchy scores of the individual and subtracting

the resulting figure from one hundred" (1954:407).










This procedure yielded possible scores ranging from

zerc to one hundred with high scores indicating high

status consistency and low scores indicating low status

consistency.

One of the major shortcomings of this type of measure

is that it does not allow for direct examination of

different types of status inconsistency. Since the major

concern of this research is to compare the mental health

of people characterized by various types of status incon-

sistency, a measure which does not facilitate such a

comparison would not be useful.

It should be pointed out that in his original

paper Lenski (1954) did examine various types of status

inconsistency, but not by using the measure of status

inconsistency explained above. In examining the

different types of status inconsistency Lenski (1954:

410-11) compared the four social statuses involved in

the inconsistency measure two at a time. He considered

a difference of 30 or more points in the cumulative

frequency distribution of any two status hierarchies

to represent a status inconsistency between those two

statuses. The status which is at least 30 points higher

in the cumulative frequency distribution is, of course,

the high status, and that which is at least 30 points

lower is the low status.










Berry and Martin (1973) have criticized this type of

approach for comparing different types of status incon-

sistency. The major problem to which Berry and Martin

devoted the largest portion of their effort was what they

referred to as the topologicall fallacy." This methodolo-

gical problem has two primary ingredients, the problem

of the inflated sample size and the problem of meaningless

comparisons. The problem of the inflated sample size

arises when a complete status profile analysis is not

employed and a difference in any two status hierarchies

is considered to constitute a case of status inconsistency.

In a complete status profile analysis, all the social

statuses considered in the status inconsistency measures

are compared at the same time, rather than two at a time.

This procedure avoids emphasizing the extreme (highest

and lowest) social statuses. Thus, in a complete

status profile analysis, all the status inconsistencies

involved are considered in a synchronous manner, allowing

the effects of all the social statuses to be assessed

simultaneously.

The problem of the inflated sample size was pointed

out by Berry and Martin in Lenski's (1954) original paper

in which he had 166 status inconsistent individuals in

the sample, but made comparisons on the basis of 598

status inconsistencies. Obviously, some of the individuals










in the sample were utilized in more than one of the com-

par isons.

This leads directly to the second issue, the problem

of meaningless comparisons. Since many of the inconsisten-

cies compared are characteristics of the same individual,

"the inconsistencies cannot be considered to be independent

events" (Berry and Martin, 1973:27). Because the status

inconsistencies measured according to the Lenski method are

not independent and not mutually exclusive, statistical ana-

lysis would be misleading (Morrison, 1976:2). Therefore,

this procedure was not used in this research.

Lenski (1964) later proposed another method of

measuring status inconsistency. He utilized a two-by-two

table, breaking statuses into categories of high and

low. According to the Lenski method, one would compare the

sum of the low-high and the high-low diagonal with the

sum of the high-high and the low-low diagonal. If the

two diagonal sums were not equal, then this interaction

effect would be attributed to status inconsistency.

Hyman (1966) criticized Lenski's (1964) methodology

for measuring status inconsistency as too conservative,

arguing that Lenski's method may be too "broad" and,

further, that it fails to specify the particular type of

status inconsistency which produces stress. It merely

indicates that at least one type of status inconsistency

has no effect. Secondly, as Hyman pointed out, if different










types of status inconsistency are associated with the

dependent variable in opposite directions they may

cancel each other when they are summed. A comparison of

the summed diagonals would show them to be equal, and

by the Lenski method one would conclude that no status

inconsistency effects were present. A closer examination

of the data, however, could reveal that those in the

high-low cell were significantly different from those in

the low-high cell with respect to the dependent variable.

Thus, Hyman criticized Lenski's method for either

(a) failing to detect the particular type of inconsistency

which produces an effect or (b) failing to detect any

inconsistency effect when one is actually present.

Therefore, according to Hyman (1966:128), "If Lenski's

test indicated that a status inconsistency effect is

present in a table, then some such effect is indeed

there." However, "If the Lenski approach indicates the

absence of a status inconsistency effect, one may never-

the less be present." Because of these weaknesses, this

method of measuring status inconsistency was deemed

unacceptable for this research.

Berry and Martin (1972) have indicated some of the

problems associated with controlling for the effects of

socioeconomic status (SES) in status inconsistency

research. A typical procedure utilized in status










inconsistency research in controlling for the effects

of SES, has been to show that the status inconsistent

did not differ significantly from the status consistent

with respect to SES. But as Berry and Martin (1972:86)

pointed out, "the fact that two or more groups have the

same mean scores on socioeconomic status says nothing

about whether the effects of the status dimensions have

been controlled at the individual level."

This research avoided this issue by controlling

for SES in a different manner. Scores on SES were entered

as a continuous independent variable along with status

inconsistency as an independent variable in a regression

equation to predict mental health scores. It should also

be noted, however, that in this analysis the various

types of status inconsistency were found to have signi-

ficantly different scores on SES, as indicated in Table

8. The association between low SES and poor mental health

has been well documented (Dohrenwend and Dohrenwend,

1969). As explained in the discussion of Table 8, the

fact that the status inconsistent groups scored differently

on the mental health measure than one would predict from

a knowledge of their SES scores indicates that status

inconsistency had an effect independent of SES.

Berry and Martin (1972) commented further concerning

the relationship between SES and status inconsistency.











It was their contention that this relationship was of

such a nature that:

(1) any given socioeconomic score limits the
range of consistency scores possible and (2)
the two vary in a curvilinear fashion so that
individuals with either high or low socio-
economic scores can have only high consis-
tency scores and it is only individuals with
middle socioeconomic scores who can possibly
have low consistency scores. (Berry and
Martin, 1972:87)

Their first point, while possibly valid, does not consti-

tute a real criticism of a status inconsistency or SES

measure. The fact that two or more measures have the

same possible range indicates nothing about the statis-

tical independence or dependence between the measures.

In fact, to some extent, having the same possible range

may be advantageous in terms of facilitating the con-

ceptual ease of comparability.

While their second criticism does apply to many

measures of SES and status inconsistency, it does not

as readily apply to the SES and status inconsistency

measures employed in this analysis. This is elaborated

further in the discussion of Table 8, which demonstrates

that, as measured in this analysis, all five SES groups,

ranking from low to high SES, were represented in each

of the status inconsistency types.

Another problem in status inconsistency research

which has received attention from several sources










(Blalock, 1966; 1967a, 1967b; 1967c; 1969; Duncan, 1975a;

1975b) is the "identification problem." In status incon-

sistency research, the identification problem arises when

one attempts to develop a structural equation model. The

identification problem is a result of having too many

unknowns for a unique solution to the simultaneous

equations. As Blalock (1967b:790) noted, the identifi-

cation problem develops when one or more or the

independent variables is a direct function of some of the

other independent variables. As a result, a unique

solution cannot be identified. How identifiable solutions

can be obtained was given as follows by Blalock:

The necessary conditions for identification
can be stated rather simply. We write a
(linear) equation for each variable that is
dependent on any of the other variables.
In order for the coefficients in a given
equation to be identifiable, the number of
variables excluded from this equation must
be at least equal to one less than the
number of equations. This criterion can
be expressed in another equivalent way.
Referring to variables that are not depen-
dent on any of the others as 'exogenous,'
and labeling the remainder as endogenouss,'
the number of endogenous variables appearing
in any given equation cannot be greater
than one more than the number of exogenous
variables left out of this equation.
(1966:57)

As previously noted, however, the identification

problem, as discussed here, arises in status inconsistency

research when one attempts a structural equations










approach. Since that approach is not taken in this

research, the identification problem is not a major

issue. This research utilizes a regression analysis

type of approach, and as Blalock has noted:

Whenever one uses a set of simul-
taneous equations to provide theo-
retical models of causal processes,
he can generally expect to encoun-
ter identification problems because
there may be more unknowns than
pieces of empirical information for
estimating the parameters. Such
identification problems do not
arise on simple prediction problems,
where one uses a simple regression
equation to predict to a single de-
pendent variable (1967:792)

There are, however, two issues related to obtaining

identifiable coefficients which are a relevant concern

in regression analysis. When multicollinearity is high

or singularity is present it becomes impossible to obtain

identificable coefficients. The problems associated with

multicollinearity and singularity also have major impli-

cations in regression procedures. If multicollinearity

is high or singularity is present the results obtained

from regression analysis may be misleading. These two

issues are discussed in greater detail in a later

section dealing with the control variables and provide

a statistical reason for employing SES as a control

variable.










The Data

The data for this analysis were obtained by com-

bining two epidemiologic field surveys collected between

1970 and 1973 in counties in central and north central

Florida. This was done for the purpose of increasing the

sample size. In addition to the statistical advantage

obtained from a large sample when employing multivariate

statistical techniques, two other considerations supported

the use of a large sample. First, the entire sample was

not utilized; a subsample consisting of all full-time

employed subjects was selected. Nelson (1973) has pointed

out the methological hazards associated with including

persons who are not employed full-time in studies of

status inconsistency. Regardless of the particular reason

for their not being employed full-time, complications

arise in assigning them scores on income, and especially

on occupation. The allocation of a rank by some type of

arbitrary method introduces unknown confounding effects.

When full-time employed peoDle are selected, the proba-

bility of their having complete scores for all three

variables (education, occupation and income) is greatly

increased. Second, a large sample increases the likeli-

hood that all the various types of status inconsistency

to be examined will be represented. The combined sample

consisted of 3674 cases, but when the full-time employed

people were selected, a working sample of 1619 cases was










retained. As a result of missing values on some of

the variables, this sample size was further reduced to

1472 cases.

One of the two component samples came from the

project "Evaluating Southern Mental Health Needs and

Services" (NIMH Grant 15900). For a detailed descrip-

tion of this sample see Warheit et al. (1973b). The

other sample came from the project "Southern Health and

Family Life Studies" (NIMH Contract HSM 42-73-9 OC).

For a detailed description of this sample see Bell et al.

(1974).

Each of these samples was drawn by means of a syste-

matic probability sampling procedure based on an enume-

ration of households provided by tuility services and

supplemented by area sampling. The Kish (1965) method for

determining the particular respondent within a house-

hold to be interviewed was also utilized for each sample.

When both samples were compared to the 1970 census of the

counties from which they were selected, they were found

to be representative in terms of sociodemographic

characteristics.



Measuement of Status Inconsistency

When utilizing regression analysis in status incon-

sistency research, one of the most common methods employed










for measuring inconsistency is to take the interaction

effects of the status variables as an indication of

status inconsistency. But, as was previously noted,

several researchers have pointed out the weakness of

this type of approach (Jackson and Burke, 1965; Hyman,

1966, Knoke, 1972; House and Harkins, 1975). The problem

is that there may be inconsistency effects which are not

revealed by the interaction effects of the status variables.

For this reason, the utilization of the interaction

effects of the status variables, as a measurement of status

inconsistency, was considered to be an unacceptable

approach for this research.

The method utilized for measuring the independent

variable of status inconsistency in this research was

developed by the United States Bureau of the Census (1967).

This procedure yields the 13 types of status inconsis-

tency listed in Table 1. These 13 categories of status

inconsistency were collapsed according to three different

sets of criteria to yield three different measures of

status inconsistency, listed in Tables 2,3, and 4. Atten-

tion will first be given to the procedure utilized to

obtain the 13 types of status inconsistency, since the

other three measures simply involve different ways of

collapsing the 13 status inconsistency types.

The three status varibles of education, occupation










and income were measured as follows. The scores for

education utilized in this analysis were obtained by

selecting the midpoint of each education category from

a cumulative percentage distribution of the years of

school completed by residents of the United States. The

scores for income were obtained in a similar manner.

The midpoint of each income category was selected from

a cumulative percentage distribution of the income of

residents of the United States (U.S. Bureau of the Census,

1973a:368-79). This procedure yielded scores for edu-

cation and income which ranged from zero to 100 and

represented the relative cumulative percentage distribution

of that level of education or income in the United States.

For example, a person who had graduated from high school

but had obtained no further formal education was assigned

an education score of 63. This indicated that according

to the 1970 census the highest formal educational

attainment of 63% of the population was the same as this

individual, a high school graduate. The same procedure

and interpretation applies to the income scores.

The occupation scores were obtained according to

the following method:

The detailed occupations were scored accor-
ding to the combined average levels of edu-
cation and income for the given occupation.
Thus, the score obtained is an average for
the occupation and it contributes an










independent* effect to the total socioeconomic
score, which includes also the individual's
actual educational and income levels. Using
the number of workers in each occupation, a
cumulative percentage distribution was obtained.
The score for a given occupation was then
determined by taking the midpoint of the cumu-
lative percentage interval for that occupation
(emphasis added). (U.S. Bureau of the
Census, 1967:XI)

The occupational scores obtained by this method were

taken from Nam et al. (1975), and they differ in two

respects from occupational scores utilized previously.

First, these occupational scores were calculated on the

basis of the 1970 U.S. census rather than the 1950

U.S. census. Second, these scores were calculated for

both males and females, rather than just males. These

differences make two important contributions. First, the

occupational scores based on the 1970 census are likely

to provide a better representation of the data collected

between 1970 and 1973 than are scores based on the 1950

census. Second, having scores for both males and females



*It should be noted that the independence referred to here
is mathematical independence, not statistical independence.
It would be contradictory to claim that a measure which
truly measured occupational status was statistically
independent of income and educational attainment. This
occupational measure is mathematically independent in the
sense that the occupation scores for the individual are
not a mathematical function of the individual's income or
education scores. The score for any particular occupation
represents the average income and education level for every-
one in the civilian labor force who claims that particular
occupation.










allows the inclusion of females in the analysis. They

are assigned a score which is valid for females as well

as males, as opposed to a score based on males in the

civilian labor force, or their husband's occupation,

which has been common practice in the past. The education,

occupation and income scores employed in this analysis

are provided in Appendix A.

Hartman (1975:715) has criticized the method developed

by the United States Bureau of the Census for measuring

status inconsistency because the occupation score is based

on a combined average of the income and education scores.

While what he said is true as far as he goes, it does not

reveal the entire procedure. The occupation score is an

average of the education and income score, but not at the

individual level. As Nam et al. (1975) indicated, the

occupational scores are an average of the income and

education level of all the people in the civilian labor

force who are in that particular occupation. Based on

the education, occupation and income scores provided in

Appendix A, the following is an individual example demon-

strating that the individual's occupation score is not a

mathematical function of his education and income scores.

An individual carpenter who has an eleventh grade education

would have an education score of 37. If his income for the

last year was $5500, his income score would be 60. His










occupation score, 42, is the average of the education

and income scores for all carDenters in the civilian

labor force (Nam et al., 1975:573). This occupation

score of 42 is obviously not an average of his education

and income scores (37 + 60 = 97 2 = u8.5). In this way,

the individual's occupation score contributes an effect

which is mathematically independent of his income and

education scores. Assigning the average score of a

collectivity who share the same occupation to an indivi-

dual who claims that occupation is a valid procedure,

since as Knoke (1972:28) has noted, "Occupational status

is a collective phenomenon which an individual may tap

by virtue of his membership in the group." Additionally,

this occupational status measure represents a behavioral

as opposed to an attitudinal measure. Rather than

assessing attitudes concerning the prestige of the

occupation, the behavior towards the occupation is assessed

in terms of how the occupation is rewarded (income) and

the average entry and functioning levels of training

(education) required.

Hartman (1975:717) has also objected to the use of

percentile distributions for the calculation of status

inconsistency on the basis that it ties the measure to a

particular sample and, therefore, limits comparability.

In making this criticism he seems to be assuming that










the percentile distributions are based on some specific

"local" sample. The percentile distributions employed

here, however, are based on "national" figures. This

does not limit comparability since anyone wishing to

undertake a comparable study could obtain the same per-

centile distributions from the U.S. Bureau of the Census.



Thirteen Status Inconsistency Types

The 13 status inconsistency types which appear in

Table 1 were obtained by the following procedure utilizing

the education, occupation and income scores described

previously.

A. If the score between the highest and lowest
scores for education, occupation and income
was 25* or less, code 1 in Table 1 was assigned.
Code 1 is the consistent type, indicating that
as measured by this procedure no status
inconsistency exists.

B. Code 2,4, or 6 from Table 1 was assigned if the


*The original census report used a score differential
of 20 points, rather than the range of 25 points which is
used here. Lenski, who maintained that status inconsis-
tency effects "result only from marked or pronounced in-
consistencies of status," employed a 30 point differen-
tial (Lenski, 1956a:369). The 25 point differential utili-
zed here represents a compromise between the two. Addi-
tionally, it represents one-fourth of the possible range
of variation in the scores, and thus is substantial enough
that intuitively it would seem to represent a noticeable
difference, without being so drastic as to select only
very unusual cases. As Kasl and Cobb (1971:3) have pointed
out, the exact cutting point, "which determines the
relative proportions of congruent or incongruent subjects,
has to be necessarily rather arbitrary."










range between the highest and lowest scores
for education, occupation and income exceeded
25, while the range between the middle and
lowest scores was 25 or less, but less than the
range between the highest and middle scores.
This indicates that two of the scores were
consistent, but one score was inconsistent
by being at least 25 points higher than the
lowest score and having a greater range between
the high and middle score than between the low
and middle score.
1. Code 2 was assigned if the income score was
the inconsistent score which was high.
2. Code 4 was assigned if the education score
was the inconsistent score which was high.
3. Code 6 was assigned if the occupation score
was the inconsistent score which was high.

C. Code 3,5 or 7 from Table 1 was assigned if the
range between the highest and lowest scores for
education, occupation and income exceeded 25,
while the range between the highest and middle
scores was 25 or less, but less than the range
between the middle and lowest scores. This
indicates that two of the scores were consis-
tent but one score was inconsistent by being
at least 25 points lower than the highest score
and having a greater range between the low and
middle score than between the high and middle
score.
1. Code 3 was assigned if the income score
was the inconsistent score which was low.
2. Code 5 was assigned if the education score
was the inconsistent score which was low.
3. Code 7 was assigned if the occupation score
was the inconsistent score which was low.

D. Code 8,9,10,11,12 or 13 from Table 1 was assigned
if the range between the highest and middle scores
and the range between the middle and lowest scores
each exceeded 25 points. This indicates that
all three scores were inconsistent, with at least
25 points between the highest and middle scores
and at least 25 points between the lowest and
middle scores.
1. Code 8 was assigned if all three scores were
inconsistent and the occupation score was
the highest score while the income score
was the lowest score.
2. Code 9 was assigned if all three scores were










inconsistent and the occupation score was
the highest score while the education score
was the lowest score.
3. Code 10 was assigned if all three scores were
inconsistent and the education score was the
highest score while the occupation score was
the lowest score.
4. Code 11 was assigned if all three scores were
inconsistent and the education score was the
highest score while the income score was the
lowest score.
5. Code 12 was assigned if all three scores were
inconsistent and the income score was the
highest score while the education score was
the lowest score.
6. Code 13 was assigned if all three scores were
inconsistent and the income score was the
highest score while the education score was
the lowest score.
(U.S. Bureau of the Census, 1967:X-XI)

The above procedure resulted in the production of the

13 status inconsistency types which appear in Table 1.

This listing constitutes the 13 basic types of status

inconsistency from which all the measure5of status incon-

sistency utilized in this research are constructed. It

should also be noted that this method has avoided the

topologicall fallacy" criticized by Berry and Martin (1973),

in that it does represent a complete status profile analy-

sis for the statuses of education, occupation and income.

All three statuses are represented in each of the 13

status inconsistency types in Table 1. These 13 basic

status inconsistency types are collapsed in three

different ways to produce three separate measures of

status inconsistency. In all three of these measures,

however, the problems associated with not having a complete










TABLE 1. THIRTEEN STATUS INCONSISTENCY TYPES



1. (E 0 I) All three components consistent*

2. (I/O E) Occupation and education consistent; income
high

3. (0 E/I) Occupation and education consistent; income
low

4. (E/O I) Occupation and income consistent; education
high

5. (0 I/E) Occupation and income consistent; education
low

6. (0/E I) Education and income consistent; occupation
high

7. (E I/O) Education and income consistent; occupation
low

8. (0/E/I) All inconsistent; occupation highest, income
lowest

9. (O/I/E) All inconsistent; occupation highest, education
lowest

10. (E/I/0) All inconsistent; education highest, occupation
lowest

11. (E/O/I) All inconsistent; education highest, income
lowest

12. (I/E/0) All inconsistent; income highest, occupation
lowest

13. (I/O/E) All inconsistent; income highest, education
lowest


*Thn table heading is stated as Thirteen Status Inconsis-
tency Types, vet the first type listed is a consistent
type rather than an inconsistent type. This is done in
order to provide a consistent type with which to compare
the inconsistent types. In this analysis, all the tables
which report status inconsistency types contain one
consistent type with which the inconsistent types may be
comDared.










status profile analysis have been avoided, since each

of the 13 component types which were collapsed to for-

mulate the three measures constituted a complete status

profile.



Three Status Inconsistency Types

The first reformulation of the 13 status inconsistency

types provided for the construction of the basic under-

rewarded, overrewarded and consistently rewarded types of

status inconsistency which were discussed in the theory

section in relation to distributive justice. Education

and occupation were considered as investments and income

as the reward for these investments. This theoretical

reasoning led to collapsing the 13 types of status incon-

sistency listed below in Table 2.



TABLE 2. THREE STATUS INCONSISTENCY TYPES



1. The Underrewarded Type includes from Table 1 types:
3(0 E/I), 4(E/O I), 6(0/E I), 8(0/E/I), 9(O/I/E),
10(E/I/0), ll(E/0/I).

2. The Overrewarded Type includes from Table 1 types:
2(1/0 E), 5(0 I/E), 7(E I/O), 12 (I/E/0), 13(I/0/E).

3. The Consistent Type includes from Table 1 type:
1(E 0 I).




It should be noted that in the underrewarded type income










is never the high status variable, while in the overrewarded

type income is always either the high status variable or

tied for high with one other variable and never appears as

the low variable. Of course, in the consistent type

all three are consistent. Thus, in the underrewarded

type the reward variable is always the low variable; in

the overrewarded type the reward variable is always the

high variable. In the consistent type, the investment

and reward variables are consistent.



Four Status Inconsistency Types

The second reformulation of the 13 status inconsistency

types provided the inconsistency measure for the first

test of the predominant process in the relationship

between status inconsistency and poor mental health dis-

cussed in the theory section. As previously noted, Dunham

et al. (1966) maintained that status inconsistency in

which educational status was higher than occupational

status was the type of inconsistency which was most likely

to be a result of poor mental health. This was based on

the premise that the educational system constitutes a

"semi protective" environment in which individuals with

poor mental health may function adequately. When compelled

to compete in the "real world" of the job market where

a certain level of skill in social functioning is required,










the person with poor mental health may be at a disadvantage

regardless of his educational attainment.

If the contention of Dunham et al. (1966) that poor

mental health results in status inconsistency is valid,

one would expect to find a strong relationship between

poor mental health and status inconsistency of the type

education higher than occupation. For the purposes of

this research this type of status inconsistency has been

labeled as the impaired type.

The second measure of status inconsistency utilized in

this research is presented below in Table 3.



TABLE 3. FOUR STATUS INCONSISTENCY TYPES



1. The Underrewarded Type includes from Table 1 types:
3(0 E/I), 6(0/E I), 8 (0/E/I), 9(0/I/E) 1 (E/O/T).

2. The Overrewarded Type includes from Table 1 types:
2(1/0 E), 5(0 I/E), 12(I/E/0), 13(I/0/E).

3. The Impaired Type includes from Table 1 types:
4(E/O I), 7(E I/O), 10(E/I/0).

4. The Consistent Type includes from Table 1 type:
1(E 0 I).




This measure was formulated by collapsing the 13 types

of status inconsistency from Table 1 into the four status

inconsistency types in Table 3. For these four status

inconsistency types, the underrewarded, overrewarded,










and consistent types remained the same, the only altera-

tion being that the impaired type was selected from among

them. The impaired type was selected on the basis that

the occupation score was always below the education score.



'. f*,ltus Inconsistency Types

The third reformulation of the 13 status inconsis-

tency types provided the inconsistency measure for the

second test of the predominant process in the relationship

between status inconsistency and poor mental health dis-

cussed in Chapter 3. This test involved determining

whether or not the subject perceived himself as under-

rewarded. This was measured by the subject's yes or no

response to the following question: "All things consi-

dered, do you think your pay is a fair wage/salary?" A

positive response to this item was taken as an indication

that the subject perceived his wages as being fair and

therefore did not perceive himself as underrewarded.

A negative response to this item was taken as an indication

that the subject perceived his wages as being unfair and

therefore did perceive himself as underrewarded. Some

might argue that the subject could claim his wages were

unfair but that they were unfair in the direction of

being overrewarded rather than underrewarded. While tech-

nically this is possible, it seems extremely unlikely




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs